United States    Office of
     Environmental Protection Research and Development
     Agency      Washington, DC 20460
EPA-600/R-94-012
January 1994
EPA  Comparison of the 1985
      NAPAP Emissions
      Inventory with the 1985
      EPA TRENDS Estimate for
      Industrial SO2 Sources

-------
                       EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Informa-
tion Service. Springfield, Virginia 22161.

-------
                                                       EPA-600/R-94-012
                                                       January 1994
                Comparison of the 1985 NAPAP Emissions Inventory
                       with the 1985 EPA TRENDS Estimate
                            for Industrial SO2 Sources

                                  Final Report
                                   Prepared by:
                                David Zimmerman
                           TRC Environmental Corporation
                            100 Europa Drive, Suite 150
                              Chapel Hill, NC  27514

                                       and

                                  Rebecca Battye
                                EC/R Incorporated
                            University Tower, Suite 404
                                 3101 Petty Road
                                Durham, NC 27707
                           EPA Contract No. 68-D2-0181
                           Work Assignment Nos. 2 and 8
                       EPA Project Officer: Charles C. Masser
                        U.S. Environmental Protection Agency
                   Air and Energy Engineering Research Laboratory
                         Research Triangle Park, NC 27711
                                   Prepared for:

U.S. Environmental Protection Agency          U.S. Environmental Protection Agency
Office of Air Quality Planning and             Office of Research and Development
Standards                                   Washington, DC  20460
Research Triangle Park, NC 27711

-------
                            ACKNOWLEDGEMENTS

      Special acknowledgements are to be made to Sue Kimbrough and Charles Mann in the
Emissions and Modeling Branch of AEERL for their insight and historical knowledge of
NAPAP and TRENDS and to J. David Mobley and his staff in the Emissions Inventory
Branch of OAQPS for their support in reviewing this document.
                                   FOREWORD

       This document compares results from 1985 EPA Emission TRENDS methodology
 with 1985 NAPAP Emission Inventory methodology. Due to findings in this report as well as
 other factors, the TRENDS methodology has been revised as of 1993; thus, references to
 TRENDS in this report will no longer be valid for years  1985 and beyond, effective with the
 1993 edition of the TRENDS report The new TRENDS methodology uses the 1985 NAPAP
 Emission Inventory as a base.  Further changes will  be seen in the TRENDS reports published
 in 1994 and thereafter. Thus, the reader is cautioned that comments on the EPA TRENDS
 report in this document are valid for editions prior to 1993, but are not valid for the editions
 1993 and thereafter.
                                        11

-------
                           TABLE OF CONTENTS

Section                                                                  Page



ACKNOWLEDGEMENTS	  ii

FOREWORD  	   ii

LIST OF TABLES	  vi

LIST OF FIGURES	  vi

EXECUTIVE SUMMARY	  vii

SECTION 1 INTRODUCTION	   1

SECTION 2 ANALYSIS OF 1985 TRENDS AND NAPAP ESTIMATES		   5
      2.1  COMBUSTION SOURCES	   6
            2.1.1  Oil Combustion	   7
            2.1.2  Coal Combustion	  22
            2.1.3  Anthracite Coal	  23
            2.1.4  Bituminous Coal and Lignite	  25
            2.1.5  Natural Gas	  31
            2.1.6  Miscellaneous Fuel	  38
            2.1.7  Wood	  52
            2.1.8  Other NAPAP Combustion Categories  	  57
      2.2  NON-FERROUS SMELTING  	  57
            2.2.1  Primary Copper	  58
            2.2.2  Combined Primary Lead and Primary Zinc	  59
            2.2.3  Primary Zinc	  59
            2.2.4  Primary Lead	  63
            2.2.5  Primary Aluminum	  67
            2.2.6  Secondary Lead	  72
            2.2.7  Other NAPAP Non-ferrous Emission Categories  	  75
      2.3  OTHER INDUSTRIAL PROCESSES	  75
            2.3.1  Kraft Pulp Production	  76
            2.3.2  Chemical Manufacturing	  83
            2.3.3  Carbon Black Production 	  83
            2.3.4  Sulfuric Acid	  88
            2.3.5  Sulfur Recovery  Plants	  93
            2.3.6  Petroleum Refineries  	  98
            2.3.7  Natural Gas Production	109
            2.3.8  Iron and Steel	113
            2.3.9  Cement Manufacturing	126

-------
                            TABLE OF CONTENTS

Section                                                                   Page

            2.3.10 Glass Manufacturing	131
            2.3.11 Lime Manufacturing	134
            2.3.12 Additional Industrial Process Emission Categories in the NAPAP
                  Inventory	137

SECTION 3  CONCLUSIONS 	141
      3.1  COMBUSTION SOURCES	142
      3.2  COMBUSTION OF OIL	142
      3.3  COAL COMBUSTION	144
      3.4  NATURAL GAS  COMBUSTION  	146
      3.5  MISCELLANEOUS FUELS	147
            3.5.1 Coke  	148
            3.5.2 Coke Oven Gas	148
            3.5.3 Kerosene	149
            3.5.4 LPG	150
      3.6  WOOD	150
      3.7  NON-FERROUS  SMELTING SOURCES	151
            3.7.1 Primary  Zinc	151
            3.7.2 Primary  Lead	152
            3.7.3 Primary  Aluminum	153
            3.7.4 Secondary Lead	154
            3.7.5 Other Non-ferrous Emissions Reported in NAPAP	154
      3.8  OTHER INDUSTRIAL PROCESS EMISSION SOURCES 	155
            3.8.1 Kraft Pulp Production	155
            3.8.2 Carbon Black Manufacture  	156
            3.8.3 Sulfuric  Acid	157
            3.8.4 Sulfur Recovery Plants	158
            3.8.5 Petroleum Refineries  	159
            3.8.6 Natural Gas  Production	161
            3.8.7 Iron and  Steel  	162
            3.8.8 Cement Manufacturing	163
            3.8.9 Glass Manufacturing	164
            3.8.10 Lime Manufacturing	165

SECTION 4  REFERENCES  	166

APPENDIX A EPA TRENDS PROCEDURE FOR INDUSTRIAL SO2 EMISSIONS . .  . A-l
      Anthracite Coal	A-l
      Bituminous Coal and Lignite 	A-2
      Residual Oil	A-3
      Distillate Oil	A-4
      Natural Gas 	A-5

-------
                             TABLE OF CONTENTS

Section                                                                      Page

      Miscellaneous Fuel  	A-6
      Primary Copper	A-10
      Primary Zinc	A-12
      Primary Lead  	A-13
      Primary Aluminum  	A-14
      Secondary Lead	A-15
      Pulp and Paper 	A-15
      Sulfuric Acid  	A-16
      Carbon Black Production  	A-17
      Sulfur Recovery Plants	A-17
      Petroleum Refining	A-18
      Iron and Steel	A-19
      Cement Manufacturing	A-21
      Glass	A-22
      Lime Manufacturing  	A-23

-------
                               LIST OF TABLES

Number                                                                   Page

1-1    Magnitude Differences Between 1985 TRENDS and NAPAP SO2 Emission
      Estimates  	 4
2-1    Emissions from Combustion TRENDS versus NAPAP	 8
2-2    Comparison of Oil Combustion Values for 1985 TRENDS and NAPAP  	  12
2-3    Weighted Average Emission Factor for Distillate Oil Combustion	  14
2-4    Comparison of Coal Combustion Values for 1985 TRENDS and NAPAP  	  22
2-5    Comparison of Natural Gas Values for 1985 TRENDS and NAPAP  	  31
2-6    Comparison of Miscellaneous  Fuels Values for 1985 TRENDS and NAPAP  ....  39
2-7    Weighted Average Emission Factor for Coke Combustion  	  43
2-8    Comparison of Non-Ferrous Smelting Values for 1985 TRENDS and NAPAP ...  58
2-9    Recovery of Sulfur as H2SO4  	  59
2-10  Other Non-Ferrous Emissions  Reported in NAPAP	  76
2-11  Comparison of Other Industrial Processes Values for 1985 TRENDS and
      NAPAP 	  77
2-12  Emissions  from Chemical Manufacturing Sources Included in NAPAP but not
      in TRENDS 	138
2-13  Emissions  from Mineral Products Sources Included in NAPAP but not in
      TRENDS  	139
2-14  Emissions  from Other Industrial Process Sources Included in NAPAP but not
      in TRENDS 	140

3.9-1  Weighted Average Emission Factors for Industrial Oil Combustion	A-5

3.11-1 Weighted Average Emission Factors for Coke	A-8

3.15-9 Capacity Data	A_i(3

3.15-10  Smelting  Emission Factor  Data  	                         A_JJ

3.15-11  Converting Emission Factor Data	                        A-12

3.15-12  Emission Factors for Uncontrolled Emissions  	                     * 99

3.15-13  Weighting Factors 	                   A --
                                                       *"*"*••••.»...... f\~'£3

                               LIST OF FIGURES

 Number
                                                                           Page
 1      Cumulative sulfur dioxide emissions in the 1985 NAPAP inventory versus
      number of plants	
                                                                           " *  *.

                                      vi

-------
                               EXECUTIVE SUMMARY

       Section 406 of the 1990 Clean Air Act Amendments (CAAA) requires that the
 Administrator of the Environmental Protection Agency transmit to Congress a report
 containing a national  inventory of annual sulfur dioxide (SO:)  emissions from industrial
 sources not later than January  1, 1995 for all years for which data are available, as well as the
 likely trend in SO2 emissions over  the following twenty year period (1995 to 2015). Under
 the Act as amended, 1985 served as the  baseline for the 5.6 million ton industrial  SO:
 emission estimate.  To provide the  1995 analysis mandated by Congress,  the 1985 baseline
 data must first be examined to identify strengths and weaknesses in the available emission
 and supporting data.  This report presents the initial analysis of two major sources of baseline
 industrial data available at this time:  the 1985  National Acid Precipitation Assessment
 Program (NAPAP) emission inventory and the  1985  national emission estimates, referred to
 as the TRENDS emission estimates.

       The 1985 NAPAP emission inventory effort supported acid precipitation deposition
 research, including atmospheric modeling, through comprehensive, detailed source emission
 estimates provided by local and state agencies.   It  is a "bottom-up" inventory and a 1985
 snapshot  The SO2 emission data for significant (>100 tons per year) sources were
 systematically quality assured,  with greater effort expended on larger sources. The inventory
 included a unique confirmation step, allowing individual plants emitting at least 2500 tons per
 year to review their emission estimates prior to finalization. The  resulting inventory is widely
 regarded as the most comprehensive and accurate national inventory compiled to date.

       The TRENDS  estimates represent both current and historic emissions (1940 to present)
and are compiled annually.   Industrial emission estimates are derived from national, published
activity data and standard emission factors; historic estimates are  altered based on the most
recent activity data and emission factors to better represent the most current understanding of
emission processes. It is essentially a "top-down"  approach and not a true inventory.  It
                                          vu

-------
presents a consistently derived national emission estimate at the emission category level (e.g-,
industrial oil combustion) rather than the source (e.g., boiler) level.

       The aim of this document is to analyze the derivation of individual industrial category
estimates from NAPAP and TRENDS.  Such  analyses are complicated by several factors:
       NAPAP is comprehensive and includes all reported industrial emission
       categories;  TRENDS is limited to categories thought to emit at least 10,000
       metric tons of a criteria pollutant per year.
       NAPAP is source and plant specific; TRENDS is national and category-
       specific.  There is no opportunity  to match individual data values between the
       inventories; in fact, category definitions differ between the two inventories.
       The 1985 NAPAP inventory is a single year inventory  and is not updated;
       TRENDS adjusts historical data based on the most current information.
       A complete understanding of a NAPAP emission category could entail plant by
       plant, source by source review of throughput data, control information, fuel
       parameters, etc.  Such an effort was not possible within the framework of this
       project  The TRENDS methods were reviewed, but still represent a large and
       complicated set of data, methods and assumptions.
It is unavoidable but in a number of cases it is difficult if not  impossible to compare
TRENDS and NAPAP on an even basis.
       This document presents a highly detailed view of the two inventories on a category
basis, principally relying on emission and activity (throughput) data.  Methods, data sources,
emissions and assumptions are documented and analyzed in as much detail as possible so that
this information need not be recompiled for future examination.  The authors have attempted
to reproduce the 1985 TRENDS emissions estimates and noted any irregularities in the
calculated and published data.  The analyses proved complex, especially when disaggregating
data to create comparable categories between NAPAP and TRENDS  datasets, and raised a
number of questions.  (In particular, categories with significant inprocess fuel emissions such
as cement proved difficult to assess because the process and fuel emissions were treated
differently in each dataset.)  As such, most of the document is technical in nature and  is
meant to be a reference tool as individual category methods and  emissions are compared
                                         viii

-------
 during the development of the 1995 Report to Congress.  Supporting data are consolidated
 and provided in Appendix A.

       Overall, the two 1985 estimates (NAPAP and TRENDS) compare favorably:  NAPAP
 estimates 5.6 million tons SO2 (as reflected in the Act as amended) and TRENDS estimates
 6.0 million tons. When broken down, however, the two estimates show greater divergence
 for individual categories.  The TRENDS estimate is larger than the 1985 NAPAP estimate
 and does not include as many source categories.  The TRENDS estimate systematically
 overestimates emissions, relative to the NAPAP inventory, and  therefore on aggregate the
 estimates are very similar. The primary reason for the overestimation in the TRENDS
 method is the exclusion of SO2 control technology that has come into being over the years.
 SO2 control  technology has been applied to the majority of the  large industrial SO2 categories
 through the  promulgation of New Source Performance Standards (NSPS), issuance of
 operating permits, and New Source Review Permits.  Several categories exceeding 100,000
 tons of SO2  differ by more than 50 percent between the two datasets:

 •      Primary lead and zinc
 •      Iron and steel
 •      Oil and natural gas production
 •      Pulp  and paper
 •      Cement

       In general terms, the authors drew the following conclusions from the analyses of the
 two datasets. These conclusions are based  on an analysis of the 1985 NAPAP SAS® data
files (annual files) and the published TRENDS methods and emissions data.
NAPAP
      The 1985 NAPAP inventory still represents the most comprehensive and
      accurate emissions estimates for 1985 because of its unprecedentedly rigorous
      quality assurance of emissions and bottom-up nature.  The inventory accounts
      for individual source operating characteristics, controls and emission factors.
                                         IX

-------
       Activity data in the 1985 NAPAP inventory were not subject to the same
       standard of quality assurance or completeness. Some data were unreported due
       to confidentiality restrictions and activity data for small sources (i.e., <100 tons
       per year) passed only the grossest quality assurance checks.  There are known
       reporting problems among miscellaneous fuels and other categories.  The
       accuracy and representativeness of activity data in the NAPAP inventory are
       best evaluated  source by source; category-level summaries are unreliable
       without adjustments.

       It is still very possible to locate questionable data values in the 1985 NAPAP
       emission inventory when examined on a source by source basis, especially for
       smaller emitters.
 TRENDS
       Some industrial emission categories, notably processes within oil and natural
       gas production, are missing from the TRENDS method.

       As a top-down approach, broad assumptions of emission factors and controls
       are used across a category. Frequently, estimates make no adjustment for
       controls and accommodation for individual operating characteristics, including
       emission factors, is impossible.

       For the most part, the underlying industrial activity data are reliable and
       probably far superior to the corresponding NAPAP estimates at the category
       level.  Any method for  1995 and beyond  should take advantage of the
       underlying data sources.

       Based on the TRENDS  documentation, the actual TRENDS execution contains
       minor to moderate errors in calculation of activity and  emissions.  As such,
       some sections of this report related to  the calculation of the TRENDS emission
       estimate are  internally inconsistent.  TRENDS could also benefit from recently
       revised standard emission factors and updated sources of activity data.
       Due to findings in this report as well as other factors, the TRENDS methodology has

been revised as of 1993; thus, references to TRENDS in this report will no longer be valid for
years 1985 and beyond, effective with  the 1993 edition of the TRENDS report.  The new

TRENDS  methodology uses the 1985 NAPAP Emission Inventory as a base.  Further changes

will be seen in the TRENDS reports published in 1994 and thereafter.  Thus, the reader is

cautioned  that comments on the EPA TRENDS report in this document are valid for editions
prior to 1993, but are not valid for the editions for 1993 and thereafter.

                                         x

-------
                                    SECTION 1
                                 INTRODUCTION

       Section 406 of the 1990 Clean Air Act Amendments (CAAA) requires that not later
 than January 1, 1995, the Administrator of the Environmental Protection Agency (EPA)
 transmit to Congress a report containing an inventory of national annual sulfur dioxide (SO2)
 emissions from industrial sources for all years for which data are available, as well as the
 likely trend in  such emissions over the following twenty-year period.

       To support the development of the report to Congress, the Air and Energy Engineering
 Research Laboratory (AEERL) conducted an analysis of the differences in the 1985 National
 Acid Precipitation Assessment Program (NAPAP) Emission Inventory and the 1985 EPA
 TRENDS Emission Estimates for industrial SO2 emission sources. This document presents
 the initial findings of this analysis.

       The 1985 NAPAP emission inventory  is the most comprehensive national inventory of
 industrial SO2 emissions  that has been compUfd to date. The NAPAP inventory was used by
 Congress in the development of the  1990 CAAA. In section 406 of the  1990 CAAA,
 Congress cites  a limit for industrial SO2 emissions of 5.6 million tons. This value was
 derived from the 1985 NAPAP emission inventory.1

       The majority of industrial SO2 emissions as reported in the 1985 NAPAP Emission
 Inventory are reported as point sources.  Over 4,000 industrial facilities were listed as
 emitting SO2, one third of these facilities emitted a small amount of SO2 (<100  tons of SO2).
 Figure  1 shows that relatively few sources (about 130) account for approximately 50  percent
 of the total industrial SO2 emissions. Based on data from the 1985 NAPAP inventory, only
 500 facilities account  for 80 percent of the industrial SO2 emissions.

       Every year the Office of Air Quality Planning and Standards (OAQPS) prepares
annual  estimates to determine emission TRENDS.2 The emission estimates are  updated as
necessary to account for changes in the activity data or the  emission estimation method.

-------
5,000,000 -
                          1,000
  2,000            3,000
# of plants
4,000
    Figure 1. Cumulative sulfur dioxide emissions in the 1985 NAPAP inventory versus
              number of plants
  Historically, the EPA TRENDS estimates have only included source categories that exceeded
  10,000 metric tons of a criteria pollutant.
        The calculation of the EPA TRENDS emission estimates is accomplished by following
  a documented procedure. The procedure that pertains to industrial SO2 emission source
  categories is included as Appendix A.  The procedures manual is  supplemented with two
  extensive spreadsheets that are utilized in the TRENDS calculation procedure. The emissions
  calculations spreadsheet includes all of the emission factors, fuel sulfur assumptions, and
  control assumptions and is utilized in the calculation of current emissions.  The second
  spreadsheet contains historic activity data and is used to project emissions for the study year.
  Data are entered into the spreadsheets and are used in other areas of the spreadsheet as
  necessary.  The only documentation of the 1985 TRENDS estimates is the procedures manual
  and the accompanying spreadsheets.  In several instances, following the documented

-------
procedures resulted in values that were inconsistent with the values in these spreadsheets.
This occasionally resulted in two sets of 1985 TRENDS SO2 emission estimates, the
published TRENDS value and the calculated TRENDS value. These inconsistencies are
documented throughout this report.

       In 1991, a scoping study was undertaken to determine the feasibility of different
approaches for developing an updated industrial SO2 emission estimate for the report to
Congress. This scoping study revealed large differences in the NAPAP and TRENDS
emission estimates.  Table 1-1 summarizes the differences between the NAPAP and TRENDS
estimates.

       The annual EPA TRENDS estimate is a logical source of industrial SO2 emissions
trends data.  Prior to using the EPA TRENDS estimate in the 1995 Report to Congress, a
detailed analysis of the differences that exist in the 1985 NAPAP and TRENDS estimates was
undertaken.  The results of that analysis are presented in this report.

       This report includes three sections.  Section 2 presents the results of the detailed
analysis.  Section 3 summarizes the conclusions for each source category covered in the
detailed analysis.

-------
      TABLE 1-1.  MAGNITUDE DIFFERENCES BETWEEN 1985 TRENDS
                   AND NAPAP SO2 EMISSION ESTIMATES
Source Category3
Coal"
Oil0
Natural Gasd
Wood
Miscellaneous Fuel
Other Fuel Combustion Emissions
Reported through NAPAP
1° Copper
1° Lead and Zinc
2° Lead
1° Aluminum
Other Primary and Secondary Metals
Emissions Reported through NAPAP
Iron and Steel
Iron and Steel Foundries
Oil and Natural Gas Production
Pulp and Paper
Cement
Glass
Lime
Sulfuric Acid
Carbon Black
Petroleum Refineries
Other Industrial Process Emissions
Reported through NAPAP
Total
TRENDS
(tons)
1,840,000
540,000
0
10,000
80,000


650,000
240,000
30,000
70,000


360,000

160,000
250,000
620,000
30,000
30,000
210,000
10,000
830,000


5,960,000
NAPAP
(tons)
1,721,000
713,000
33,000
42,000
14,000
74,000

655,000
106,000
21,000
58,000
42,000

204,000
16,000
332,000
130,000
291,000
23,000
32,000
217,000
28,000
640,000
220,000

5,612,000
Delta Delta
(tons) (percent)
119,000
-173,000
-33,000
-32,000
66,000
-74,000

-5,000
134,000
9,000
12,000
-42,000

156,000
-16,000
-172,000
120,000
329,000
7,000
-2,000
-7,000
-18,000
190,000
-220,000

348,000
6.9
-24.3
-100.0
-76.2
471.4
-100.0

-0.8
126.4
42.9
20.7
-100.0

76.5
-100.0
-51.8
92.3
113.1
30.4
-6.3
-3.2
-64.3
29.7
-100.0

6.2
"Except where noted, the emissions for a source category represent process level emissions
only and do not include emissions from the combustion of fuel.
"Excludes bituminous coal and lignite consumed at cement and lime manufacturing facilities
Excludes both  distillate and residual oil consumed at cement plants and petroleum refineries
 and residual oil consumed at iron and steel mills.
"Excludes natural gas consumed in cement manufacturing, petroleum refining, the  iron and
 steel industry,  glass manufacture, and at crude petroleum and natural gas production
 facilities.

-------
                                    SECTION 2
               ANALYSIS OF 1985 TRENDS AND NAPAP ESTIMATES

       A detailed analysis of the differences and a determination of the validity of emission
estimates presented by both sources is provided in this section.  The analysis of the
differences is limited to the source categories which appear in the TRENDS method.  This
section is divided into three subsections: combustion, non-ferrous smelting, and other
industrial categories.  Combustion includes coal, oil, natural gas, and miscellaneous fuels from
all industrial processes except the largest fuel-consuming industries (specifically cement
manufacturing, petroleum refining, iron and steel processing, glass manufacturing, and lime
manufacturing).  Non-ferrous smelting includes primary copper, primary zinc, primary lead,
primary aluminum, and secondary lead. Other industrial processes include the pulp and paper
industry, chemical manufacturing (sulfuric acid and carbon black production), petroleum
refining, the iron and steel industry, and the minerals processing industries (cement, glass, and
lime manufacturing).

       A table summarizing the emission estimates for the source categories is presented at
the beginning of each subsection.  The summary table lists the applicable Source
Classification Codes (SCC) and Standard Industrial Classification (SIC) codes for each source
category.  In addition, the emissions are broken down to illustrate the differences between the
NAPAP and TRENDS estimate.  Except for fuel combustion, these industrial categories are
wholly represented in the stationary point source NAPAP categories.

       Each subsection holds to a standard format for comparison of the TRENDS and
NAPAP emission estimates by industrial category. Each of the source  category discussions
begins with a comparison of the overall estimate for the subcategory.  Following this
comparison, the remaining discussion is divided into six parts:
      TRENDS Activity
      TRENDS Emission Factor
      TRENDS Emissions
      NAPAP Activity

-------
      NAPAP Emissions
      Conclusion
       Within the first three divisions covering TRENDS, the TRENDS estimate is recreated
and  discussions of the derivation of the activity data, the emission factor, and the  final
emission estimate are provided.  Many of the TRENDS emission estimates do not  include all
of the emission  sources within the source category. For example, there may be three types of
furnaces in a metallurgical operation and the TRENDS method may only provide emissions
estimates from two of the three furnace types.

       The next two divisions provide the NAPAP activity and emission estimate for the
corresponding emissions categories. It is important to note that the NAPAP emissions are
reported emissions and have probably been derived using source tests, mass balance and both
AP-42 and State emission factors, along with source-specific control information.  Additional
emissions (not included explicitly in TRENDS) for the source category which are included in
NAPAP are discussed.  For the above metallurgical example, the NAPAP emission estimate
for the third type of furnace is also  listed.

       Finally, conclusions on the validity of the estimates are made.  The conclusions
section includes a "revised" TRENDS estimate where potential errors or gaps in the TRENDS
method are discovered.

2.1  COMBUSTION SOURCES

       The TRENDS combustion categories are separated into coal, oil, natural gas,
miscellaneous fuels, and wood.  Coal  combustion is developed from two numbers,  one for
anthracite coal and one for bituminous and lignite.  Oil combustion emissions are calculated
for distillate and residual oil.  Miscellaneous fuels include coke, coke oven gas, kerosene, and
liquified petroleum gas (LPG).  NAPAP emissions are source emissions derived from a
national fuel balance step which compares published fuel use from several DOE publications
to fuel use reported among point sources.  Any fuel unaccounted in point sources is assigned

-------
to the relevant area source category. Emissions are estimated using standard emission factors
and DOE point source data on fuel sulfur.  A summary of the differences in the emission
estimates for these combustion categories is presented in Table 2-1.

2.1.1  Oil Combustion

       The 1985 TRENDS emission value for industrial oil combustion was 540,000 tons
SO2.  The TRENDS estimate includes residual and distillate oil and excludes oil consumed
from cement plants and petroleum refineries, and residual oil consumed by  steel mills.  The
emission estimates for these industry-specific  uses are included in the specific-industry
estimates.

       The 1985 NAPAP value for distillate and residual oil, including external combustion
boilers and in-process fuel use, and excluding cement plants, petroleum refineries, and
residual oil consumed by steel mills, is 107,385 + 605,200 = 712,600 tons SO2.

       The apparent difference between the two inventories is 117,600 tons of SO2
(26 percent).  Table 2-2 summarizes the fuel consumption and emission estimates for the
1985 TRENDS and NAPAP inventories.

2.1.1.1  Distillate Oil

TRENDS activity

       Distillate oil consumption is calculated in the TRENDS method by subtracting the
quantity of distillate oil consumed by cement  plants and petroleum refineries from the
"adjusted" quantity of distillate oil sales to industrial and  oil companies.

       The "adjusted" quantity of distillate oil sales to industrial and oil companies  in 1985 is
obtained  from Table 13 "Adjusted Sales of Distillate Fuel Oil by End Use in the United
States:  1985-1989" of Fuel Oil and Kerosene Sales3 For 1985, the value is:

-------
TABLE 2-1. EMISSIONS FROM COMBUSTION TRENDS VERSUS NAPAP
Industrial Combustion
Anthracite
External combustion
In-process fuel
Anthracite (total)
Bituminous
Subbituminous
External combustion
In-process fuel

Lignite
External combustion
In-Process fuel
Bituminous and Lignite (total)'
Coal (total)
Residual Oil
External Combustion
Internal Combustion

In-process fuel use
Residual Oil (totalf
Distillate Oil
External Combustion
Internal Combustion

Space Heaters
In-process fuel
Distillate Oil (total)*
Oil (total)
sec

1-02-001
3-90-001


1-02-002
3-90-002
Area

1-02-003
3-90-003



1-02-004
1-02-014-04
2-02-005-01
Area
3-01-900-12
3-05-900-02
3-90-004


1-02-005
1-02-014-03
2-02-001
Area
1-05-001-05
3-04-900-01
3-90-005f


1985 TRENDS 1985 NAPAP
(tons)' (tons)"

10,998
177
900 77,200

1,272.795
3,137
356,000

78,417
72
1,687,400 1,710,000
1,840,000 1,721,000

349,932
96
242,000
13,132
460,000 67S,226

47,557
727
55,000
578
3,512
72,500 707,400
540,000 712.finn
                        (continued)

-------
   TABLE 2-1. EMISSIONS FROM COMBUSTION TRENDS VERSUS NAPAP
                                 (Continued)
Industrial Combustion
Natural Gas (Boilers)
External Combustion

Space Heaters
Internal Combustion
In-process fuel
Natural Gas Boilers (total/
Natural Gas Production (total)
sec

1-02-006
1-02-014-01
Area
1-05-001-06
2-02-002
3-01-900-03
3-01-900-13
3-03-900-03
3-04-900-03
3-07-900-03
3-07-900-13
3-90-006
3-99-900-03
3-99-900-13

3-10-900-04
3-10-900-14
Other8
1985 TRENDS 1985 NAPAP
(tons)8 (tons)"

19,085
1,000
11
1,501
11,223
1,400 32,800
400 7,660
Miscellaneous Fuels
 Coke
   External Combustion

 Coke (totalf
 Coke-oven Gas
   External Combustion
   In-process fuel
1-02-008
1-02-007-07
3-90-007-01
3-90-007-02
3-90-007-89
                         36,000
11,253

77,253

     2
 2,687
 Coke-oven Gas (total)1
 Kerosene
  Internal Combustion
 Kerosene (total)
2-02-009
                         43,255
                           2,497
 2,659

   421
   427
                                 (continued)

-------
TABLE 2-1. EMISSIONS FROM COMBUSTION TRENDS VERSUS NAPAP
                         (Continued)
Industrial Combustion
Liquified Petroleum Gas
External Combustion
Space Heaters
In-process fuel
Liquified Petroleum Gas (total)
Miscellaneous Fuels (total)
Wood
Wood/Bark Waste

In-process fuel
Wood (total)
Other NAPAP Industrial Fuel
Combustion Categories
External Combustion
Process Gas
Bagasse
Solid Waste


Liquid Waste
Internal Combustion
Gasoline and Diesel
In-Process Fuel Use
Process Gas
sec

1-02-010
1-05-001-10
3-90-009-89
3-90-010-89



1-02-009
Area
3-90-008-89
3-90-009-99



1-02-007-02
1-02-007-10
1-02-007-99
1-02-014-02
1-02-011
1-02-012
5-03
Area
1-02-013

2-02-003
2-02-004

3-03-900-24
3-90-007-97
3-90-007-99
3-99-900-24
1985 TRENDS 1985 NAPAP
(tons)" (tons)"

27
18
7
709 52
80,000 14,400

15,141
17,000
9,568
10,000 41,700


44,804
170
2,753
10.686
1,000
12.199

135

1,233
                         (continued)
                            10

-------
     TABLE 2-1.  EMISSIONS FROM COMBUSTION TRENDS VERSUS NAPAP
                                        (Continued)
         Industrial Combustion
      SCC
 1985 TRENDS
	(tons)8
1985 NAPAP
   (tons)"
    Liquid Waste
   Miscellaneous Processes
 Other NAPAP Industrial Fuel
 Combustion Categories
 Total
3-90-013
3-99-999
                           2,470,000
                               166
                               717
                            74,500

                         2,597,000
'  National Air Pollutant Emission Estimates. 1900 -1991, EPA^54/R-92-013, October 1992.
"  The 1985 NAPAP Emissions Inventory (version 2):  Development of the Annual Data and Modelers' Tapes, EPA-
600/7-89-012a, November 1989.
c  Excludes bituminous and lignite burned in cement and lime kilns.
d  Excludes residual oil burned in cement plants, petroleum refineries and iron and steel mills.
'  Excludes distillate oil burned in cement plants and petroleum refineries.
'  Excludes natural gas burned in cement plants, petroleum refineries, iron and steel mills, glass manufacture, and oil
and natural gas production.
1  Other includes all natural gas combustion emissions with an SIC code of 1311 or 1321.
k  Excludes iron and steel industry.
'  Excludes iron and steel industry.
                                            11

-------
              TABLE 2-2.  COMPARISON OF OIL COMBUSTION
                           VALUES FOR 1985 TRENDS AND NAPAP
Category
Distillate Oil
Emissions (tons)
Fuel Consumed (106 gallons)
Residual Oil
Emissions (tons)
Fuel Consumed (106 gallons)
Total Oil Combustion
Emissions (tons)
Trends
Published


3,426.8


3,562
540,000

Trends
Calculated

72,500
3,429.8

460,000
3,555
532,500

NAPAP
Published

107,400
1,902

605,200
5,615
712,600

       2,592,678,000 + 876,505,000 = 3,469,183,000 gallons.

       The quantity of oil consumed by cement plants is obtained from Table 7, "Clinker
Produced in the United States, by Fuel" of Minerals Yearbook 1986 "Cement".* The quantity
for 1985 is 755,000 barrels (31,710,000 gallons).  TRENDS assumes that one-third of the oil
consumed is distillate oil  (10,570,000 gallons).

       The quantity of distillate oil consumed by petroleum refineries is obtained from
Table 43, "Fuels Consumed at Refineries by PAD District, 1985" of the Petroleum Supply-
Annual.5  The figure for 1985 is 758,000 barrels (31,836,000 gallons).

       Therefore the TRENDS  activity number for industrial distillate oil consumption is:

3,469,183,000 - 10,570,000  31,836,000 = 3,426,777,000 gallons.

       The value in the current TRENDS activity data file is 3,426,800,000 for 1985.
                                        12

-------
TRENDS emission factors

       The emission factors cited in the TRENDS procedure are for external combustion of
grades 1 and 2 oil (combined) and grade 4 oil.  The emission factors listed in the TRENDS
procedure are provided below.
  SCC             Description                      Emission Factor  Units
  1-02-005-01      External Combustion                      143.6S  lbs/103 gallons
                   Industrial: Distillate grades 1                       burned
                   and 2 oil
  1-02-005-04      External Combustion -                    150.0S  lbs/103 gallons
                   Industrial: Distillate grade 4                       burned
                   oil

       The 1985 sulfur content for distillate oil is obtained from Table 1 "Summary of
Grade 1 Fuels" and Table 2 "Summary of Grade 2 Fuels" of Heating Oils6 which reports
average sulfur contents for five regions of the country for both grade 1 and 2 oil.  This report
does  not include an average sulfur content for grade 4 oil due to a lack of data. The average
sulfur content for grade 1 oil ranged from 0.042 to 0.123 weight percent. The average sulfur
content for grade 2 oil ranged from 0.228 to 0.267 weight percent. The average sulfur
contents were averaged across grade and region based on the number of samples, and the
average grades 1  and 2 sulfur content was determined to be 0.206 weight percent.  By
contrast, the  average sulfur contents reported through NAPAP (weighted on the number of
records reporting fuel sulfur content) are 0.36 for grades 1 and 2 oil and 0.85 for grade 4 oil.
The cement procedure (see Section 2.3.7) cites a distillate  oil sulfur content of 0.3 percent.

       The TRENDS procedure requires that the emission factors be weighted based on  AIRS
data and provides a table to assist in the weighing. Table 2-3 illustrates the values used to
produce a weighted emission factor for distillate oil combustion. Table 2-3 requires the
emission factor for internal combustion engines burning distillate oil.  For turbines, the
emission factor is  140.0S lbs/103 gallon burned and for reciprocating engines, the emission
factor is 31.22 lbs/103 gallon  burned.7 Table 2-3 also requires activity data obtained from the
1985  NAPAP emission inventory.  Average  sulfur content for grades  1 and 2 oil are from
                                           13

-------
Heating Oils as derived above and, for the other SCCs, the average NAPAP sulfur contents
were used.
            TABLE 2-3. WEIGHTED AVERAGE EMISSION FACTOR
                        FOR DISTILLATE OIL COMBUSTION
Combustion Category
Boilers burning grade 1 or 2
Boilers burning grade 4
Turbines
Internal Combustion Engines
Weighted Average (lb/103 gal)
NAPAP
Consumption
(103 gallons)
843,851'
125,949"
352b
399

AP-42 S02
Emission Factors
(lb/103 gal)
143.6S
150.0S
140.0S
31.2
42.3
Average sulfur
content
(weight percent)
0.206
0.85
0.27


  1 Heating Oils, 1985
  " NAPAP
       The weighted average emission factor was calculated by multiplying the activity
 number (consumption) by the emission factor (and the sulfur content as needed) and dividing
 by the total activity.  The average emission factor for 1985 was calculated as 42.3 lb/103
 gallon.  This value is higher than the emission factors that were used in the 1990 (39.2 lb/103
 gallon) and the 1991 (38.5 lb/103 gallon) TRENDS estimates.

       The TRENDS procedure calls for an SO2 control efficiency, which is obtained from
 EIA  767 data.  If a control efficiency is applied, the TRENDS value  is not documented and
 the control efficiency does not appear in the TRENDS spreadsheets. In addition, the EIA-767
 data pertains  to utility boilers and the  suitability of transferring control data from the utility
 sector to the  industrial sector has not been investigated or documented.
                                          14

-------
TRENDS emissions

       The TRENDS emission estimate for distillate oil in 1985 is:

       3,426,777 * 42.3 / 2000 = 72,476 tons of SO:.

NAPAP activity

       The overall NAPAP activity should match the TRENDS number.  During the
development of the NAPAP inventory, the same national industrial distillate oil consumption
value was used to produce a "fuel balance" among point and area sources. In NAPAP, the oil
consumed by point sources [as reported through the National Emissions Data System
(NEDS)] is subtracted from the national number and any remaining balance is allocated to the
area source inventory. As shown in Table 2-1, there are more emissions reported among area
sources than for external combustion point sources. This may be due to the under reporting
of fuel usage by point sources.  The reporting of fuel usage was a secondary priority in
NAPAP and many sources consider throughput data confidential and did not report it. In
addition, the entire State of Texas did not report industrial  fuel usage by plant, instead fuel
usage was reported at the county level.

       The distillate oil activity data reported through  NAPAP include external combustion
boilers, space heaters, internal combustion engines, and in-process fuel use. To  better
compare the TRENDS and NAPAP combustion estimates,  the fuel used in the cement and
petroleum refining industries must be subtracted from  the total NAPAP estimate.
Unfortunately,  for the NAPAP inventory development effort, fuel usage was not a priority
data element and facilities were not required  to report  their fuel usage.  Therefore, subtracting
the fuel used by these industries as reported in  NAPAP will still not provide a reliable value
for comparing  the fuel combustion and will in fact make the two estimates more dissimilar.

       The emission factors for industrial distillate oil combustion range from 31.2 for
internal combustion reciprocating  engines to  143.6S for the majority of the boiler descriptions.

                                         15

-------
In addition, the emission factor for distillate in-process fuel use (general) is 0.0, and it is
intended that this emission factor be used to report fuel usage for industries where the process
emission factor accounts for the fuel sulfur emissions. Both cement and lime kilns have
substantially lower emission factors (98S and 72S respectively), due to the affinity for sulfur
of both lime and cement.

NAPAP emissions
       The NAPAP industrial distillate oil emissions are reported as follows:
 SCC              Description
 1-02-005-01       External combustion boilers - Industrial: Distillate
                   oil, grade  1 and 2
 1-02-005-02       External combustion boilers - Industrial: Distillate oil
 1-02-005-03       External combustion boilers - Industrial: Distillate oil
 1-02-005-04       External combustion boilers - Industrial: Distillate
                   oil, grade  4
 1-02-014-03       External combustion boilers - Industrial: Distillate
                   oil, cogeneration
 1-05-001-05       External combustion boilers - Space heaters,
                   Industrial, Distillate oil
 2-02-001-01       Internal combustion engines - Industrial, Distillate
                   (Diesel) oil, turbine
 2-02-001-02       Internal combustion engines - Industrial, Distillate
                   (Diesel) oil, reciprocating
 2-02-001-03       Internal combustion engines - Industrial, Distillate
                   (Diesel) oil, turbine cogeneration
 3-04-900-01       Secondary metal, Process heaters, distillate oil
 3-90-005-01       In process fuel use, Distillate  oil
 3-90-005-02       In process fuel use - Distillate Oil: Cement kiln/dryer
 3-90-005-03       In process fuel use - Distillate Oil: Lime kiln
 3-90-005-89       In process fuel use, Distillate  oil. General
 3-90-005-99       In process fuel use. Distillate  oil, General
 Total
Tons SO. Emitted
           39,359

            2,071
             800
            3,625
            3,836

             578

             284

             399

               68

                8
               59
            2,404
               12
            1,624
            1.827
                                                                                  56,954
       In addition, NAPAP reports area source emissions for distillate oil of 55,000 tons
This results in total NAPAP distillate oil emissions of 111,954 tons of SO,.
                                            16

-------
       To compare the NAPAP estimate with the TRENDS estimate, emissions for cement
kilns/dryers and petroleum refineries must be subtracted from this total.  For the above listed
SCCs, NAPAP reports that 2,450 tons are emitted from cement manufacturing (SIC 3241) and
2,119 tons are emitted at petroleum refineries (SIC 2911).  This results in 107,385 tons of
SO2 being emitted through distillate oil combustion excluding cement manufacturing and
petroleum refineries.

Conclusion

       The distillate oil estimates are 72,480 tons for TRENDS versus 107,358 tons for
NAPAP. The NAPAP point source estimate and the TRENDS  estimate  are quite similar,
although no conclusion can be drawn from this similarity.  This may indicate that area source
emissions were overestimated due to an underestimate of fuel use in the point source
inventory.

       There  is a huge discrepancy in the reported quantity of oil bumed in the TRENDS
estimate versus the NAPAP  inventory.  TRENDS reports 3,426 x 106 gallons and NAPAP
reports approximately 2,000 x  106 gallons.  Table 3 "Total Inputs of Energy for Heat, Power,
and Electricity Generation by Census Region, Industry Group and Selected Industries, 1985"
of Manufacturing Energy Consumption Survey: Consumption of Energy, 1985* reports
31,684,000 barrels (bbls) consumed (1,330.7 x 106 gallons).  This survey reports an activity
that is approximately one third the value reported in TRENDS.  If the value  used in TRENDS
was also  used to  compute an area source activity in NAPAP, this could  have resulted in a
large overestimation in distillate oil consumed and resulting emissions reported in NAPAP
through the area source category.

       The development of an average sulfur content for distillate oil is not  well documented
in the TRENDS procedure.  The sulfur content assumptions have a large impact on the
overall emission factor. Additional effort should be expended to determine a reasonable
average sulfur content, to determine if industrial distillate oil consumers are  electing to use
lower sulfur content oil, and if so, the overall effects on emissions from this category.

                                         17

-------
2.1.1.2  Residual Oil

TRENDS activity

       The residual oil consumption is calculated by subtracting the quantity of oil consumed
by cement plants, petroleum refineries, and steel mills  from the "adjusted" quantity of residual
oil sales to industrial and oil companies.

       The adjusted quantity of residual oil sales to industrial and oil companies in  1985 is
obtained from Table 14 "Adjusted Sales of Residual Fuel Oil by End  Use in the United
States:  1985-1989"  of Fuel Oil and Kerosene Sales 1989? For 1985, the value is:

       4,011,361,000 + 776,019,000 = 4,787,380,000 gallons.

       The quantity of oil consumed by cement plants is obtained from Table  7  "Clinker
produced in the United States, by fuel" of Minerals Yearbook 1986 "Cement".4 The quantity
for 1985 is 755,000 barrels (31,710,000  gallons). TRENDS assumes that two  thirds of the oil
consumed is residual oil (21,140,000 gallons).

       The quantity of residual oil consumed by petroleum refineries  is obtained from
Table 43 "Fuels Consumed at  Refineries by PAD District, 1985"  in the Petroleum Supply
Annual5  The figure for 1985  is 13,326,000 barrels (559,692,000 gallons).

       The quantity of residual oil consumed by steel  mills was calculated by multiplying the
quantity in tons of raw steel produced in 1985  by a conversion factor for the value of residual
oil consumed per ton  of raw steel produced.  The conversion factor used is 7.38 gal/ton raw
steel. The quantity of raw steel produced is  obtained from the Bureau of Economic Analysis,
Survey  of Current Business. The quantity of raw steel produced  is more readily available
from Table  1 "Salient Iron and Steel Statistics" of Minerals Yearbook 1986 "Iron and Steel."-
The  value for 1985, in both references, is 88,259,000 short tons.  The quantity of residual oil
consumed by steel mills is calculated  as:
                                           18

-------
is:
       88,259,000 * 7.38 gal/ton raw steel = 651,350,000 gallons.

       Therefore the TRENDS activity value for residual oil consumed by industrial sources


       4,787,380 - 21,140 - 559,692 - 651,350 = 3,555,198 x 103 gallons.

       The value in the TRENDS activity table is 3,562.300 x 103 gallons.

TRENDS emission factors

       The emission factor cited in TRENDS is for grade 6 oil, SCC 1-02-004-01, and the
value is:

       158.6S lbs/103 gallons burned.

       This emission factor matches the AIRS Facility Subsystem Source  Classification Codes
and Emission Factor Listing for Criteria Air Pollutants7 document   The average sulfur
content of grade 6 Fuel Oil is obtained from Heating Oils, J985.6  Again, this reference
provides averages for each of five regions.  The average sulfur contents range from 1.20 to
1.75  percent in a total of 44 samples.  The average national figure based  on the number of
samples is 1.63 percent. The average  sulfur content  for residual oil can also be obtained from
Table 11 "Receipts of No. 6 Fuel Oil at Electric Utilities" of Cost and Quality of Fuels for
Elecmc Utility Plants 1985.9  For 1985, the average  sulfur content of grade 6 fuel  oil was
1.09  percent by weight. The emission factor for residual oil (based on the Heating Oils,
79S56 data) is:

       158.6 * 1.63 = 258.5 lb/103 gallons burned.
                                          19

-------
TRENDS emissions
       The emissions are calculated as:
       258.5 * 3,555,198 / 2000 = 459,510 tons of SO:.
NAPAP activity



       The residual oil activity data reported through NAPAP include external combustion

boilers, space heaters, internal combustion engines, and in-process fuel use.  Collection and

quality assurance of all throughput data were not a priority item in the NAPAP inventory.

The NAPAP inventory also includes approximately 932,000 x 106 gallons of crude oil burned

during production in the total residual oil throughput value.  This oil was placed in the

residual oil category because no other category existed in NAPAP.



NAPAP emissions



       The estimated emissions reported through NAPAP for residual oil combustion are

listed below:


  SCC             Description                                       Tons SO. Emitted

  1-02-004-01       External combustion boilers - Industrial: Residual               349,348
                   oil, grade 6 oil

  1-02-004-02       External combustion boilers - Industrial: Residual                39 304
                   oil, 10-100 MMBTU/hr

  1-02-004-03       External combustion boilers - Industrial- Residual                 2 676
                   oil, <10 MMBTU/hr

  1-02-004-04       External combustion boilers - Industrial: Residual                 5,836
                   oil, grade 5

  1-02-004-05       External combustion boilers - Industrial: Residual                 1 655
                   oil, Cogeneration

  1-02-004-06       External combustion boilers - Industrial: Residual                     4
                   oil

  1-02-014-04       External combustion boilers   Industrial: CO Boiler,                375
                   Residual oil


                                          20

-------
  SCC            Description                                        Tons SO. Emitted
  2-02-005-01      Internal combustion engines - Industrial, Residual/                   96
                  Crude oil, reciprocating
  3-01-900-12      Chemical Manufacturing, Incinerators, Residual oil                  657
  3-05-900-02      Mineral Products, Process Heater, Residual oil                     134
  3-90-004-02      In-process fuel use. Residual oil Cement Kiln/dryer              3,574
  3-90-004-03      In-process fuel use, Residual oil. Lime Kiln                        738
  3-90-004-89      In-process fuel use. Residual oil. General                       19,451
  3-90-004-99      In-process fuel use, Residual oil, General                        1.644
  Total                                                                      425,493
       In addition, the NAPAP estimate includes 242,000 tons of SO, from the combustion of
 residual fuel oil by area sources, bringing the NAPAP total for residual  oil combustion to
 667,493 tons of SO2.

       To compare the NAPAP and TRENDS estimates, it is necessary to exclude the
 residual oil emissions from cement manufacturing, petroleum refineries, and steel mills.  The
 emission estimates for the above listed SCCs are 3,807 tons for cement  manufacturing (SIC
 3241); 44,208 tons for petroleum refining (SIC 2911); and 14,318 tons for iron and steel mills
 and foundries (SIC 3312 and 3325).  Subtracting these emissions results in a NAPAP estimate
 of 605,160 tons of SO,.

 Conclusion

       The TRENDS estimate is 459,510 tons for residual oil excluding cement plants,
petroleum refineries and steel mills. The NAPAP estimate for the same category of emissions
is 605,160 tons of SO2. The two estimates would be much closer if the area source
component of the NAPAP total were not included.

       There is a huge discrepancy in the reported quantity of residual  oil burned in the
TRENDS  estimate versus  the NAPAP inventory.  TRENDS reports 3,555 x 106 gallons and
NAPAP reports approximately 6,000 x 106 gallons.  Table 3  "Total Inputs of Energy for Heat.

                                         21

-------
Power, and Electricity Generation by Census Region, Industry Group and Selected Industries,
1985" of Manufacturing Energy Consumption Survey: Consumption of Energy,  1985s reports
80,252,000 bbls consumed (3,370.6 x  106 gallons).

2.1.2  Coal Combustion

       The 1985 TRENDS SO: emission value for coal is comprised of two separate
categories; anthracite, and bituminous coal and lignite.  The published TRENDS emission
estimate of 1,840,000 tons of SO; excludes emissions from bituminous coal and lignite
consumed at cement and lime manufacturing facilities.  The 1985 NAPAP value, excluding
bituminous coal and lignite consumed at cement and lime manufacturing facilities, was
 1,721,000 tons of SO2.  The apparent difference between the two inventories is 119,000 tons
of SO2 (7 percent).  Table 2-4 illustrates the  differences in the coal consumption and emission
estimates  for the 1985 TRENDS and NAPAP inventories.
                TABLE 2-4.  COMPARISON OF COAL COMBUSTION
                             VALUES FOR 1985 TRENDS AND NAPAP
Category
Anthracite
Emissions (tons)
Fuel Consumed (103 tons)
TRENDS
Published


658.8
TRENDS
Calculated

10,900
800
NAPAP
Published

11,000
522
  Bituminous & Lignite
   Emissions                                        1,687,000        1,710,000
   Fuel Consumed (106 tons)              61.6              62.131            74.535
  Total Coal Combustion            1,840,000          1,698,000        1,721,000
  Emissions (tons)
       The total published TRENDS estimate is 1.84 x 106 tons SO2. In following the
 TRENDS procedure, the emission estimates for anthracite as a category and bituminous and
 lignite as a category are 10,900 and 1,687,000 tons of S(X respectively. The total of
 1,698,000 tons SO: differs from the published TRENDS estimate by 152,000 tons SO,.
                                         22

-------
       The sulfur contents and resulting emission factors that were actually used in 1985 are
not documented and may explain the difference.  In addition, the TRENDS procedure manual
refers to the use of control assumptions as documented in the EIA-767 data.  If control
assumptions have been applied (EIA-767 data pertain to utilities and should not be used to
estimate controls on the industrial sector), they are  not documented.

2.1.3 Anthracite Coal

2.1.3.1  1985 TRENDS  Activity

       Anthracite combustion activity obtained from the distribution of anthracite coal from
District 24 (District 24 is the anthracite-producing district of Pennsylvania) to industrial users
except coke plants.  The value for 1985 is obtained from Table 6 "Distribution of U.S. Coal
by Origin, Destination, and Consumer" of Coal Distribution.10  The value for  1985  is
800,000 short tons.

       The TRENDS activity data spreadsheet has a 1985 activity value of 658,800 tons. The
wording in the TRENDS procedure document is "obtain the distribution of anthracite from
Pennsylvania to industrial less coke plants."  This could be interpreted as direction  to subtract
the coke coal (29,000 short tons), but this would  be an error (because the coal coke is already
excluded from the value) and it does  not resolve the difference between the two values.

2.1.3.2  TRENDS Emission  Factors

       Emission factors  for three types of boilers, pulverized, traveling grate stoker, and
hand-fired, that burn anthracite coal are 39.0S Ibs/ton burned.  The TRENDS method assumes
a sulfur content of 0.7 percent. This  provides an emission factor of:

       39.0 * 0.7 = 27.3 Ib/ton burned.
                                          23

-------
2.1.3.3  TRENDS Emissions

       TRENDS emissions were calculated using the activity value found in the TRENDS
spreadsheet

       658,800 * 27.3 / 2000 = 8,990 tons of SO.

       Emissions calculated using the activity value derived following the TRENDS
procedure manual results in emissions of:

       800,000 tons * 27.3 / 2000 = 10,920 tons of S0:.

2.13.4 NAPAP Activity

       The NAPAP activity value for anthracite coal combustion is based on the SCC codes
for industrial external combustion and in-process fuel use. However, collection and quality
assurance of all throughput data were not priority items.

2.13.5 NAPAP Emissions


 SCC             Description                                       Tons SO. Emitted
 1-02-001-01      External combustion boilers - Industrial: Anthracite               9,099
                  coal, Pulverized coal
 1-02-001-02      External combustion boilers - Industrial: Anthracite                 17
                  coal
 1-02-001-04      External combustion boilers - Industrial: Anthracite               1  872
                  coal, Traveling grate, (overfeed)  stoker
 1-02-001-07      External combustion boilers - Industrial: Anthracite                 10
                  coal, Hand-fired
 3-90-001-89      In process fuel use, Anthracite coal, General                        99
 3-90-001-99      In process fuel use, Anthracite coal, General                        7g
 Total                                                                        1U75
                                         24

-------
2.1.3.6  Conclusion

       The emission estimates are in agreement, 11,000 tons of S02 in both NAPAP and
TRENDS.  The NAPAP inventory does not report all of the anthracite coal that is consumed,
and it is unclear what fraction of the anthracite coal combustion is represented  by the
emissions that are reported through NAPAP. The  NAPAP estimates result in an overall
emission factor of 42.8 Ibs/ton of coal burned.  This translates into an average  sulfur content
of 1.1 percent.  The reported  sulfur content for anthracite coal varies in the NAPAP
inventory.  The average reported for pulverized coal is 1.15 percent (3 records  reporting).
The average reported for traveling grate (overfeed) stoker and the majority of records
reporting is 0.7 (24 records).  The assumptions on  sulfur content have a large influence on the
emission estimate.

       The activity value that is published in the TRENDS activity spreadsheet could not be
replicated.  The NAPAP inventory may overestimate SO; emissions relative to  the reported
quantity of coal burned but anthracite coal is a relatively minor category.  Additional research
into a reasonable  average sulfur content for anthracite coal is warranted.

2.1.4 Bituminous Coal and  Lignite

2.1.4.1  TRENDS Activity

      The 1985 TRENDS activity for bituminous coal and lignite is calculated by
subtracting coal consumed in  lime and cement plants from the national total for industrial
users. The national value is obtained from Table 23 "Coal Consumption by End-use Sector"
of Quarterly Coal Report.11 The 1985 value for "other industrial" is 75,317,000 tons.

      The consumption by cement plants is obtained from Table 7 "Clinker produced in the
United States, by  fuel" of Minerals Yearbook 1986 "Cement.1*  In 1985, 11,606,000 tons of
coal were consumed by cement plants.
                                          25

-------
       National annual lime production is obtained from "Facts and Figures for the Chemical
Industry" published in Chemical & Engineering News.12 The 1985 primary production
figure was 15,800,000 short tons. Coal consumption is estimated using a multiplier of 0.1 ton
coal/ton lime produced.  Therefore, the estimated coal consumption by lime plants is:

       15,800,000 * 0.1  =  1,580,000 tons.

       The TRENDS bituminous coal and lignite consumption is:

       75,317,000 - 11,606,000 - 1,580,000 = 62,131,000 tons.

       The TRENDS activity spreadsheet has a value of 61,600,000 tons of coal consumed.

2.1.4.2 TRENDS Emission Factors

       The documented  procedure for determining an overall emission factor for this category
is complex and the documentation states  that the procedure has not been applied.  The
primary complexity is in the development of an average sulfur content.  The TRENDS
procedure uses an overall emission factor of 38.IS Ib SO2/ton coal burned.  In the complex
procedure, an average sulfur content is developed from each coal production district and these
average sulfur contents are weighted based on shipments data listed in Coal Distribution.10

       Recent TRENDS procedures for developing an emission factor for this category are
unclear.  The emission factor for 1990 was 54.3 Ib/ton  burned and the emission factor for
1991 was 51.5 Ib/ton burned. Why and how these factors were changed is unclear.  To back
calculate the 1990 sulfur content  using the emission factor of 38.IS Ib/ton burned results in
an overall sulfur content of 1.4 percent:

       54.3 / 38.1 =  1.4 percent sulfur.
                                          26

-------
2.1.4.3  TRENDS Emissions



       TRENDS emissions, using the  1990 emission factor and the activity value in the

TRENDS spreadsheet, are calculated as follows:



       61,600,000 * 54.3 / 2000 = 1,672,440 tons of SO:.



       Emissions calculated using the  activity data derived following the TRENDS procedure

manual results in emissions of:



       62,131,000 * 54.3 / 2000 = 1,686,900 tons of SO2.



2.1.4.4  NAPAP Activity



       Activity data were not priority  items for collection and quality assurance for the entire

NAPAP inventory. Activity data are not reproduced here by SCC.



2.1.4.5  NAPAP Emissions



       The following emission estimates were reported for bituminous coal and lignite

combustion.


 SCC             Description                                       Tons SO. Emitted

 Bituminous:

 1-02-002-01      External combustion boilers - Industrial: Bituminous            101.735
                  coal, Pulverized coal: wet bottom

 1-02-002-02      External combustion boilers - Industrial: Bituminous            571.457
                  coal, Pulverized coal: dry bottom

 1-02-002-03      External combustion boilers - Industrial: Bituminous             81.758
                  coal, Cyclone furnace

 1-02-002-04      External combustion boilers - Industrial: Bituminous            330.099
                  coal, Spreader stoker

 1-02-002-05      External combustion boilers - Industrial: Bituminous            102.120
                  coal, Overfeed stoker
                                         27

-------
sec
1-02-002-06

1-02-002-10

1-02-002-12

1-02-002-17

1-02-002-19

Subbituminous:
1-02-002-21

1-02-002-22

1-02-002-24

1-02-002-25

Lignite:
1-02-003-01

1-02-003-03

1-02-003-04

1-02-003-06

In-process fuel
use:
3-90-002-01

3-90-002-03
3-90-002-89
3-90-002-99
3-90-003-89
Total
Description
External combustion boilers - Industrial: Bituminous
coal, Underfeed stoker
External combustion boilers - Industrial: Bituminous
coal, Overfeed stoker
External combustion boilers - Industrial: Bituminous
coal, Pulverized coal: Dry bottom (Tangential)
External combustion boilers   Industrial: Bituminous
coal, Atmospheric Fluidized bed
External combustion boilers - Industrial: Bituminous
coal, Cogeneration
External combustion
Subbituminous coal,
External combustion
Subbituminous coal,
External combustion
Subbituminous coal,
External combustion
Subbituminous coal,
stoker
boilers - Industrial:
Pulverized coal: wet bottom
boilers - Industrial:
Pulverized coal: dry bottom
boilers - Industrial:
Spreader stoker
boilers - Industrial:
Traveling grate (overfeed)
External combustion boilers - Industrial: Lignite,
Pulverized coal
External combustion boilers - Industrial: Lignite,
Cyclone furnace
External combustion boilers - Industrial: Lignite,
Traveling grate (overfeed) stoker
External combustion boilers - Industrial: Lignite,
Spreader stoker
In process fuel use, Bituminous coal, Cement
kiln/dryer
In process fuel use, Bituminous coal, Lime kiln
In process fuel use, Bituminous coal, General
In process fuel use, Bituminous coal, General
In process fuel use, Lignite, General
SO. Emitted
     18,088

       1,518

     20,286

       5,438

     12,128


         83

       8,547

     11,195

       8,396


     45,759

        341

     29,905

       2,412
                                          77,859

                                           6,384
                                           6,828
                                          10,925
                                         	72
                                       1,453,333
                                         28

-------
       In addition, NAPAP reports area source emissions of 356,000 tons of SO2 from coal
combustion.  This brings the total combustion from bituminous coal, subbituminous coal and
lignite to  1,809,333 tons.  In order to compare the NAPAP and TRENDS estimates, emissions
from cement and lime kilns must be excluded from the NAPAP estimate.  For the above
listed SCCs, 92,524 tons are for cement manufacturing (SIC  3241) and 6,386 tons are for
lime manufacturing (SIC 3274). The adjusted NAPAP emissions for bituminous and lignite
combustion are 1,710,423 tons  of S(X

2.1.4.6  Conclusions

       The total estimates for both emissions and coal burned for this category are very close.
The total initial activity value, including coal burned in lime  manufacturing and cement
plants, is 75.3 x 106 tons of coal in TRENDS versus 74.5 x 106 tons of coal in NAPAP.
Table 3 "Total Inputs of Energy for Heat, Power, and Electricity Generation by Census
Region, Industry Group and Selected Industries, 1985" of Manufacturing Energy Consumption
Survey: Consumption of Energy, 1985 reports 59.195 x 106 tons of coal  burned in the
industrial  sector. This activity  is consistent with both the TRENDS and NAPAP totals. The
emissions, excluding lime and cement-related emissions, are  1.69 x  106 tons of SO2 versus
1.71 x 106 tons of SO2. Again, the NAPAP inventory relies  on a total fuel balance to
determine the area  source emissions for this category. For this category, the addition of the
area source emissions brings the NAPAP and TRENDS estimate closer together.

       The TRENDS method utilizes a factor of 0.1 tons of  coal consumed/ton of lime
produced. This factor should be verified.  The Manufacturing Energy Consumption Survey:
Consumption of Energy, 1985* and subsequent available editions do not separate lime from
other mineral products industries. The  survey for 1991 is expected  to provide additional
resolution and the preliminary numbers should be available in late summer of  1993.

      The three types of coal that constitute this category each have a slightly different
emission factor. The  emission factors for external combustion in industrial boilers are
generally as follows.

                                         29

-------
       Bituminous          1-02-002-01,19             39.0S
       Subbituminous       1-02-002-21,29             35.0S
       Lignite              1-02-003                  30.0S

       There are smaller emission factors for some types of bituminous coal combustion (for
example fluidized bed), but these constitute a very small amount of the coal combustion
activity.

       Sulfur contents reported in the NAPAP inventory for these SCCs range as follows.

       1-02-002-01,19              1.0 to 1.9 percent
       1-02-002-21,29              0.4 to 1.5 percent
       1-02-003                   0.5 to 0.9 percent

       Average sulfur contents for the types of coal are estimated based on emissions, and
reported coal consumption are as follows.

       1-02-002-01,19              1.4 percent
       1-02-002-21,29              0.7 percent
       1-02-003                   0.7 percent

       Use of these sulfur contents results in the following  average emission factors.

       Bituminous          1-02-002-01,19             54.6 Ib/ton  burned
       Subbituminous       1-02-002-21,29             24.5 Ib/ton  burned
       Lignite              1-02-003                  21.0 Ib/ton  burned

       The TRENDS emission factor of 54.3 is probably an overestimation.  In addition
there are probably some emission controls that are not reflected in the TRENDS method
                                          30

-------
2.1.5  Natural Gas

      The 1985 TRENDS does not publish emission estimates for categories if the total
estimated emissions are less than 10,000 tons.  The emissions from natural gas combustion
are adjusted to exclude natural gas consumed by cement plants, petroleum refineries, the iron
and steel industry,  glass manufacturing, and crude petroleum and natural gas production.
Once the TRENDS estimates are adjusted  to exclude the above mentioned categories, the
emission estimate is less than 10,000 tons  and therefore there is no published SO2 estimate for
natural gas combustion.  The adjusted NAPAP emissions are 32,800 tons S02.

      The TRENDS procedure includes a subsection adjustment to estimate the emissions
from natural gas combustion during  crude  petroleum and natural gas production. The
combustion emissions estimated in this section are combined with  sulfur recovery emissions
(estimated in Section 2.3.3) and reported in Section 2.3.5.  Table 2-5 summarizes the
information for natural gas combustion and natural gas plants and pipelines for the 1985
TRENDS and NAPAP inventories.
               TABLE 2-5.  COMPARISON OF NATURAL  GAS
                            VALUES  FOR 1985 TRENDS AND NAPAP
Category
Natural Gas Combustion
Emissions (tons)
Fuel Consumed (109 ft3)
Natural Gas Plants and
Pipelines
Emissions (tons)
Fuel Consumed (109 ft3)
Trends
Published


4,764.8


1,469.8
Trends
Recreated

1,400
4,852.3

400
1,469.8
NAPAP
Published

32,800
6,700

7,660
15
                                        31

-------
2.1.5.1  TRENDS Activity

       TRENDS separates the natural gas emissions into combustion in boilers and emissions
from gas pipelines and plants.  The emissions from the combustion of natural gas at natural
gas pipelines and plants are reported in  Section 2.3.5.
Boilers
       The total industrial consumption of natural gas consumed in industrial boilers is
 obtained from Table 26 "Natural Gas Consumption in the U.S. 1930-1985" of Natural Gas
 Annual, 79S5.13  For 1985, this value is 5,901,288 million cubic feet (106 ft3). The
 TRENDS procedure subtracts consumption by cements plants, petroleum refineries, the iron
 and steel industry, and the glass manufacture industry from the total.

       Natural gas consumption for cement plants is provided in Table 7 "Clinker produced
 in the United States, by fuel" of Minerals Yearbook 1986 "Cement."*  The value for 1985 is
 10,644.314 106ft3.

       The total natural gas  consumption by petroleum refineries is obtained from Table 43
 "Fuels Consumed at Refineries" of Petroleum Supply Annual 1985.5 The 1985 value  is listed
 as 487,830 106 ft3.

       Natural gas consumption by the iron and steel industry is estimated from the annual
 production of raw steel and a conversion factor of 4.25 x 106 ft3/103 ton raw steel  produced.
 Annual raw steel production is obtained from  Table  1 "Salient Iron and Steel Statistics" of
 Minerals Yearbook  1986 "Iron and Steel."*  The  1985 production was 88,259,000  tons
 Therefore natural gas consumption for the iron and steel industry is calculated as:

       4.25 *  88,259 = 375,100 x 106 ft3.
                                          32

-------
       Natural gas consumption by the glass manufacture industry is also estimated using the
annual glass production and a conversion factor of 10.8 x 106 ftVlO3 tons glass produced.
Annual glass consumption is computed as the sum of production of flat glass, container glass,
and miscellaneous glass products. The procedure  for calculating annual glass production is
described in Section 2.3.8.  Rat glass production in 1985 was obtained from Table 1A
"Summary of Flat Glass Production, Shipments, and Inventories: 1986 and 1985" of Current
Industrial Reports Flat Glass Summary for 1986.u Container glass production was obtained
from Table 5 "Shipments, Production and Stocks of Glass Containers:  1985" from Current
Industrial Reports Glass Containers Summary for  1986.15  Miscellaneous glass products are
assumed to be an additional 10 percent of the production of flat glass and glass containers.
Annual glass production in  1985 was 16,245,837 tons. Therefore,  the amount of natural gas
consumed by the glass manufacture industry for 1985 is:

       10.8 *  16,246 = 175,457 x 106 ft3.

       The resulting TRENDS consumption in boilers is calculated as:

       5,901,288 - 10,644 - 487,830 - 375,100 - 175,457 = 4,852,257  x 106 ft3.

       The value reported in the TRENDS activity spreadsheet is 4,764,800 x 106 ft3.

Gas pipelines and plants

       The natural gas consumption for gas pipelines and plants is the  sum of pipelines fuel
and lease and plant fuel.  These values are obtained from Table 13  "Consumption of Natural
Gas" of Natural Gas Annual, 1985.13 The value for 1985 for lease and plant fuel is 966,047
x 106 ft3 and the 1985 value for pipelines fuel is 503,766 x 106 ft3.  Therefore, the total
pipelines and plants natural  gas combustion rate is:

       966,047 + 503,766 = 1,469,813 x  106 ft3.
                                          33

-------
2.1.5.2  TRENDS Emission Factor

       The TRENDS emissions factor is 0.6 Ib SO2/106 ft3 burned.  This emission factor is
consistent with all of the natural gas combustion emission factors listed in AIRS Facility
Subsystem Source Classification Codes and Emission Factor Listing for Criteria Air
Pollutants1

2.1.5.3  TRENDS Emissions

       TRENDS emissions for industrial boilers using  the activity rate in the TRENDS
spreadsheet are:

       0.6 * 4,764,800 / 2000 = 1,429  tons of SO2.

       TRENDS emissions for natural  gas pipelines are calculated as:

       0.6 * 1,469,813 / 2000 = 441 tons of SO2.

       Because TRENDS does not report emissions of fewer than 10,000 tons from a source
category, the published TRENDS estimate is zero tons for 1985.

2.1.5.4  NAPAP Activity

Boilers

       Activity data collection  and quality assurance were not priority items for all sources
Activity data for natural gas combustion in boilers are  not itemized here.
                                          34

-------
Natural gas production

       The TRENDS activity for natural gas production is the amount burned during the
development of the natural gas (extraction, transportation, etc.).  In NAPAP, the following
two SCCs appear to correspond to the combustion of natural gas in natural gas production as
reported through TRENDS.
 SCC             Description                                           106 ft3 Burned
 3-10-004-04      Oil and Gas Production   Process Heaters: Natural               12,187
                  gas
 3-10-004-14      Oil and Gas Production   Steam generators: Natural               2.852
                  gas
 Total                                                                        15,039
       Additional combustion of natural gas at natural gas pipelines and plants is reported
through boilers, engines, as in-process fuel, and in flares.

2.1.5.5 NAPAP Emissions

       Industrial natural gas consumption is listed for external combustion boilers, process
heaters, internal combustion boilers, and in-process fuel.  The emissions  at the 8-digit SCC
level are as follows.
Boilers
SCC
1-02-006-01
1-02-006-02
1-02-006-03
1-02-006-04
1-02-014-01
Description
External combustion boilers -
gas, Over 100 MMBtu/Hr
External combustion boilers -
gas, 10-100 MMBtu/Hr
External combustion boilers -
10 MMBtu/Hr
External combustion boilers -
gas, Cogeneration
External combustion boilers -
Industrial: Natural
Industrial: Natural
Industrial: Less than
Industrial: Natural
Industrial: CO Boiler,
Tons SO^ Emitted
23,655
10,733
267
315
9
                  Natural gas
                                          35

-------
 SCC             Description                                        Tnns SO, Emitted
 1-05-001-06      External combustion boilers - Space heaters -                        14
                  Industrial, Natural gas
 2-02-002-01      Internal combustion engines- Industrial: Natural gas,              1,523
                  Turbine
 2-02-002-02      Internal combustion engines- Industrial: Natural gas,              4,198
                  Reciprocating
 2-02-002-03      Internal combustion engines-Industrial: Natural gas,                211
                  Turbine: Cogeneration
 2-02-002-04      Internal combustion engines- Industrial: Natural gas,                  4
                  Reciprocating: Cogeneration
 3-01-900-03      Chemical manufacturing - Process heaters: Natural                1.908
                  gas
 3-01-900-13      Chemical manufacturing   Incinerators: Natural gas               2.016
 3-03-900-03      Primary metal production   Process heaters: Natural               1,370
                  gas
 3-04-900-03      Secondary metal production - Process heaters:                       25
                  Natural gas
 3-07-900-03      Pulp & Paper and Wood products - Process heaters:                  39
                  Natural gas
 3-07-900-13      Pulp & Paper and Wood products - Incinerators:                     13
                  Natural gas
 3-90-006-02      In process fuel use: Natural gas, Cement kiln/dryer                   0
 3-90-006-03      In process fuel use: Natural gas, Lime kiln                           0
 3-90-006-05      In process fuel use: Natural gas, Metal melting                       0
 3-90-006-89      In process fuel use: Natural gas, general                         12,700
 3-90-006-99      In process fuel use: Natural gas, general                            993
 3-99-900-03      Miscellaneous Manufacturing Industries - Process                     3
                  heaters: Natural gas
 3-99-900-13      Miscellaneous Manufacturing Industries -                           37
                  Incinerators:  Natural gas                                        	
 Tota]                                                                         60.033
       In addition, NAPAP reports area source emissions of  1,000 tons of SO-, from natural
gas combustion. This brings the  total combustion from natural gas to 61,033 tons of SO,  In
order to compare the NAPAP and TRENDS estimates, emissions from cement plants,
petroleum refineries,  the iron and steel industry, and the glass manufacture industry must be
excluded from the NAPAP estimate.  For the above listed SCCs, 5 tons are from cement
manufacturing (SIC 3241), 9,377 tons are from petroleum refineries (SIC 2911), 10,044 tons

                                          36

-------
are from the iron and steel industry (SIC 3312, & 3325), and 1,582 tons are from glass
manufacture (SIC 3211, 3221, & 3229).  In addition, NAPAP reports 6,797 tons of S02 from
combustion during crude petroleum and natural gas production (SIC 1311) and 404 tons of
SO2 from natural gas liquids production (SIC 1321)  for the above listed SCCs. The adjusted
NAPAP emissions for natural gas combustion in  boilers are 32,824 tons of SO2.

Natural gas pipelines and  plants

       Emissions from combustion  of natural gas at natural gas pipelines and plants are
reported in NAPAP under as follow.
  SCC             Description                                        Tons SO- Emitted
  3-10-004-04       Oil and Gas Production - Process heaters: Natural                 442
                   gas
  3-10-004-14       Oil and Gas Production - Steam generators: Natural                  18
                   gas
  Total                                                                          460

       In  addition, emissions of 6,797 tons of S02 from crude petroleum and natural gas
(SIC 1311) and 404 tons of SO2 from natural gas liquids production (SIC 1321) are included
in the category Natural Gas Pipelines and Plants. This brings the total for this category  to
7,661 tons of SO2.

2.1.5.6 Conclusions

       The activity rates for natural gas combustion that are reported in the NAPAP inventory
are not consistent with the emissions. The emissions appear to  be overestimated by less than
an order of magnitude and, given the relatively small amount of SO2 emitted,  were not closely
reviewed in the NAPAP inventory.  The total natural gas reportedly consumed is about 7,000
x 109 ft3 in NAPAP versus 5,901 x  109 ft3 in TRENDS (unadjusted).  The TRENDS value is
fairly consistent with the 4,512 x 109 ft3 reported through the Manufacturing Energy
Consumption Survey: Consumption of Energy, 1985*
                                          37

-------
      The TRENDS conversion factors for both the iron and steel industry and the glass
manufacturing industry are to be periodically updated.  The value for the steel industry that is
currently used is 4.25 x 106 ft3 of natural gas/103 tons of raw steel.  The value for the iron
and steel industry can  be updated with information provided in Table 3 "Total Inputs of
Energy for Heat, Power, and Electricity Generation by Census Region, Industry Group, and
Selected  Industries,  1985" of Manufacturing Energy Consumption Survey:  Consumption of
Energy, 79SS.16  Natural gas consumed by blast furnaces and steel mills was 400 billion
cubic feet (TRENDS calculated  375 billion cubic feet for raw steel).  Based on a 1985 raw
steel production of 88,259,000, a revised  factor for iron and steel would be:

       400,000 / 88,259 = 4.53  x 106 ft3/!000 ton raw steel.

       The value for glass cannot be recalculated at this time because the reference cited
combines stone, clay,  and glass  products.

2.1.6 Miscellaneous Fuel

       TRENDS includes industrial SO2 emission estimates for four categories of fuel. The
estimate  is published for all four of the fuels combined.  The TRENDS estimate for
miscellaneous fuels  is  80,000 tons of SO2.  NAPAP reports emissions of  14,400 tons of SO:
for those same four  fuels.  The apparent difference in the emission estimates is 65,600 tons
(455 percent). Table 2-6 compares the NAPAP and TRENDS values for these four fuels.

2.1.6.1  TRENDS Activity

       In TRENDS, there are four subcategories for miscellaneous fuel SO, emissions: coke,
coke oven gas, kerosene, and LPG. Both the coke and coke oven gas categories exclude  fuel
burned in the iron and steel industry.  Each of these subcategories is discussed separately
below.
                                          38

-------
           TABLE 2-6.  COMPARISON OF MISCELLANEOUS FUELS
                        VALUES FOR 1985 TRENDS AND NAPAP
Category
Coke Combustion
Emissions (tons)
Fuel Consumed (103 tons)
Coke oven Gas Combustion
Emissions (tons)
Fuel Consumed (109 ft3)
Kerosene Combustion
Emissions (tons)
Fuel Consumed (106 gallons)
LPG Combustion
Emissions (tons)
Fuel Consumed (106 gallons)
Total Miscellaneous Fuels
Emissions (tons)
TRENDS
Published

36,000
1,343

43,000
79

2,000
463

0
1,979
80,000
TRENDS
Calculated

26,645
1,621.2

1,075
85

1,434
463

63
5,756
29,200
NAPAP
Published

11,000
1,656

2,700
4

421
0.656a

52
286b
14,400
'Excludes activity assigned to area sources as part of distillate oil.
"Excludes activity assigned to area sources as natural gas equivalents.
Coke
      Industrial coke consumption outside the iron and steel industry is an adjusted sum of

coke from coal and petroleum coke.  Coke from coal is obtained from Table A5 "Coke and

Breeze Production at Coke Plants" of the Quarterly Coal Report.11  Total breeze production at

coke plants in 1985 was 2,155,000 short tons.  TRENDS assumes that 24 percent is sold for

use as boiler fuel.  Therefore, industrial breeze consumption is:


      2,155,000  * 0.24 = 517,200 short tons.
                                       39

-------
       Coke sales to "other industrial plants" are obtained from Table A6 "Coke and Breeze
Distributed from Coke Plants" in the Quarterly Coal Report.11 The 1985 value was 873,000
short tons.  Therefore, total industrial coke, produced from coal, consumed outside  the iron
and steel industry in  1985 was:

       517,200 + 873,000 = 1,390,200 short tons.

       TRENDS provides two sources for petroleum coke. The first reference is Table 12
"Receipts of Petroleum Coke at  Electric Utilities  for Steam Plants of 50-Megawatt Installed
Nameplate Capacity or Larger, 1985" of Cost and Quality of Fuels for Electric Utility Plants
19859 which lists 279,900 short  tons of petroleum coke received at electric utilities.  The
second reference is a footnote to Table 7 "Consumption of Coal, Petroleum, and Gas by
Electric Utilities, 1984-1985" of the Electric Power Annual 1985" which states that
petroleum coke consumption in  1985 was 231,000 short tons by  the electric utility industry.

       The total coke consumed is calculated by  adding the total amount of coke produced
from coal, and the amount of petroleum coke consumed by power plants. For 1985, the total
is:
       1,390,200 + 231,000 = 1,621,200 short tons.

       This value differs from the value in the TRENDS activity spreadsheet  which was
1,343,000 short tons.  It is unclear why the petroleum coke consumed at electric utilities
should be included in the industrial fuel combustion category.  Assuming this  is an error in
the TRENDS procedure document would bring the values closer together (1,390,000 short
tons  here versus 1,343,000 short tons in activity spreadsheet).

Coke oven gas

       The TRENDS procedure is to obtain coke oven gas production from Quarterly  Coal
Report"  however, coke oven gas production is not provided in that report.  Table 23 "Coal
Consumption  by End-Use Sector" of Quarterly Coal Report" provides coal consumption bv
                                          40

-------
coke plants in 1985 of 41,056,000 short tons.  Figure 15 "Production of Coke and Coal
Chemicals" of Coal Data: A Reference1* indicates that 11,000 ft3 of coke oven gas are
produced per ton of metallurgical (coking) coal.  This is consistent with the value used in the
Iron and Steel section of the TRENDS procedure.  This results in 1985 coke oven gas
production of:

       41,056,000  *  11,000 / 1,000,000 = 451,616 x 106 ft3.

       The TRENDS method assumes that 18.8 percent of the coke oven gas is consumed
outside of the iron  and steel industry.  Therefore the coke oven gas consumption is:

       451,616 * 0.188 = 84,903 x 106 ft3.

       The value in the TRENDS activity spreadsheet is 79,300.x 106 ft3.

Kerosene

       Kerosene consumption is obtained from Table 15 "Adjusted Sales of Kerosene by End
Use in the United States 1985-1989" of Fuel Oil and Kerosene Sales J989.3  The 1985 value
is 254,491,000 gallons for industrial use and 208,139,000  for all  other.  Therefore the
TRENDS activity value for industrial kerosene is:

       254,491,000 + 208,139,000 = 462,630,000 gallons.

LPG

       Most (88 percent) of the LPG is used as a feed stock. The TRENDS procedure
attempts to account for this using an ad hoc procedure to  determine consumption for use as  a
fuel. LPG consumption  by industrial sources is calculated by multiplying the 1985 production
with the ratio of 1982 sales of LPG to the 1982 product supplied (the last available data).
                                          41

-------
       In  1982 the total sales were 5,397,000,000 gallons. The 1982 products supplied were
 1,499,000 barrels/day.  Therefore the ratio is:

       5,397,000,000 / 1,499,000 = 3,600 gallons/bbl/day.

       The 1985 product supplied is obtained from Table S7 "Liquified Petroleum Gases
 Supply and Disposition" of Petroleum Supply Annual 1985 5  In 1985 the products supplied
 were 1,599,000 barrels/day, therefore, the 1985  industrial LPG  activity figure is:

        1,599,000 * 3,600 = 5,756,000,000 gallons.

       The value listed in the TRENDS activity spreadsheet is  1,979 x 106 gallons. Table 3
 "Total Inputs of Energy for Heat, Power, and Electricity Generation  by Census Region,
 Industry Group, and Selected Industries,  1985" of the Manufacturing Energy Consumption
 Survey: Consumption of Energy, 1985 lists the industrial LPG  consumption as  1,116 x 106
 gallons.

 2.1.6.2 TRENDS Emission Factors
Coke
       The TRENDS method recommends the development of a weighted emission factor for
coal coke and petroleum coke. The emission factor listed for petroleum coke is 38.8 and
TRENDS multiplies this value by 3.25 percent sulfur content for petroleum coke.  There is no
emission factor in the AIRS Facility Subsystem SCC and Emission Factor Listing for Criteria
Pollutants1 or in Compilation of Air Pollutant Emission Factors.  Volume I: Stationary Point
and Area Sources.  Fourth Edition. AP-4219 for combustion of petroleum coke.

       The TRENDS procedure document lists  an emission factor of 30.3 Ib/ton burned for
coal coke.  The emission factor listed in the AJRS Facility Subsystem SCC and  Emission
                                          42

-------
Factor Listing for Criteria Pollutants' for industrial coke combustion is 39.0S Ib/ton burned
(SCC 1-02-008-02). This implies that TRENDS is using 0.77 percent sulfur for coal coke.

       The TRENDS procedure requires that the emission factors be weighted based on AIRS
data and provides a table to assist in the weighing. Table 2-7 illustrates the values used to
produce the weighted emission factor of 43.9 Ibs/ton coke burned.  The TRENDS spreadsheet
lists an emission factor of 53  Ib/ton for  both the 1990 and 1991 study years.
           TABLE 2-7.  WEIGHTED AVERAGE EMISSION FACTOR
                        FOR COKE COMBUSTION
Coke Type
Petroleum Coke
Coal Coke'*
Weighted Average
(lb/103 ton burned)
Trends Activity SOX Emission Factor
(103 tons) " (Ib/ton)
231 126*
1,390.2 30.3
43.9
* Assumes a constant sulfur content value of 3.25 percent for petroleum coke.
** Total of breeze production plus coke industrial boilers.
Coke oven gas

      The emission factor for coke oven gas is 680.0S Ib SO^IO6 ft3.  This is consistent with
the AIRS Facility Subsystem SCC and Emission Factor Listing for Criteria Pollutants7
document for SCC 1-02-007-007. The TRENDS procedure document assumes a sulfur
content of 1.605 percent. Therefore the emission factor is:
      680.0 * 1.605 = 1,091 Ib SO2/106 ft3.
                                         43

-------
Kerosene

       TRENDS lists the kerosene emission factor for SCC 1-02-005-01 as 143.6S Ibs
SO2/103 gallons burned.  TRENDS also lists an average sulfur content of 0.075 percent for
kerosene.  This results in an overall emission factor for kerosene combustion of:

       143.6 * 0.075 = 10.77 lb/103 gallons burned.

       The SCC 1-02-005-01 is for grades  1 and 2 distillate oil. An appropriate SCC for
kerosene would be 2-02-009-01.  All of the emission  factors listed in AIRS Facility Subsystem
SCC and Emission Factor Listing for Criteria Pollutants1 for SO, from kerosene combustion
are 6.2 lb/103 gallons burned.
 LPG
       TRENDS lists the LPG emission factor for SCC 1-02-010-02 as 86.5S Ibs SCyiO3
gallons burned. This matches the emission factor in the AIRS Facility Subsystem SCC and
Emission Factor Listing for Criteria Pollutants1 for both butane and propane. TRENDS also
lists an average sulfur content of 0.0013 percent for LPG and results in the following overall
emission factor:

       86.5 * 0.0013 = 0.11  lb/103  gallons burned.
2.1.6.3  TRENDS Emissions

       The estimates presented below utilize the activity rates found in the TRENDS activity
spreadsheet and emission factors found in the TRENDS procedure document  Differences
between these numbers and TRENDS estimates are discussed under conclusions.
                                          44

-------
Coke

       Emissions from industrial coke combustion are calculated using the activity value and
emission factor found in the TRENDS spreadsheets.

       1,343,000 * 53.0  / 2000 = 35,589 tons of SO:

Coke oven gas

       Emissions are calculated  using the activity value found in the TRENDS activity
spreadsheet
       79,300 * 1,091 / 2000 = 43,258 tons of SO:

Kerosene

       Emissions are calculated using the emission factor listed in the TRENDS procedure
document

       462,600 * 10.77 / 2000 = 2,491  tons of SO,

LPG

       Emissions are calculated using the activity value found in the TRENDS activity
spreadsheet.
       1,979,000 * 0.11 / 2000 = 109 tons of SO2

Total  miscellaneous fuels

       The total emissions are the sum  of coke, coke oven gas, kerosene and LPG:

       35,589 + 43,258 + 2,491  +  109  = 81,447 tons S0;

                                         45

-------
2.1.6.4 NAPAP Activity

       Although these data were not priority NAPAP items, the data are presented here for
comparison.
Coke
       The NAPAP activity for coke combustion include industrial external boilers and in-
process fuel use.  The NAPAP inventory reports 55 percent of the activity in SIC codes  1011,
3312 and 3321  (all iron and steel).  The remainder includes petroleum refining, electric
utilities, chemical manufacturing and mineral products.
  SCC         Description                                                Tons burned
  1-02-008-02   External combustion boilers - Industrial: Coke, all boiler sizes       392,652
  1-02-008-04   External combustion boilers  Industrial: Coke, cogeneration         834,218
  3-90-008-99   In process fuel use, Coke, general                               2.414.424
  Total                                                                      3,641,294

Coke oven gas

       The NAPAP activity for coke oven gas include external combustion boilers and  in-
process fuel use.  The inventory reports 99 percent of the activity in SIC codes 1011, 3312
and 3321, with the remainder in chemical manufacturing, carbon black and petroleum
refining.

  SCC             Description                                            106 ft3 Burned
  1-02-007-07      External  combustion boilers - Industrial: Process                111,686
                  gas, coke oven gas
  3-90-007-01      In process fuel use, Process gas. Coke oven or blast             423,091
                  furnace
  3-90-007-02      In process fuel use, Process gas. Coke oven gas                 193,971
  3-90-007-89      In process fuel use, Process gas, Coke oven gas                 65 119
  Total                                                                        793,867
                                          46

-------
Kerosene

       The NAPAP activity for kerosene include internal combustion boilers.  No adjustment
for area source activity could be made because kerosene is combined into the distillate oil
category.
 SCC
 2-02-009-01

 Total

LPG
Description
Internal combustion engines  Industrial:
Kerosene/Naphtha (Jet fuel), Turbine
10' Gallons Burned
              656
                                                              656
       The NAPAP activity for LPG include external combustion boilers, space heat, internal
combustion boilers, and in-process fuel use.  No adjustment for area source activity could be
made because LPG is combined as natural gas equivalents in the natural gas category.
 SCC

 1-02-010-01

 1-02-010-02

 1-05-001-10

 2-02-010-01

 2-02-010-02

 3-90-009-89
 3-90-010-89
 3-90-010-99
 Total
 Tons burned.
    Description

    External combustion boilers - Industrial: LPG -
    Butane
    External combustion boilers - Industrial: LPG -
    Propane
    External combustion boilers - Space heaters:
    Industrial - LPG
    Internal combustion engines - Industrial: LPG,
    propane: reciprocating
    Internal combustion engines - Industrial: LPG,
    butane: reciprocating
    In process fuel use, LPG, general
    In process fuel use, LPG, general
    In process fuel use, LPG, general
           103 Gallons
              Bumed
                 714

              100,060

                   47

                1,854
              91,944'
                9.702
               81.864
              286,188
                                         47

-------
2.1.6.5 NAPAP Emissions

Coke

       The NAPAP emission estimates for coke combustion include industrial external boilers
and in-process fuel use.

  SCC             Description                                       Tons SO. Emitted
  1-02-008-02       External combustion boilers   Industrial: Coke, all                2.246
                   boiler sizes
  1-02-008-04       External combustion boilers - Industrial: Coke,                   9,665
                   cogeneration
  3-90-008-99       In process  fuel use. Coke, general                               	89
  Total                                                                        12,000

       The NAPAP inventory  reports 747 tons SO2 in these three SCCs among SICs 1011,
 3312 and 3321 within the iron and steel industry. The remaining 11,252 tons are  categorized
 with  miscellaneous fuel.

 Coke oven gas

       The NAPAP emission estimates for coke  oven gas include external combustion boilers
 and in-process  fuel use.


  SCC             Description                                       Tons SO. Emitted
  1-02-007-07       External combustion boilers   Industrial: Process                 17,637
                   gas, coke oven gas
  3-90-007-01       In-process fuel use, Process gas,  Coke oven or  blast                  6
                   furnace
  3-90-007-02       In-process fuel use, Process gas,  Coke oven gas                  4,331
  3-90-007-89       In-process fuel use, Process gas,  Coke oven gas                  4.795
  Total                                                                         26,770
                                          48

-------
      The NAPAP inventory reports 24,081 tons SO2 in these SCCs among SICs 1011, 3312
and 3321 (iron and steel). The remaining 2,689 tons are categorized within miscellaneous
fuel.

Kerosene

      The NAPAP emission estimates for kerosene include internal combustion boilers.
 SCC             Description                                       Tons SO. Emitted
 2-02-009-01      Internal combustion engines - Industrial:                          421
                  Kerosene/Naphtha (Jet fuel), Turbine
 Total                                                                           421
LPG
      The NAPAP emission estimates for LPG include external combustion boilers, space
heat, internal combustion boilers, and in-process fuel use.

 SCC             Description                                       Tons SO, Emitted
 1-02-010-01       External combustion boilers - Industrial: LPG -                      0
                  Butane
 1-02-010-02       External combustion boilers - Industrial: LPG                       27
                  Propane
 1-05-001-10       External combustion boilers - Space heaters:                        18
                  Industrial - LPG
 2-02-010-01       Internal combustion engines - Industrial: LPG,                       0
                  propane: reciprocating
 2-02-010-02       Internal combustion engines - Industrial: LPG,                       0
                  butane: reciprocating
 3-90-009-89       In process fuel use, LPG, general                                  2
 3-90-010-89       In process fuel use, LPG, general                                  5
 3-90-010-99       In process fuel use, LPG, general                                 _0
 Total                                                                            52
                                         49

-------
2.1.6.6 Conclusions

       There are many numbers in this category that are not well justified in the TRENDS
method.  The categories are discussed separately.
Coke
       The coke emissions, excluding iron and steel, are 11,300 tons in NAPAP, and 36.000
tons in TRENDS.  The coke emissions are probably underestimated in  the NAPAP inventory.

       There are discrepancies in the TRENDS activity data.  First, it is unclear why the
petroleum coke that is delivered to electric utilities should be included  in this category. (The
NAPAP total for this petroleum coke is 3,111 tons.) Second,  the activity value in the
TRENDS spreadsheet could not be reproduced by following the TRENDS procedure.
TRENDS lists 1,343,000 tons in the  activity spreadsheet.  Following the TRENDS  procedure
resulted in a value of 1,621,000 tons. The Manufacturing Energy Consumption Survey:
Consumption of Energy, 1985* lists 1,952,000 tons of coke and breeze  for industries outside
of blast furnaces and steel mills.  Therefore, the activity value used to estimate the 1985 coke
combustion emissions is probably too low in the  TRENDS procedure.

       The emission factor which is  used for coke combustion may overestimate the
TRENDS  emission estimate.  TRENDS uses an emission factor of 53.0 Ib/ton burned.  The
AIRS Facility Subsystem SCC and Emission Factor Listing for Criteria Pollutants1  has an
emission  factor of 39.OS Ib/ton burned.  The NAPAP inventory lists an average coke sulfur
content of 0.7 percent which results in an overall emission factor of 27.3 Ib/ton burned.  If
the petroleum coke delivered to electric utilities is not included and the value from the
Manufacturing Energy Consumption  Survey: Consumption of Energy, 1985* is used and the
revised emission factor is utilized, the following TRENDS coke emission estimate would
result.

       1,952,000 * 27.3 / 2000 = 26,645 tons of SO:
                                          50

-------
Coke oven gas

       The coke oven gas emissions, excluding iron and steel, are 2,700 tons in NAPAP, and
43,000 tons in TRENDS. It appears as though TRENDS overestimates SO2 emissions for
combustion of coke oven gas outside the iron and steel industry.  TRENDS assumes 18.8
percent of the coke oven gas produced is burned in boilers outside the iron and steel industry.
The 18.8 percent is not documented. In addition, the TRENDS iron and steel section assumes
that 40 percent of coke oven  gas is  used in the iron and steel process equipment (see Roll and
Finish subsection of iron and steel). The remaining 40 percent of the coke oven gas being
consumed in  the TRENDS procedure is not accounted for. Table 12 "Production and
Disposal of Coke Oven Gas in the United States by Producing State:  1980" of Coke and Coal
Chemicals in 79S020 reports that in  1980 coke gas use was 39 percent by producers in
heating ovens, 58 percent other use by producers, 1.4 percent commercial sales, and 1.5
percent wasted. These statistics are consistent with the NAPAP distribution of coke oven gas
combustion.

       TRENDS lists a coke  oven gas average sulfur value of 1.605 percent.  The NAPAP
inventory lists an average sulfur content for coke oven gas of 0.5 percent.  Using a factor of
1.4 percent of coke oven gas  burned in industrial boilers outside the iron and  steel industry
and using the NAPAP average sulfur content, results in the following emissions.

       451,616 * 0.014 * 680 * 0.5 / 2000 = 1,075 tons of SO2

Kerosene

       The kerosene emissions are 421  tons in NAPAP and 2,491 tons in  TRENDS.  The
kerosene  emissions are probably overestimated in the TRENDS document.  The emission
factor used in the TRENDS procedure to estimate kerosene emissions is actually an emission
                                         51

-------
factor for distillate oil. The emission factor cited is 10.77 lb/103 gallon burned. Using a
kerosene emission factor of 6.2 lb/103 gallons burned results in the following emissions.

       462,630 * 6.2 / 2000 = 1,434 tons of SO2

LPG

       The LPG emissions are 52 tons in NAPAP and 109 tons in TRENDS.   The LPG
emissions are probably overestimated in the TRENDS document.  Following the TRENDS
procedure manual did not result in the same activity value for LPG combustion as is
published in the TRENDS activity spreadsheet.  The LPG activity value used in the 1985
TRENDS estimate is higher (1,979 million gallons versus 1,116 million gallons) than the
value reported through the Manufacturing Energy Consumption Survey:  Consumption of
Energy, J985.s  Using the value reported through the survey results in the following
emissions.

       1,116,000 * 86.5  * 0.0013 / 2000 = 63 tons of SO,

       The TRENDS procedure for determining LPG combustion activity is difficult to
understand.  A preferred approach may be to hold the value constant and update it every three
years with a new Manufacturing Energy  Consumption Survey.

       Total TRENDS emissions  for the miscellaneous fuel category would then be:

       26,645 +  1,075 + 1,434 +  63 = 29,217 tons of SO:

2.1.7 Wood

       The published TRENDS emission  estimate for wood combustion is 10,000 tons of
SO2.  The NAPAP estimate is 41,700 tons of SO2. The apparent difference in  the emission
estimates is 31,700 tons (76 percent).

                                         52

-------
2.1.7.1  TRENDS Activity

       The TRENDS procedure for determining the activity value for wood combustion is
poorly documented and difficult to follow.  The total industrial combustion of wood for both
the 1984 and the 1989 study years is available from Table 1 "U.S. Consumption of Wood
Energy by Sector, 1949-1990" of Estimates of U.S. Biofuels Consumption 1990.21 The value
for 1984 is 1,679 trillion Btu. The TRENDS procedure document provides a conversion
factor of 17.2 million Btu per oven-dried short ton. Therefore the 1984 wood consumption is:

       (1,679 x 1012) / (17.2 x 106) = 97.616 x 106 oven-dried short tons.

       The TRENDS procedure initially instructs the user to obtain the consumption figures,
in tons, for the previous year and to assume that 15 percent of the heating value is lost to
moisture on a typical basis. The TRENDS procedure further states that as of 1990, wood
consumption was published in therms of Btu's and an average Btu content per oven-dried
short ton is provided. TRENDS then states "No adjustment to the calculated tonnage is
necessary."  This last statement is interpreted as instruction not to apply the 85 percent factor
to account for lost moisture.

       TRENDS assumes that 75 percent of the industrial wood is consumed by the pulp and
paper industry and 25 percent is used in lumber and wood products. TRENDS requires that
the converted (from Btu to oven-dried ton) consumption figure be projected to the update year
following a procedure, outlined in the LPG section, for paper and wood separately.  Therefore
the portion of the industrial wood combustion from pulp is:

       0.75 * 97,616,300  = 73,212,000 oven-dried tons.

       The portion of the industrial wood combustion from the lumber industry is:

       0.25 * 97,616,300  = 24,404.000 oven-dried tons.
                                         53

-------
      The production of sulfite (kraft) and sulfite is found in Table 4 "Production and
Shipments of Woodpulp, by Type of Pulp: 1986 and 1985" of Current Industrial Reports,
Pulp, Paper, and Board.22  The 1985 production figure for sulfate was 42,563,831 short tons
and for sulfite was 1,620,084 short tons.  Therefore, the total 1985 pulp production value was
44,183,915 short tons. The 1984 value is obtained from the TRENDS activity spreadsheet
and was 42,758,800 short tons. Therefore the projection for the paper portion of the wood
combustion is as follows:

       '85 woodpaper  = '84 woodpaper * [1 + ('85 paper - '84 paper) / '84 paper]
                    = 73,212,000 * [1 + (44,183.9 - 42,758.8)7 42,758.8]
                    = 75,652,388 oven-dried tons

       The lumber production is found in Table 1  "Lumber Production: 1980 to 1991" of
Current Industrial Reports, Lumber Production and Mill Stocks.23 The 1985 total production
was  36,445 million board feet.  The 1984 total production, as documented in the TRENDS
activity spreadsheet, was 37,065 million board feet Therefore the projection for the lumber
portion of the wood combustion is as follows:

       '85 woodlumber       =  '84 woodlumbCT * [1  + ('85 lumber  '84 lumber) / '84 lumber]
                          =  24,404,000 * [1 + (36,445   37,065)/ 37,065]
                          =  23,995,795 oven-dried tons

       Total industrial wood combustion is the sum of these two figures:

       75,652,388 + 23,995,795 = 99,648,183 oven-dried  tons.

      The value in the TRENDS activity spreadsheet is 113,380,000 oven-dried tons.
                                          54

-------
2.1.7.2  TRENDS Emission Factors

       TRENDS lists the emission factor for wood combustion as 0.15 Ib/ton burned.  This is
consistent with the AIRS Facility Subsystem Source Classification Codes and Emission Factor
Listing for Criteria Air Pollutants1 document which lists the emission factors for industrial
external combustion of wood as 0.15 Ib/ton burned for all types of boilers.

2.1.7.3  TRENDS Emissions

       Using the activity value in the TRENDS activity spreadsheet results in emissions of:

       113,380,000 * 0.15 / 2000 = 8,504 tons of SO2.

2.1.7.4  NAPAP Activity

       Collection  and quality assurance of the activity data were not priority elements for all
sources in the inventory.  The data are not presented here.

2.1.7.5  NAPAP Emissions

       The emissions from wood combustion reported in NAPAP include wood and bark
waste burned in external boilers, wood burned as in-process fuel, and area source wood
combustion.  Because the collection and quality assurance of the activity data were not high
priority items for all sources, these data are not presented  here.
 SCC             Description                                        Tons SO, Emitted
 1-02-009-01      External combustion boilers - Industrial: Wood/Bark              6,863
                  waste, Bark-fired boiler (>50,000 Ib steam)
 1-02-009-02      External combustion boilers - Industrial: Wood/Bark              5,664
                  waste, Wood/Bark-fired boiler (>50,000 Ib steam)
 1-02-009-03      External combustion boilers - Industrial: Wood/Bark               1,102
                  waste, Wood-fired boiler (>50,000 Ib  steam)
 1-02-009-04      External combustion boilers - Industrial: Wood/Bark                  4
                  waste, Bark-fired boiler (<50,000 Ib steam)

                                         55

-------
 SCC             Description                                       Tons SO. Emitted
 1-02-009-05      External combustion boilers - Industrial: Wood/Bark                 45
                  waste, Wood/B ark-fired boiler (<50,000 Ib steam)
 1-02-009-06      External combustion boilers - Industrial: Wood/Bark                 16
                  waste, Wood-fired boiler (<50,000 Ib steam)
 1-02-009-07      External combustion boilers   Industrial: Wood/Bark               1,446
                  waste, Wood cogeneration
 3-90-008-89      In process fuel. Wood, general                                  9,567
 3-90-009-99      In process fuel. Wood, general                                  	1_
 Total                                                                       24,708
       In addition, NAPAP reports 17,000 tons as area source emissions. Therefore, the total
NAPAP wood combustion emission estimate is 41,700 tons SO2.
2.1.7.6  Conclusions

       The TRENDS emission estimate for wood combustion is 8,504 tons of SO2. The
NAPAP estimate is 41,700 tons of SO2.  Nearly half of the point source emissions in the
NAPAP inventory are from a general in-process wood combustion category. The emission
factor for this SCC (3-90-008-89) is 38.0S Ibs/ton burned.  NAPAP also reports an average
sulfur content of 1.5 percent for the SCC. This emission factor is  substantially higher than
the emission factor used in TRENDS and the rest of the NAPAP categories (0.15 Ib/ton
burned).  As a result, this category is responsible for a disproportionate share of the wood
combustion point source emissions reported in NAPAP.  Due to the high emissions for this
one category of wood combustion, the NAPAP inventory probably overestimates the wood
combustion emissions.

       Following the TRENDS procedure did not recreate the activity value that was used in
the calculation of the 1985 emission estimate for wood combustion.  Even when rounding the
emission estimate to the nearest 10,000 tons,  the difference is not insubstantial.  Table 3
"Industrial Woodfuel  Consumption by Sector, 1990" of Estimates of U.S. Biofuels
Consumption 199CP states that paper and allied product consume 79 percent of the industrial
wood fuel, lumber and wood products consume 18 percent and other industries consume the
                                         56

-------
remaining 3 percent.  Using these breakdowns by industry results in a 1985 wood
consumption value of 99,892,644 oven-dried tons.  This results in emissions of:

       99,892,644 * 0.15 / 2000 = 7,492 tons of SO2.

2.1.8 Other NAPAP Combustion Categories

       Additional emissions are reported in the NAPAP inventory as combustion emissions
for sources not discussed in this section (because they have no corresponding TRENDS
estimate). These emissions include liquid waste, waste oil, solid waste, bagasse, and process
gas, and  amount to an additional 74,477 tons of SO,.

2.2  NON-FERROUS SMELTING

       The TRENDS nonferrous smelting category includes primary copper, primary lead,
primary zinc, primary aluminum and  secondary lead. Additional categories are included in
TRENDS for other criteria pollutants (especially particulate) and other non-ferrous smelting
categories are included in the NAPAP inventory. Table 2-8 presents the comparison of the
NAPAP and TRENDS non-ferrous smelting SO2 emission estimates.

       The non-ferrous smelting source categories contribute significantly  to the industrial
SO2 emissions.  The majority of the smelters recover sulfur that is emitted from the ores
being processed.  Due to the increasing environmental pressures exerted on smelters and
global competition, smelters in this country are continuing to close, dramatically affecting the
trends in  emissions from these categories.  This  discussion begins with the reporting of sulfur
recovered as H:SO4 from non-ferrous smelters.

       Byproduct  1985 sulfuric acid production  was reported in Table 7 "Byproduct Sulfuric
Acid Produced in  the United States" of Minerals Yearbook 1989 "Sulfur" and in Table 22
"Byproduct Sulfuric Acid (100% Basis) Produced in the United States" of Minerals Yearbook
1989  "Copper". Values  are converted to SO, based on molecular weights of sulfur (32), SO:

                                         57

-------
           TABLE 2-8. COMPARISON OF NON-FERROUS SMELTING
                       VALUES FOR 1985 TRENDS AND NAPAP
Category (SIC)
Primary Copper (3331)
Emissions (tons)
Activity
Primary Zinc (3339)
Emissions (tons)
103 Tons of concentrate
Primary Lead (3339)
Emissions (tons)
Lead processing (103 tons)
Primary Aluminum (3334)
Emissions (tons)
Aluminum produced (103 tons)
Secondary Lead (3341, 3364)
Emissions (tons)
Reverberatory furnace (103 tons)
Blast furnace (103 tons)
Total Emissions (tons)
TRENDS
Published

650,000


67,400'
686.2

34,500'
759.3

70,000
3,850

30,000
394.0
391.3
990,000"
TRENDS
Calculated




93,864
734.3

98,775
975.4

71,039
3,855.6

27,147
393.6
430.4
940,825
NAPAP
Published

655,300


7.600


98,800


58,400


20,700
216.6
332.8
840,800
"Published together as one value of 240,000 tons of SO2.  The discrepancy between the
 total and individual sum (67,400 + 34,500) could  not be resolved.
""Includes 240,000 tons of SO2 for primary lead and zinc.
(64), and sulfuric acid (98). Table 2-9 lists the 1985 sulfur recovery breakdown/


      The TRENDS activity spreadsheet lists sulfur recovered as H:SO4 as 327,900 tons of

SO2 for primary lead and zinc versus 501,000 tons published in Mineral Yearbook, 79S94


2.2.1  Primary Copper
      NAPAP and TRENDS report essentially identical SO2 emissions for this category

(655,000 tons vs. 650,000 tons).  Because these smelters are large and few in number, it is

understood that TRENDS actually tracks  individual primary copper smelter emissions, as does
NAPAP   No further analysis of the category is warranted.

                                       58

-------
              TABLE 2-9.  RECOVERY OF SULFUR AS H2S04

Copper
Lead &
Molybdenum
Zinc
Total
As Sulfur"
(metric tons)
729,000
87,000

141.000
957,000
As SO2a
(short tons)
1,604,000
191,400

310.200
2,105,600 .
As H2SO4b
(metric tons)
2,230,257
267,159

430.946
2,928,362
As SO2b
(short tons)
1,602,240
191,930

309.600
2.103,770
 "Reported in Table 7 Minerals Yearbook 1989 "Sulfur."
 "Reported in Table 22 Minerals Yearbook 1989 "Copper."
 2.2.2 Combined Primary Lead and Primary Zinc

       The 1985 TRENDS value is 240,000 tons SO; for primary lead and zinc, which are
 reported together as one category. As described below, this total could not be reproduced.
 The 1985 NAPAP value is 98,775 for primary lead and 7,642 for primary zinc.  The
 discrepancy between the two inventories is 133,600 tons (126 percent).

 2.2.3 Primary Zinc

 2.2.3.1 TRENDS Activity

      National zinc production is obtained from Table 1 "Salient Zinc Statistics" of Minerals
 Yearbook 1989 "Zinc."* The 1985 total slab U.S. zinc production was 333,772 metric tons
 (367,149 short tons).  This value is multiplied by 2 because there are 2 units of concentrate
per ton slab zinc. This provides a value of 734,298  tons of concentrate.

      The value listed in the TRENDS activity spreadsheet is 686,200 tons for 1985.
                                         59

-------
2.23.2 TRENDS Emission Factor
       The emission factor for zinc roasting cited in the TRENDS method is as follows.
 SCC         Description                             Emission  Units
                                                        Factor
 3-03-030-02  Primary Metal  Zinc Production:             1,100  Ib/ton concentrated
              Multiple hearth roaster                             ore processed
2.2.3.3  TRENDS Emissions

       TRENDS accounts for the recovery of sulfur at primary zinc smelters. The 1985
recovery as shown in Table 2-9 is 310,000 tons of S(X Therefore,  1985 emissions are
calculated as follows (using the activity value found in the TRENDS spreadsheet):

       1,100 * 686,200 / 2000 - 310,000 = 67,410 tons of SO2.

       In contrast, the 1985 emissions using  the zinc activity published in Minerals Yearbook4
are:

       1,100 * 734,298 / 2000 - 310,000 = 93,864 tons of SO2.

2.2.3.4  NAPAP Activity

       The NAPAP activity is listed below.  The units for most processing with associated
SO2 emissions are tons of concentrated ore processed.  Recall that these activity data were not
high priority for all sources  in the inventory.

 SCC             Description                                                 Tons of
                                                                    Concentrated Ore
                                                                            Processed
 3-03-030-02      Primary Metal  Zinc Production: Multiple hearth               35 Q82
                  roaster
 3-03-030-03      Primary Metal  Zinc Production: Sinter strand                 472 550

                                          60

-------
 SCC              Description                                                 Tons of
                                                                     Concentrated Ore
                                                                            Processed
 3-03-030-05       Primary Metal - Zinc Production: Vertical retort/             1,954,037
                   electrothermal furnace
 3-03-030-06       Primary Metal - Zinc Production: Electrolytic                3,407.830
                   processor
 3-03-030-07       Primary Metal   Zinc Production: Rash roaster                       0
 3-03-030-11       Primary Metal   Zinc Production: Zinc casting                  79,410'
 3-03-030-14       Primary Metal - Zinc Production: Crushing/                      2,376
                   screening
 3-03-030-15       Primary Metal   Zinc Production: Zinc melting                 26,208"
 3-03-030-99       Primary Metal   Zinc Production: Not classified              5,682,107

a Tons zinc produced
b Tons processed
       As shown above, one record lists 79,410 tons of zinc cast (3-03-030-11), however, it
is doubtful that this represents all of the zinc produced in 1985. The largest value for
concentrated ore processed is 3,407,830 tons.  Neither the total zinc cast nor the concentrated
ore processed in the electrolytic processor are consistent with the TRENDS activity value.

2.2.3.5 NAPAP Emissions

 SCC             Description                                        Tons SO. Emitted
 3-03-030-02      Primary Metal -  Zinc Production: Multiple hearth                   849
                  roaster
 3-03-030-03      Primary Metal -  Zinc Production: Sinter strand                   5,175
 3-03-030-05      Primary Metal -  Zinc Production: Vertical retort/                 1,104
                  electrothermal furnace
 3-03-030-06      Primary Metal -  Zinc Production: Electrolytic                        0
                  processor
 3-03-030-07      Primary Metal -  Zinc Production: Flash roaster                     258
 3-03-030-11      Primary Metal -  Zinc Production: Zinc casting                       0
 3-03-030-14      Primary Metal -  Zinc Production: Crushing/                          0
                  screening
 3-03-030-15      Primary Metal -  Zinc Production: Zinc melting                       0
 3-03-030-99      Primary Metal -  Zinc Production: Not classified                     256

                                          61

-------
 SCC             Description                                       Tons SO. Emitted
 Total                                                                          7.642
       Only 849 tons are reported in NAPAP for zinc roasting in the multiple hearth roaster
and this process is not the largest source of SO2 emissions in the NAPAP inventory (although
it does have the largest emission factor).

2.2.3.6  Conclusion

       The NAPAP and TRENDS estimates for emissions from primary zinc production are
very different. NAPAP reported emissions of 7,642 tons of SO: and the TRENDS method
resulted in an emission estimate of 93,864 tons of S0:.

       The TRENDS published estimate for primary zinc production is combined  with the
primary lead estimate and the published total of 240,000 tons of SO2 from both  industries
could not be recreated.  In the TRENDS method, sulfur recovered as H2SO4  is used in the
development of both the zinc and lead emission estimate.  The value in the TRENDS
spreadsheet for recovered sulfur for  this section also could not be reproduced. The TRENDS
activity spreadsheet stated that 327,900 tons of SO2 were produced at primary lead and
primary zinc facilities in  1985.  The references for these data are sections in the Minerals
Yearbook4 and the values published  are 501,500 tons of SO:.  The difference in  the recovered
sulfur could account for the  inability to recreate the published TRENDS value.

       The TRENDS method needs  to be updated. TRENDS assumes all roasting is done in
a multiple hearth roaster; two additional SCCs for roasting exist, flash roaster (3-03-030-07)
and fluid bed roaster (3-03-030-08).   Both have a smaller emission factor (404.4 and 223.5
Ibs/ton of concentrated ore processed, respectively) than  the multiple hearth roaster (1,100
Ibs/ton of concentrated ore processed).

       The NAPAP inventory did not report the majority of emissions through the multiple
hearth roaster. The NAPAP inventory may have  overestimated SO: emissions from some of
                                         62

-------
 the other processes in zinc production (specifically the sinter strand and the vertical
 retort/electrothermal furnace SCC 3-03-030-03/05). The discrepancy in NAPAP where the
 majority of emissions were not reported through the roasting process needs to be investigated.
 but could be an artifact of the NEDS software as well as misreporting of emissions.

 2.2.4  Primary Lead

 2.2.4.1 TRENDS Activity

       The TRENDS procedure has a complicated method to determine "Lead Processing" as
 opposed to primary lead production.  Primary lead production can be obtained from Table 1
 "Salient Lead Statistics" of Minerals Yearbook 1989 "Lead."* The 1985 production value is
 the sum from domestic  ore and base bullion and from foreign ores and base bullion and was:

       422,650 + 71,353 = 494,003 metric tons = 543,403 short tons.

       According to the TRENDS procedure manual, lead processing is calculated in four
 steps.  First, total copper and zinc SO2 emissions are calculated.  Second, byproduct sulfuric
 acid recovered from copper,  zinc, and lead smelters is estimated.  Third, total SO; lead
 emissions are  calculated using the following equation:

       Total SO:   = SO2 (lead+zinc) -  SO2(zinc) + AFS Lead Emissions.

 Fourth, the lead processing value is backcalculated using the following equation.

 Lead Processing = Total SO. Lead Emissions * 2000
                        595

       The derivation and reasoning behind this procedure are not documented.  In addition.
this procedure includes steps to calculate values that are not used  in the remaining steps.
                                          63

-------
Following the TRENDS procedure for determining lead production leads to the following
value for 1985.

Step 1, copper and zinc SO2 emissions  (based on the previous sections) are:

       650,000 + 67,410 = 717,410 tons of SO2.

Step 2, byproduct sulfuric acid for copper, lead  and zinc are listed in Table 2-9 as 2,100,000
tons of SO2.  Byproduct sulfuric acid for lead and zinc are 501,600 tons of SO:.  (It is unclear
why the TRENDS procedure references the sulfur recovered from copper smelting).

Step 3,  NAPAP lists the  1985 SO2 emissions (SCC 3-03-010-XX) as 98,775 short tons.

Total SO2    = SO2 (lead+zinc) - SO2(zinc) + AFS Lead Emissions

             = 501,600 - 310,200 + 98,775 = 290,175 tons of SO2

Step 4,

Lead Processing =   290.175 * 2000 = 975,378 short tons
                        595

       The TRENDS spreadsheet lists a lead processing value for 1985 as 759,300 tons.  There is
probably an error in the TRENDS procedure manual.  It is unclear why the lead sulfur recoverv
would be added to the zinc sulfur recovery only to subtract the zinc value.  The copper sulfur
recovery value is referenced, but does not  appear to enter into the equation.

2.2.4.2 TRENDS Emission Factor

      The emission factors cited in the TRENDS method are as follows.
                                             64

-------
 SCC          Description                             Emission  Units
                                                         Factor
 3-03-010-01   Primary Metals - Lead production:             550  Ib/ton lead produced
               Sintering, single stream
 3-03-010-02   Primary Metals - Lead Production:              45  Ib/ton lead produced
               Blast furnace operation
 Total                                                     595  Ib/ton lead produced
       These emission factors  are twice those listed in the AIRS Facility Subsystem Source
Classification Codes and Emission Factor Listing for Criteria Air Pollutants1 document but are
consistent with  AP-42.  The units are different: AP-42 units are tons of lead produced, the SCC
emission factor units are tons of concentrated ore processed.  In addition, the SCC emission factor
document lists 3-03-010-06 as  a major source (sintering dual stream versus single stream above) but
this appears to be a new SCC as there are no values in the NAPAP inventory.

2.2.4.3  TRENDS Emissions

       Using the TRENDS emission factor and activity rate published in the TRENDS spreadsheet
and subtracting the recovered sulfur results in the following emissions.

       759,300 * 595 / 2000 - 191,400 = 34,500 tons of SO2.

       Using the activity data derived  following the TRENDS procedure manual and subtracting the
recovered sulfur results in emissions of:

       975,382  * 595 / 2000 - 191,400 = 98,775 tons of SO:.

2.2.4.4 NAPAP Activity

      The following activity data were published in the NAPAP inventory.
                                              65

-------
 SCC             Description                                                      Tons of
                                                                         Concentrated Ore
                                                                                Processed
 3-03-010-01      Primary metal - Lead production: Sintering single stream           1,062,662
 3-03-010-02      Primary metal  Lead production: Blast furnace operation           1,006,182
 3-03-010-09      Primary metal - Lead production: Lead dressing                     275.599
 3-03-010-23      Primary metal  Lead production: Lead casting                      131.601
 2.2.4.5  NAPAP Emissions

       There are 21 SCCs or processes listed for lead production. Of these 21 SCCs only four have
 associated SO2 emissions and they are listed below.
  SCC            Description                                        Tons  SO. Emitted
  3-03-010-01      Primary metal - Lead production: Sintering single               78,496
                  stream
  3-03-010-02      Primary metal  Lead production: Blast furnace                 20,109
                  operation
  3-03-010-09      Primary metal - Lead production: Lead dressing                      1
  3-03-010-23      Primary metal - Lead production: Lead casting                     169
  Total                                                                       98,775

2.2.4.6  Conclusion

       The NAPAP and TRENDS estimates for emissions from primary lead production are also
very different. NAPAP reported emissions of 98,775 tons of SO2 and the TRENDS method resulted
in an emission estimate of 34,500 tons of S02.  (The TRENDS published estimate for primary lead
production is combined with the primary zinc estimate and the published total of 240,000 tons of
SO2  from both industries could  not be recreated.)

       In the TRENDS method, sulfur recovered as H2S
-------
Minerals Yearbook4 and the values published are 501,500 tons of S02.  The difference in the
recovered sulfur could account for the inability to recreate the published TRENDS value for primary
lead and primary zinc emissions.

       The TRENDS method  is outdated relative to the data that are currently provided in the
Minerals Yearbook, 1989*  It  is unclear why such a complicated procedure is introduced to
determine lead processing activity.  The TRENDS method includes four steps to determine lead
processing, however, many  of the steps do not seem logical. After following the four steps, the
result was a lead processing value of 975,378 short tons, which did not match the value of 759.300
tons in the TRENDS activity spreadsheet.  After analyzing the four steps that currently comprise the
TRENDS emission estimation procedure for primary lead, it appears that the final number is a simple
sum of the emissions reported through NEDS (now AFS) and the sulfur recovered as sulfuric acid.
The recovered sulfur is then subtracted from the emission estimate. Therefore, there should be
complete agreement between NAPAP and TRENDS for this category.  If the TRENDS method is
intended to be different from the simple sum, there are errors in the TRENDS procedure manual that
need to be addressed.

       The NAPAP activity for this category is  fairly close to the activity published in the Minerals
Yearbook.* The Minerals Yearbook cites a  1985 production of 543,403 short tons. TRENDS
includes the assumption that there is a 2:1 ratio  of concentrated ore processed to lead produced.
NAPAP reports 1,006,182 tons of concentrated ore processed in the blast furnace which would
correspond to a lead production rate of 503,000  tons of lead.

2.2.5 Primary Aluminum

       The 1985 TRENDS  emission value is 70,000 tons of S02 for primary aluminum.  The 1985
NAPAP value is 58,400 tons of SO2 for primary aluminum.  The discrepancy between the two
inventories is 11,600 tons (20  percent).
                                             67

-------
2.2.5.1 TRENDS Activity

       In TRENDS, the primary aluminum production is obtained from Table 20 "Salient Aluminum
Statistics" of Minerals Yearbook 1989 "Bauxite, Alumina, and Aluminum."4  In 1985 the U.S primary
production was reported as 3,500,000 metric tons (3,850,000 short tons).  The TRENDS activity
spreadsheet reports 3,855,600 tons of aluminum produced.

2.2.5.2 TRENDS Emission Factor

       SO2 emissions are calculated using an emission factor of 33.5 Ibs/metric ton  (reported in the
TRENDS spreadsheet).  This factor converts to 36.85 Ibs/ton. The emission factor is documented as
an average emission factor based on NEDS data from Washington State (February 1980).

2.2.5.3 TRENDS Emission Estimate

       Emissions using TRENDS method and activity data confirm the reported TRENDS estimate:

       36.85 * 3,855,600 / 2,000 = 71,039 tons SO2.

2.2.5.4 NAPAP Activity

       NAPAP activity data were not a high priority for collection or quality assurance for all
sources. The activity data reported through the NAPAP inventory are as follows for comparison.
 SCC             Description                                         Tons of Molten
                                                                         Aluminum
                                                                           Produced
 3-03-001-01       Primary metal -  Aluminum Ore: Electro-Reduction:           2 996007
                  Prebaked reduction cell
 3-03-001-02      Primary metal -  Aluminum Ore: Electro-Reduction:             307 880
                  Horizontal stud soderberg cell (MSS)
 3-03-001-03      Primary metal   Aluminum Ore: Electro-Reduction-             575 497
                  Vertical stud soderberg cell (VSS)
 3-03-001-04      Primary metal   Aluminum Ore: Electro-Reduction-
                  Materials handling                             '

                                            68

-------
 SCC             Description
 3-03-001-05      Primary metal - Aluminum Ore: Electro-Reduction:
                  Anode baking furnace

 3-03-001-06      Primary metal - Aluminum Ore: Electro-Reduction:
                  Degassing

 3-03-001-07      Primary metal - Aluminum Ore: Electro-Reduction:
                  Roof vents

 3-03-001-99      Primary metal  Aluminum Ore: Electro-Reduction:
                  Other not classified

 3-03-002-01      Primary metal - Aluminum Hydroxide Calcining:
                  Overall process

Tons of Alumina Produced
   Tons of Molten
       Aluminum
        Produced

        1,950,248


         290,785


         677,601


        7,861.822


       4,893,410a
2.2.5.5 NAPAP Emissions



      The emissions from primary aluminum production reported through the NAPAP inventory are

as follows.
 SCC             Description

 3-03-001-01       Primary metal - Aluminum Ore: Electro-Reduction:
                  Prebaked reduction cell

 3-03-001-02       Primary metal - Aluminum Ore: Electro-Reduction:
                  Horizontal stud  soderberg cell

 3-03-001-03       Primary metal - Aluminum Ore: Electro-Reduction:
                  Vertical stud soderberg cell

 3-03-001-04       Primary metal - Aluminum Ore: Electro-Reduction:
                  Materials handling

 3-03-001-05       Primary metal - Aluminum Ore: Electro-Reduction:
                  Anode baking furnace

 3-03-001-07       Primary metal - Aluminum Ore: Electro-Reduction:
                  Roof vents

 3-03-001-99       Primary metal - Aluminum Ore: Electro-Reduction:
                  Other not classified

 3-03-002-01       Primary metal - Aluminum Hydroxide Calcing:
                  Overall process

 Total
Tons SO. Emitted

          31,178


           1,976


           4,909
           3,448


           3,230


             741


             750


          12.154


          58,386
                                           69

-------
2.2.5.6  Conclusion

       The validity of the TRENDS emission factor for primary aluminum could not be confirmed
and appears suspect for two reasons.  First, it relies on one set of old emissions (not test) data.
Second, there is no documentation of an adjustment due to controls.  There are three emission factors
in the AIRS Facility Subsystem Source Classification Codes and Emission Factor Listing for Criteria
Air Pollutants1  document. Prebake (3-03-001-01) has an emission factor of 57.3 Ibs/ton, HSS  (3-03-
001-02) has an emission factor of 10.0 Ibs/ton, and  VSS (3-03-001-03) has an emission factor  of 17.0
Ibs/ton. (None of these emission factors was derived from AP-42. In fact, AP-42 does suggest a
method for calculating SCX emissions based on other process parameters, but reports no emission
factor.) Many  of the particulate controls on primary aluminum  would be effective controls for S02.
An investigation into the  distribution of the three types of electro-reduction processes and their
controls and how they dominate the primary aluminum industry should be  undertaken to develop an
appropriately weighted emission factor.

       Further, the emission factor used by TRENDS is not consistent with the TRENDS indicators
for industrial processes for the primary metals industry production process  breakdown (as listed in
the TRENDS procedure document as provided in Appendix A).

       Prebake:     71.0 percent of production
       HSS:        18.5 percent of production
       VSS:        10.5 percent of production

AP-42 confirms that Prebake is the most common.19  However,  this is not the weighing used in the
SO: emission factor for primary aluminum.  Using this weighing would provide an emission factor
of:

       57.3(0.71) + 10.0(0.105) + 17.0(0.105) = 43.5  Ibs/ton.

       Using the 43.5 Ibs/ton of aluminum produced emission factor results in a total emission
estimate of 84,000 tons of SO2, exceeding the current TRENDS estimate by 14,000 tons. (Other
                                              70

-------
 weightings used in TRENDS for the emission categories other than primary aluminum may also be
 suspect)

       The NAPAP reported production for materials handling 3-03-001-04 (which should in theory
 represent the sum of the three process types) is 5,371,185 tons of molten aluminum produced.
 However, this SCC represents paniculate emissions and no priority was placed on collecting or
 quality assuring particulate data under the 1985 NAPAP effort.  This materials handling value does
 exceed the sum of the three aluminum processes listed above  (3,879,384 short tons of aluminum
 processed).  Note that the sum of the three NAPAP aluminum process  SCCs approximates the
 production reported in Minerals Yearbook4 (3,850,000 short tons of aluminum).

       Using the  1985 NAPAP production estimates would result in a breakdown between the three
 process types as follows:
       Prebake:     77.2 percent of production
       HSS:         7.9 percent of production
       VSS:         14.8 percent of production
 Using this weighing instead of the TRENDS weighing would  result in an even larger emission factor:

       57.3C772) + 10.0(.079) + 17.0(.148) = 47.5 Ibs/ton.

       If the  1985 NAPAP primary aluminum emission estimate is used to develop a revised
emission factor, an overall factor of 38,063 tons SO2 / 3,850,000 short  tons Al = 19.8 Ibs/ton  Al
produced.  The absolute factor is lower because reported NAPAP emissions from these processes is
only 38,063 tons,  but presumably  accounts for the effect of SO2 controls.

      In the NAPAP inventory, only 38,063 tons of SO: are  attributable to the three processes
covered by the TRENDS method. Of the processes that are not included in TRENDS, aluminum
hydroxide  calcining is the most important.  TRENDS  includes the  aluminum hydroxide calcining
process in the estimation of TSP and PM-10 emissions but not in the SO2 estimate.
                                            71

-------
2.2.6  Secondary Lead

       The published 1985 TRENDS value is 30,000 tons SO2 from secondary lead processing.  The
 1985 NAPAP value is 20,720 for secondary lead. The discrepancy between the two inventories is
 9,300 tons (45 percent).

 2.2.6.1  TRENDS Activity

       In TRENDS, the secondary lead production'is obtained from Table 1 "Salient Lead Statistics"
 of Minerals Yearbook, 1989  "Lead."* The  1985 production of secondary1 lead (based on lead content)
 was 615,695 metric tons (677,264 short tons).  The consumption of lead scrap is obtained from
 Table 8 "Stocks and Consumption of New and Old Lead Scrap in the United States, by Type of
 Scrap" of Minerals  Yearbook, 1986 "Lead."4 The total 1985 consumption of scrap is 804,832 metric
 tons (885,315 short tons).  SO2 emissions are only calculated for reverberatory and blast furnaces.
 There are no emission estimates or emission factors for SO2 from pot furnaces.

       For reverberatory furnaces the TRENDS activity is the fraction of lead recovered as soft lead
 to total lead recovered multiplied by the consumption of scrap. The amount recovered as soft lead is
 obtained  from Table 11 "Lead Recovered from Scrap Processed in The United States, by Kind of
 Scrap and Form of Recovery" of Minerals Yearbook, 1986 "Lead"4 and the value was 273,698 metric
 tons (301,068 short tons).  The 1985 production from reverberatory furnaces is calculated as  follows:

       301,068 / 677,264 * 885,315 = 393,554 tons.

       This value matches the TRENDS activity spreadsheet.

       For  blast furnaces the activity is the fraction of lead recovered as antimonial lead to total lead
recovered multiplied by the total  consumption of lead scrap.  The amount recovered as antimonial
lead is also obtained from Table  11 and was 299,307 metric tons (329,238 short tons).  The  1985
production from blast furnaces is calculated as follows:
                                             72

-------
       329,238 / 677,264 * 885,315 = 430,377 tons.

       The TRENDS activity spreadsheet has a value of 391,300 tons for the secondary lead blast
furnaces.  It appears as though the value in the TRENDS activity spreadsheet was not convened from
metric tons to short tons.

2.2.6.2  TRENDS Emission Factors

       The emission factors cited in the TRENDS method are as follows.
  SCC         Description                            Emission  Units
                                                        Factor
  3-04-004-02   Secondary metal - Secondary lead              80  Ibs/ton lead produced
               production: Reverberatory furnace
  3-04-004-03   Secondary metal - Secondary lead              53  Ibs/ton lead produced
               production: Blast furnace (Cupola)
       These emission factors match the AIRS Facility Subsystem Source Classification Codes and
Emission Factor Listing for Criteria Air Pollutants document.7
2.2.6.3  TRENDS Emissions

       Using the activity data published in the TRENDS activity spreadsheet and the TRENDS
emission factor results in the following 1985 emissions:

       (394,000 * 80 + 391,300 * 53)/ 2000  = 26,129 tons of S
-------
2.2.6.4  NAPAP Activity

       The secondary lead production reported in NAPAP is 216,554 tons of metal charged to the
reverberatory furnace and 332,773 tons of metal charged to the blast furnace (cupola).

2.2.6.5  NAPAP Emissions

       There are ten SCCs or processes listed for secondary lead production.  Of these ten SCCs
only five have associated SO2 emissions. These are listed below.
  SCC         Description                                                         Tons SO,
                                                                                    Emitted
  3-04-004-01   Secondary metal - Secondary lead: Pot furnace
  3-04-004-02   Secondary metal - Secondary lead: Reverberatory furnace
  3-04-004-03   Secondary metal - Secondary lead: Blast furnace (Cupola)
  3-04-004-07   Secondary metal - Secondary lead: Pot furnace heater natural gas
  3-04-004-99   Secondary metal - Secondary lead: Other not classified
  Total                                                                               20,720

2.2.6.6  Conclusion

       There is an apparent error in the TRENDS estimate because the activity value for the blast
furnace was not converted to english units. Following the TRENDS published procedure  and
converting the activity data results in emissions of 27,147 tons of SO2. This is not a significant
difference in the published TRENDS estimate,  because the values that are published are rounded to
the nearest 10,000 tons.

       The TRENDS and NAPAP emission values are somewhat similar, however, the TRENDS
value is 30 percent higher than the NAPAP emissions.  TRENDS does not account for SO, controls
such as baghouses and wet scrubbers.
                                             74

-------
       There is a new SCC, with an SO, emission factor of 144.0S Ibs SCyiO3 gallons burned, for
this category. The SCC is 3-04-004-07 for pot furnace heater burning distillate oil.  This SCC is not
in the TRENDS method and it is not in the NAPAP inventory.

2.2.7  Other NAPAP Non-ferrous Emission Categories

       In addition to the categories discussed above, NAPAP reports additional non-ferrous
emissions of 41,511 tons of SO2. The categories and the NAPAP emission estimates are summarized
in Table 2-10.

2.3 OTHER INDUSTRIAL PROCESSES

       TRENDS estimates SO2 emissions from other industrial processes including pulp and paper,
chemical manufacturing, petroleum refining, iron and steel, and mineral products.  Within the
chemical manufacturing group, TRENDS provides SO2 estimates for sulfuric acid and carbon black
manufacture. Within  the mineral products group, TRENDS estimates SO2 emissions for cement,
glass, and lime manufacturing. Table 2-11  summarizes the 1985 TRENDS and NAPAP emission
estimates for these industrial processes.

       The  1985 NAPAP emission  inventory is a detailed database  with emission estimates reported
in a record format.  The total emissions reported for a source category is dependent upon the way in
which the emissions are summed.  For industrial sources, the inclusion or exclusion of SO: emissions
from fuel combustion  as opposed to process emissions can dramatically effect the emissions
associated with a particular source category. The NAPAP emission estimates reported in Table 2-11
were summed to allow comparison  with the TRENDS source category estimates. For some source
categories fuel combustion emissions are included and for some source categories  they are  not
included.
                                            75

-------
            TABLE 2-10.  OTHER NON-FERROUS EMISSIONS REPORTED
                         IN NAPAP
Primary and Secondary Metals
Source Category (SIC)
Ferroalloy (3313)
Molybdenum Ore Mining (1061)
Barium Ore Processing (3295)
Taconite Iron Ore Processing (1011)
Secondary Aluminum Production (3341,
sec
3-03-006
3-03-007
3-03-011
3-03-014
3-03-023
3-04-001
1985 NAPAP
(tons)3
10,016
986
2.514
1,547
3.768
  3353, 3354, 3355, 3363, 3365)

  Secondary Copper Production (3341, 3364,
  3366)
3-04-002
Lead Battery Manufacture (3691)
Magnesium (3341)
Secondary Zinc Production (3341)
Furnace Electrode Manufacture (3624)
Fugitive Emissions (1000, 3300)
Process Heaters, Incinerators, and
Miscellaneous Not Classified
Total
3-04-005
3-04-006
3-04-008
3-04-020
3-03-888
3-03-900
3-03-999
3-04-999

1,347
27
3,094
3,979
12,060
2,165
41,511
*The 1985 NAPAP Emissions Inventory (version 2): Development of the Annual Data
 and Modelers' Tapes, EPA-600/7-89-012a, November 1989.
2.3.1  Kraft Pulp Production


      The 1985 TRENDS emission value is 250,000 tons SO2. The 1985 NAPAP value is 130,400

tons for both Kraft and sulfite pulp products.3 The discrepancy between the two inventories is
   The NAPAP estimate of 130,400 tons if for emissions from the kraft and sulfite process.  The
NAPAP estimate for all pulp and paper plants, including combustion emissions is 608 000 tons of
SO:

                                         76

-------
TABLE 2-11. COMPARISON OF OTHER INDUSTRIAL PROCESSES
          VALUES FOR 1985 TRENDS AND NAPAP
Category
Kraft Pulp Production (2611, 2621,
2631)
Emissions (tons)
Air-dry unbleached pulp
(103 tons)
Carbon Black Production (2895)
Emissions (tons)
Tons Produced (103 tons)
Sulfuric Acid Production (2819)
Emissions (tons)
Gross, new and fortified
(103 tons)
Sulfur Recovery Production
At Petroleum Refineries
(103 tons)
At Natural Gas Facilities
(103 tons)
Petroleum Refineries (2911)
F.C.C. Emissions (tons)
T.C.C Emissions (tons)
Flares Emissions (tons)
Process Heaters - Oil
Emissions (tons)
Process Heaters - Gas
Emissions (tons)
Sulfur Recovery Emissions
(tons)
Other NAPAP Refinery
Emissions (tons)
Total Emissions (tons)
TRENDS
Published

250,000
44,184

10,000
1,285.5

210,000
39,890

2,940
2,373

326,317
522
35,078
44,360
231,106
202,125

830,000*
TRENDS
Calculated

167,766
44,184

14,585
1,285

215,405
39,890

3,234
2,610

326,317
522

76,911
238,547
172,696

814,993
NAPAP
Published

130,400
29,399

28,000
1,111

217,000





204,647
7,273
15,671
117,512
117,237
29,117
149,925
640,000
(continued)
                      77

-------
        TABLE 2-11.  COMPARISON OF OTHER INDUSTRIAL
                    PROCESSES VALUES FOR 1985 TRENDS AND
                    NAPAP (Continued)
Category
Oil and Gas Production (1311)
Combustion Emissions (tons)
Sulfur Recovery Emissions (tons)
Other NAPAP Natural Gas
Production Emissions (tons)
Total Emissions (tons)
Iron and Steel Industry
Coke Emissions (tons)
Sintering Emissions (tons)
Open Hearth Furnace
Emissions (tons)
Roll and Finish (tons)
Other NAPAP Iron and
Steel Emissions (tons)
Total Emissions (tons)
Raw steel production (103
tons)
Cement Manufacturing (3241)
Emissions (tons)
Clinker produced (103 tons)
Glass Manufacturing (3211, 3221,
3229)
Emissions (tons)
103 Tons produced
Lime Manufacturing (3274)
Emissions (tons)
103 Tons Produced
Total Emissions (106 tons)
TRENDS
Published

441
163,143

160,000

162,000
21,000
4,650
168,000

360,000
88,300

620,000
77,895

30,000
16,245.8

30,000
15,800
2.5
TRENDS
Calculated

441
139,374

139,815

162,000
35,058
1,169
86,948

285,000
88,300

210,500"
77,895

28,508
16,245.8

42,000
15,800
1.9
NAPAP
Published

7,660
59,498
265,000
332,158

74,629
33,058
1,169
25,304
67,985
212,000


290,653
60,166

23,000
10,404

32,000
16,634
1.9
'Does not add to total due to independent rounding.
"Assuming no S0; controls.
                                78

-------
119,600 tons of SO2 (92 percent).

2.3.1.1  TRENDS Activity

       In TRENDS, the production of sulfate and sulfite is obtained from Table 4 "Production and
Shipments of Woodpulp, by Type of Pulp: 1986 and 1985" of Current Industrial Reports: Pulp,
Paper and Board 1986."  The production of sulfate (Kraft pulping) was 42,563,831 short tons in
1985.  The production for sulfite was 1,620,084 short tons in 1985.  The total production of
44,183,900 tons is consistent with the activity value published in the TRENDS activity spreadsheet.

2.3.1.2  TRENDS Emission Factor

       The TRENDS procedure document lists SO2 emissions from two categories:  Kraft pulping
(sulfate)  and sulfite pulping. The emission factor for sulfate (kraft) pulping is 7.0 Ibs/air-dry ton of
unbleached pulp (SCC 3-07-001-04). This emission factor matches the AIRS Facility Subsystem
Source Classification Codes and Emission Factor Listing for Criteria Air Pollutants document.7

       In the case of sulfite, the  emission factor is derived from the uncontrolled factor (52 Ibs/ton)
and the controlled factor (20 Ibs/ton), both based on a now obsolete version of AP-4219, and the
assumption that 90 percent of production is at the controlled rate.  The SO2 emission factor for
sulfite pulping is calculated as:

       (52.0 * 0.10 + 20 * 0.90) = 23.2 Ibs/ton.

2.3.1.3 TRENDS Emissions

      Emissions for kraft (sulfate) pulping are calculated from the Kraft production data.

      42,563,831 * 7.0 / 2000 = 148,973 tons of SO,
                                              79

-------
       Emissions from sulfite pulping (assuming 90 percent control) are calculated from the
weighted emission factor and sulfite production.

       1,620,084 * 23.2 / 2000 = 18,793 tons of SO2

       Total emissions are simply the sum of Kraft and sulfite processes.

       Total emissions = 148,973 + 18,793 = 167,766 tons of SO2

       The sum of these estimates does not match  the published TRENDS estimate.  Because the
activity value did match the TRENDS spreadsheet, the emission factor was researched further.
Backcalculating the TRENDS emission factor from the published emission estimate (250,000 tons of
SO2) and the production data results in an overall emission factor of:

       250,000/44,183,915 * 2000 =11.3 Ibs/ton.

       An emission factor of 11.3 was apparently used in both the 1990 and 1991 TRENDS
estimate.

2.3.1.4  NAPAP Activity

       The NAPAP production cannot be determined without a better understanding of the
woodpulping process.  The production number for the recovery furnace direct contact evaporator in
the kraft pulping category is 29,398,475 air-dry tons of unbleached pulp. It is unclear if this
production should be added to other production numbers or if this value would represent all of the
reported pulp produced through the kraft process.

2.3.1.5  NAPAP Emissions

       The NAPAP emissions for the Kraft and sulfite SCC codes are as follows.
Kraft:

                                             80

-------
 SCC             Description                                        Tons SO. Emitted

 3-07-001-01      Pulp & Paper and Wood Products - Sulfate (Kraft)                  159
                  Pulping: Digester Relief and Blow tank, general

 3-07-001-02      Pulp & Paper and Wood Products  Sulfate (Kraft)                   17
                  Pulping: Washer/screens general

 3-07-001-03      Pulp & Paper and Wood Products  Sulfate (Kraft)                   23
                  Pulping:
                  Multi-effect evaporator general

 3-07-001-04      Pulp & Paper and Wood Products  Sulfate (Kraft)               85.407
                  Pulping: Recovery Furnace Direct contact
                  evaporator

 3-07-001-05      Pulp & Paper and Wood Products  Sulfate (Kraft)                2.357
                  Pulping: Smelt dissolving tank general

 3-07-001-06      Pulp & Paper and Wood Products - Sulfate (Kraft)               11,220
                  Pulping: Lime kiln general

 3-07-001-08      Pulp & Paper and Wood Products - Sulfate (Kraft)                  171
                  Pulping: Fluid bed calciner: general

 3-07-001-09      Pulp & Paper and Wood Products - Sulfate (Kraft)                 455
                  Pulping: Liquor oxidation tower,  general

 3-07-001-10      Pulp & Paper and Wood Products - Sulfate (Kraft)               22,870
                  Pulping: Recovery furnace indirect contact
                  evaporator

 3-07-001-99      Pulp & Paper and Wood Products - Sulfate (Kraft)              	43
                  Pulping: Other not classified

 Total                                                                         122,722

Sulfite:

 SCC             Description                                        Tons SO. Emitted

 3-07-002-03      Pulp & Paper and Wood Products - Sulfite  Pulping:                 44
                  Digester blow pit dump tank:  all  bases except
                  Calcium

 3-07-002-11      Pulp & Paper and Wood Products - Sulfite  Pulping:               1,445
                  Digester blow pit dump tank:  Calcium

 3-07-002-14      Pulp & Paper and Wood Products - Sulfite  Pulping:               1,362
                  Digester blow pit dump tank:  NH3 with process
                  change

 3-07-002-21      Pulp & Paper and Wood Products - Sulfite  Pulping:                 417
                  Recovery system MgO

 3-07-002-22      Pulp & Paper and Wood Products - Sulfite  Pulping:                 543
                  Recovery system NH3

 3-07-002-23      Pulp & Paper and Wood Products - Sulfite  Pulping:               1,814
                  Recovery system Na
                                             81

-------
 SCC             Description                                        Tons SO. Emitted
 3-07-002-31      Pulp & Paper and Wood Products - Sulfite Pulping:                193
                  Acid plant NH3
 3-07-002-32      Pulp & Paper and Wood Products   Sulfite Pulping:                  5
                  Acid Plant Na
 3-07-002-33      Pulp & Paper and Wood Products - Sulfite Pulping:                249
                  Acid plant Ca
 3-07-002-34      Pulp & Paper and Wood Products   Sulfite Pulping:                  8
                  Knotters washers screens etc.
 3-07-002-99      Pulp & Paper and Wood Products   Sulfite Pulping:                608
                  Other not classified
 Total                                                                         6,688
       Additional emissions of 975 tons are listed in NAPAP for semi-chemical (neutral sulfite)
woodpulp pulping. Therefore, total pulp production emissions equal  130,385 tons of SO2.

2.3.1.6  Conclusion

       For kraft pulping, the NAPAP and TRENDS activity values are 29,398,475 and 42,563,831
tons, respectively.  The NAPAP production value does not report production for all of the records
where emissions are reported and therefore the production value is low relative to the emission
estimate.

       The TRENDS method apparently used an emission factor of 11.3 to calculate kraft emissions,
when a more appropriate value would have been 7  Ibs/ton of air-dry unbleached pulp.  It is unclear
how the overall emission factor of 11.3 Ibs/ton of air-dry unbleached pulp  was derived; it was not
derived from the methodology listed in the TRENDS procedures manual, at least not for  the 1985
study year.  The 11.3 emission factor is significantly higher (nearly 50 percent) than one calculated
following  the TRENDS procedure.  The TRENDS emission estimation procedures from the pulp
industry should be revisited based on seeming discrepancies in the TRENDS emission factor(s) and
the development of new emission factors.  Based solely on the discrepancy, TRENDS emissions may
be overstated by nearly 70,000 tons. Also, TRENDS does not account for the  effect of any controls.
These two issues could result in an overestimation  of SO: emissions from  wood pulping processes.

                                             82

-------
       The TRENDS procedure does not include a third type of paper pulping process, semi-
chemical. Activity data for semi-chemical pulping are available and emission factors exist in the
AP-42.19  Published statistics indicate that semi-chemical has recently overtaken sulfite (3.9 x 106
versus 1.6 x 106 tons of production).  If the semichemical production is increasing with fewer
associated SO: emissions, this is an industry trend that should be reflected in the emission estimates.

       The TRENDS procedure needs to be revised to account  for the new emission factors that
have been developed in 1990.  The vast majority of pulp is produced through the Kraft process.
therefore expenditure of significant effort to improve the sulfite factor may not be warranted. The
TRENDS number should be produced as at least two separate numbers, Kraft and sulfite, but
possibly as three numbers to include semi-chemical. The TRENDS procedure should also be
rewritten to document the development of the emission factor and control assumptions.

2.3.2 Chemical Manufacturing

       Within the chemical manufacturing group, TRENDS estimates SO2 emissions for carbon black
and  sulfuric acid manufacture.  TRENDS also estimates elemental sulfur production, however, the
emissions are categorized as recovered sulfur within the petroleum refining and natural gas
production categories.

2.3.3 Carbon Black Production

       The 1985 TRENDS emission value is 10,000 tons SO2.  The 1985 NAPAP value is 28.000
for carbon black production. The discrepancy between the two inventories is 18,000 tons of SO: (64
percent).

TRENDS carbon black activity

      The total  production of carbon black produced as reported in "Facts and Figures for the
Chemical Industry" in the Chemical and Engineering News is 2.57 billion pounds in 1985.1: This is
equivalent to a 1985 production value of 1.285,000 tons.  This value matches the TRENDS activity

                                             83

-------
spreadsheet  Carbon black production is divided into oil and gas processes; TRENDS creates this
split using a fixed assumption (i.e., this split is not updated annually).
       Oil Process:  90 percent of total production
       Gas Process:  10 percent of total production

TRENDS carbon black emission factors

       The TRENDS methodology relies  on the emission factor for the flared furnace published in
Table 5.3-3, AP-42  (Fourth Edition)19 (there are no SO, emission factors in the  AIRS SCC
document).  The TRENDS procedure starts with uncontrolled emissions.

       Description                                           Ibs SO./ton
       Flared Furnace Exhaust (Oil Process)                     50

       The TRENDS methodology for the remaining steps is not clear.  The procedures manual
states that a controlled emission factor (based on CO control efficiency) is calculated as follows:

       EF = CO Control Efficiency * 50  Ibs/ton.

The emission factor reported in the TRENDS printout is 22.7 Ibs/ton, indicating a CO control
efficiency of:

       CO control efficiency = 22.7 / 50  = 0.454 or 45 percent.

       It is not clear if this control efficiency is to  be applied to CO emissions  or if the control
efficiency is the result of a CO boiler and incinerator. The CO boiler and incinerator has a lower
SO, emission factor than the flare (35.2 Ib/ton versus 50 Ib/ton).

       There is no emission factor reported for the gas process.  In addition, it  does not appear that
the activity value  is  weighted toward oil versus gas.
                                              84

-------
TRENDS carbon black emissions



       Using the TRENDS emission factor results in emissions of:



       1,285,000 * 22.7 /  2000 =  14,585 tons of SO,.



       The published TRENDS value is rounded to the nearest 10,000 tons and is therefore reported

as 10,000 tons of SO:.



NAPAP carbon  black activity
       The 1985 NAPAP inventory reports the following nine separate SCCs.
 SCC             Description

 3-01-005-01      Chemical manufacturing - Carbon Black Production:
                  Channel Process

 3-01-005-02      Chemical manufacturing - Carbon Black Production:
                  Thermal Process

 3-01-005-03      Chemical manufacturing - Carbon Black Production:
                  Gas Furnace Process, Main Vent

 3-01-005-04      Chemical manufacturing - Carbon Black Production:
                  Oil Furnace Process, Main Vent

 3-01-005-06      Chemical manufacturing - Carbon Black Production:
                  Transport Air Vent

 3-01-005-07      Chemical manufacturing - Carbon Black Production:
                  Pellet Dryer

 3-01-005-08      Chemical manufacturing - Carbon Black Production:
                  Bagging/Loading

 3-01-005-09      Chemical manufacturing - Carbon Black Production:
                  Furnace Process, Fugitives

 3-01-005-99      Chemical manufacturing - Carbon Black Production:
                  Other Not Classified

NAPAP carbon black emissions
Tons produced

       29,000


            0


       98,179


     1,013,232


      821,308


      981.213


       38,600


       15,824


       134.879
      The following emissions are reported for Carbon Black Production in the NAPAP inventory:
                                            85

-------
 SCC

 3-01-005-01


 3-01-005-02


 3-01-005-03


 3-01-005-04


 3-01-005-06


 3-01-005-07


 3-01-005-08


 3-01-005-09


 3-01-005-99

 TOTAL


Conclusion
Description

Chemical manufacturing   Carbon
Channel Process

Chemical manufacturing - Carbon
Thermal Process

Chemical manufacturing - Carbon
Gas Furnace Process, Main Vent

Chemical manufacturing   Carbon
Oil Furnace Process, Main Vent

Chemical manufacturing - Carbon
Transport Air Vent

Chemical manufacturing   Carbon
Pellet Dryer

Chemical manufacturing - Carbon
Bagging/Loading

Chemical manufacturing - Carbon
Furnace Process, Fugitives

Chemical manufacturing - Carbon
Other Not Classified
                                                                      s SO. Emitted
                                                Black Production:


                                                Black Production:


                                                Black Production:


                                                Black Production:


                                                Black Production:


                                                Black Production:


                                                Black Production:


                                                Black Production:


                                                Black Production:
    10


   197


 1,075


 3.958


 2.410


15.183


   642


 4,013


   543


28,031
       The carbon black production emission estimates are 28,031 tons of SO2 in NAPAP versus

14,585 tons of SO2 in TRENDS. The TRENDS method appears to underestimate the emissions from

carbon black manufacture.



       The total NAPAP production for the oil furnace of 1,013,232 is very similar to the value

TRENDS references  (90 percent of total production) 1,156,500.  The NAPAP value for the gas

furnace 98,179 is less similar to the value TRENDS references (10 percent of total production of

128,500 tons).



       Two questionable items need to be addressed regarding the NAPAP emission estimates and

the TRENDS emission factor.  The NAPAP emission estimates by SCC show only a minority of

emissions from the oil furnace (3,958 out of 28,031 tons), although logically this would be the

source of most emissions.  However, the pellet dryer combustion furnace (with  emissions of 15.183
                                            86

-------
tons) is, in essence, a thermal incinerator and emissions associated with the furnace itself are emitted
here. It is likely that engineers coding the NAPAP inventory indicated the vents as discrete emission
points in addition to the oil furnace emission source.

       Second, the TRENDS emission factor appears to be too small.  The TRENDS procedure
initially has a fairly high emission factor of 50 Ibs/ton for the flare from an oil furnace process (this
emission factor is supported by AP-42).  If a CO boiler and incinerator exist, primarily to control CO
emissions, the AP-42 SO; emission factor drops to 35.2 Ibs/ton.  It is unlikely that the emission
factor would drop as low as 22.7  Ibs/ton  (the TRENDS number), even if all  sources had a CO boiler
and incinerator.

       If the TRENDS estimate were  computed with  the original emission factor of 50 Ibs/ton, the
TRENDS emissions would be:

       1,285,000 * 50 / 2000 = 32,125 tons of SO2.

This value is very close to the NAPAP estimate of 28,031 tons of SO2.

       If all the sources are assumed to be controlled with a CO boiler and incinerator, the emission
factor would be  35.2 Ibs/ton and the emissions would be:

       1,285,000 * 35.2 / 2000 = 22,616 tons of SO2.

       As noted, the largest source of emissions in NAPAP is for the pellet  dryer. There is no
corresponding category in TRENDS for the pellet dryer, although it is likely that emissions have
been accounted for.   As stated earlier, there is no NAPAP category specifically for the flare.
however, the flare actually represents otherwise uncontrolled oil furnace emissions.

       The TRENDS  documentation probably needs to be modified to  ensure that all emission points
and sources are included.  Further investigation of the NAPAP value is warranted to determine why
emissions associated  with the oil furnace were distributed to other emission points (vents) if possible.

                                              87

-------
Finally, the assumption that flares represent otherwise uncontrolled emissions could be confirmed by
looking at the control equipment for these sources coded in NAPAP.

2.3.4  Sulfuric Acid

       The 1985 TRENDS value is 210,000 tons SO2.  The  1985 NAPAP value is 217,000 for
sulfuric acid production. The discrepancy between the two inventories is 7,000 tons of SO; (3
percent).

TRENDS sulfuric acid activity

       The 1985 production of sulfuric acid was obtained from Table 1 "Production  and Shipments
of Selected Inorganic Chemicals: 1982 to 1986" of Current Industrial Reports, Inorganic
Chemicals2* and was 39,889,900 short tons, gross (new and fortified).

       The 1984 production is used in the calculation of the emission factor. The 1984 production
was obtained from the same source and was 41,801,900 short tons.  The TRENDS activity
spreadsheet lists the 1984 production as 39,683,000 tons  of sulfuric  acid.

TRENDS sulfuric acid emission factor

       The TRENDS procedure manual provides the following equation for the development of an
SO2 emission factor.  The equation assumes that each year 5 percent of the existing  production
comes into compliance with the New Source Performance Standard  (NSPS)b and  any new production
will also be in compliance with the NSPS.

       EF, = (0.95 * EF., * P.,,) + (0.05 * EF..W * P..,) + ((P. - P.,) * EFM,0J
                                        P,
      40 CFR Section 60 Subpart H.
                                             88

-------
where,

    i          =  1985
    EFNSPS    =  NSPS emission factor (4 Ibs SCyton of 100 percent H2SO4 production)
    P         =  Total Production
       Because the production for 1985 (P,) is less than the production for 1984 (P^,) (based on the
information in Current Industrial Reports), the last term in the equation should be set to zero.
Emissions from new capacity at the  NSPS emission level should only apply to production above the
previous record-high production level.

       In order to apply this equation, one must know the 1984 emission factor.  The 1984 emission
factor is backed out based on the 1984 production (as reported in the TRENDS activity spreadsheet)
and the 1984 published TRENDS emission estimate.  Note that the published TRENDS 1984
activity is lower than the 1985 production which reintroduces the last term of the equation. The
1984 emission factor is:

       210,000 /  39,683,000 * 2000 = 10.6 Ib/ton of H,SO4 production.

The 1985 emission factor is then calculated using the above equation:

      EF1985  = [(0.95 * EF1984 * P19M) + (0.05 * 4 * P1984) + (P1985 - P1984 * 4] / P1985
             = [(0.95*10.6*39,683,000)+(0.05*4*39,683,000)+(206,900*4)]/39,889,900
             = 10.2 Ib/ton of sulfuric acid produced

       (The emission factor for 1985 that is backcalculated based on the published TRENDS
emission estimate and the  1985 production is  10.5 Ib/ton of sulfuric acid produced.)
                                              89

-------
TRENDS sulfuric acid emissions

       The TRENDS emissions were not calculated using the emission factor derived from the
equation, rather the emissions were calculated using an emission factor of 10.5 Ib/ton of sulfuric acid
produced. The emissions were calculated as follows:

       39,889,900 * 10.5 / 2000 = 209,421  tons of SO,.

       If the TRENDS procedure is followed and the correct 1984 activity data are used, the
emission factor for 1985 would be 10.8 Ib/ton of sulfuric acid produced and the 1985 emissions
would be slightly higher.

       39,889,900 * 10.8 / 2000 = 215,405 tons of SO2.

NAPAP sulfuric acid activity

       The 1985 NAPAP inventory reports production for sulfuric acid production as follows.
 SCC            Description                                              Tons 100%
3-01-022-01
3-01-023-01
3-01-023-04
3-01-023-06
3-01-023-08
3-01-023-10
3-01-023-12
3-01-023-18
Chemical Manufacturing - Sulfuric Acid
process: general
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 99.9% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 99.5% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 99.0% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 98.0% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 97.0% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 96.0% conversion
Chemical Manufacturing Sulfuric Acid
- Chamber
- Contact
- Contact
- Contact
- Contact
- Contact
Contact
- Contact
211,236
14,886,621
1,592,842
3,462,965
7,113.929
2,842,025
40,588
3,552,216
                  process: absorber @ 93.0% conversion
                                             90

-------
sec
3-01-023-19
3-01-023-20
3-01-023-21
3-01-023-22
3-01-023-99
NAPAP sulfuric
The 1985
sec
3-01-022-01
3-01-023-01
3-01-023-04
3-01-023-06
3-01-023-08
3-01-023-10
3-01-023-12
3-01-023-18
3-01-023-19
3-01-023-20
3-01-023-21
3-01-023-22
Description
Chemical Manufacturing - Sulfuric Acid
process: Concentrator
Chemical Manufacturing Sulfuric Acid
process: Tank car and truck unloading
Chemical Manufacturing - Sulfuric Acid
process: Storage tank vent
Chemical Manufacturing - Sulfuric Acid
process: Process equipment leaks
Chemical Manufacturing Sulfuric Acid
process: Other Not Classified
acid emissions

- Contact
- Contact
Contact
- Contact
Contact

Tons 100%
H.SO.
40,050
247,520
2,335.790
278,551
437.373

reported emissions for sulfuric acid production are as follows.
Description
Chemical Manufacturing - Sulfuric Acid
process: general
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 99.9% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 99.5% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 99.0% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 98.0% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 97.0% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 96.0% conversion
Chemical Manufacturing - Sulfuric Acid
process: absorber @ 93.0% conversion
Chemical Manufacturing Sulfuric Acid
process: Concentrator
Chemical Manufacturing - Sulfuric Acid
process: Tank car and truck unloading
Chemical Manufacturing - Sulfuric Acid
process: Storage tank vent
Chemical Manufacturing - Sulfuric Acid
process: Process equipment leaks
Tons
- Chamber
- Contact
- Contact
- Contact
- Contact
- Contact
- Contact
- Contact
- Contact
Contact
- Contact
- Contact
SO, Emitted
2,091
32,947
4,300
13,551
76,707
25,893
1,445
10,905
0
1
94
46,930
91

-------
 SCC             Description                                        Tons SO. Emitted
 3-01-023-99      Chemical Manufacturing - Sulfuric Acid - Contact                2.302
                  process: Other Not Classified
 Total                                                                       217,166
Conclusions
       The NAPAP estimate of 217,166 tons of SO: is extremely close to the TRENDS estimate of
215.405 tons of SO,. Nevertheless, the origin of the initial emission factor in the TRENDS
procedure is not documented and therefore the emission factor for the category is somewhat suspect.
One concern about this category is the production of H;SO4 from recovered sulfur.  The NSPS does
not apply to sulfuric acid production in conjunction with SO2 controls.  It is unclear if the NAPAP
data reflect only the chemical companies producing sulfuric acid or also include byproduct H2S04
production.

       There appears to be an error in the emission factor used to develop the published TRENDS
estimate. If the TRENDS procedure is followed in the development of the 1985 emission factor
(using  the 1984 emission factor and the correct 1984 production), the 1985 emission factor is 10.8
Ib/ton of sulfuric acid produced as opposed to the value of 10.5 Ib/ton of sulfuric acid produced,
which  was apparently used in the published TRENDS report.

       The NSPS emission factor of 4 Ib SOj/ton of 100 percent sulfuric acid produced is consistent
with the emission factor for sulfuric acid contact process, 99.9 percent conversion (SCC 3-01-023-
01).  As shown in the NAPAP data, the activity for that  SCC dominates the category. An analysis of
the penetration of the NSPS emission limit into this source category could be conducted by
contacting the Stationary Source Compliance Division of OAQPS and by  analyzing the BACT/LAER
Clearinghouse. As stated above, the estimates are extremely close and  may not warrant any further
investigation.

       An analysis of the production data  provided in Current Industrial Reports, Inorganic
Chemicals24 reveals that production had a low value of 33,233,000 tons of sulfuric acid in 1982 and
a high  of 44,336,818 in 1990. Because the NSPS was promulgated  in the 1970's, production over

                                             92

-------
33,233,000 (and at least 25 percent of production) should be at the NSPS level. The NAPAP data
indicate that approximately 50 percent of 1985 production was at the NSPS level.

2.3.5  Sulfur Recovery Plants

       TRENDS reports SO: emissions from sulfur recovery plants in two categories, natural gas
production and  petroleum refining.  As a result, it is not possible  to directly assess whether the
published TRENDS emission estimates were successfully recreated due to errors in both the
published TRENDS activity data and emission factors.  The TRENDS emission estimates using the
erroneous TRENDS spreadsheet information are 202,000 tons  for petroleum refineries and 163,000
tons for natural gas production.  The corresponding NAPAP emission estimates are 29,000 tons for
petroleum refineries and 59,000 tons for natural gas production.

2.3.5.1  TRENDS Activity

       The quantity of sulfur recovered by natural gas plants and petroleum refineries in 1985 is
reported in Table 4 "Recovered Sulfur Produced and Shipped in the United States" of Minerals
Yearbook 1989  "Sulfur."4 The production at petroleum refineries  was 2,940,000 metric tons
(3,234,000 short tons) and the production at natural gas plants was 2,373,000 metric tons (2,610,000
short tons).

       The TRENDS activity spreadsheet erroneously  left these activity rates in metric units.

2.3.5.2 TRENDS Emission Factors

       According to the TRENDS procedure manual, the emission factor is derived  annually from
the emissions and throughput reported to  AIRS/FS in SCC 3-01-032-01  through 3-01-032-04. In
1985 the  throughput values were reported as follows.
 SCC            Description                                             Tons 100% Sulfur
 3-01-032-01     Chemical Manufacturing  Elemental Sulfur Production:              154,728
                 Mod. Glaus: 2 stage w/o control (92-95% removal)

                                             93

-------
 sec
 3-01-032-02

 3-01-032-03

 3-01-032-04

 Total
Description

Chemical Manufacturing - Elemental Sulfur Production:
Mod. Claus: 3 stage w/o control (95-96% removal)

Chemical Manufacturing  Elemental Sulfur Production:
Mod. Claus: 4 stage w/o control (96-97% removal)

Chemical Manufacturing  Elemental Sulfur Production:
Sulfur removal process (99.9% removal)
       In 1985 the emissions were reported as follows.
  SCC
  3-01-032-01
  3-01-032-02
  3-01-032-03
  3-01-032-04
 Total
 Description

 Chemical Manufacturing  Elemental Sulfur Production:
 Mod. Claus: 2 stage w/o control (92-95% removal)

 Chemical Manufacturing - Elemental Sulfur Production:
 Mod. Claus: 3 stage w/o control (95-96% removal)

 Chemical Manufacturing - Elemental Sulfur Production:
 Mod. Claus: 4 stage w/o control (96-97% removal)

 Chemical Manufacturing - Elemental Sulfur Production:
 Sulfur removal process (99.9% removal)
Tons 100% Sulfur

         754,087


         105,890


        1.374.263


        2.388,968



Tons SO. Emitted

          32.566


          30,301


           5,281


          59.386
                                                               127,534
       The emission factor is obtained by dividing the total emissions by the sum of the operating

rates. The emission factor for 1985 should be:



       127,534 / 2,388,968 * 2000 = 106.8 Ibs/ton of sulfur.



       The emission factor reported through the TRENDS spreadsheets for both 1990 and 1991 is

137.5 Ib/ton of sulfur produced.



2.3.5.3  TRENDS Emissions



      The TRENDS emission estimates are calculated for both natural gas production and petroleum

refining using the emission factor of 137.5 Ib/ton of sulfur produced and the production rates in the

TRENDS activity spreadsheet For petroleum refineries the emissions are  202,125 tons:



      2,940,000 * 137.5 / 2000 = 202,125 tons SO,
                                            94

-------
The emissions for natural gas fields are 163,143 tons:

       2,373,000 *  137.5 / 2000 = 163,143 tons SO2

The total emissions  are 365,269 tons:

       202,125 + 163,143 = 365,269  tons SO2

       Using the emission factor of 106.8 Ibs/ton of sulfur produced derived by following the
TRENDS procedure manual  and the correct activity data would result in the following emission
estimates for petroleum refineries:

       3,234,000 *  106.8 / 2000 = 172,696 tons SO2

The emissions at natural gas fields are 139,374 tons:

       2,610,000 *  106.8 / 2000 = 139,374 tons SO2

The total emissions  are 312,070 tons:

       172,696 + 139,374 = 312,070  tons SO2

2.3.5.4  NAPAP Activity

       The 1985 reported throughput  values were reported above.  In addition NAPAP reports
191,676 tons product under Other not classified.  The production sums to 2,580,644 tons of
recovered  sulfur versus the 5,844,000 tons reported through the Minerals Yearbook.4 Only half of
the  records that reported sulfur emissions reported a sulfur production  or throughput value.
                                             95

-------
2.3.5.5 NAPAP Emissions

       In order to split the emissions between petroleum refineries and natural gas production, the
sulfur production emissions were put in a matrix of the SCC versus the SIC. The matrix is presented
below.
                                   SO2 EMISSIONS  (tons)
SIC
1300
1311

1321
2801
2819
2869
2911
3011
4922

Total
Description
Oil and Gas Extraction
Crude Petroleum and Natural
gas
Natural gas liquids
Chemicals and Allied Products
Industrial Inorganic Chemicals,
Not elsewhere classified
Industrial Organic Chemicals,
Not elsewhere classified
Petroleum Refining
Tires and Inner Tubes
Natural Gas Transmission
Other Not Classified

SCC
30103201

9,829

4,482

5,366
5,674
7,175
40


32,566
sec
30103202
541
9,897



4,875
1,133
12,100

1,756

30,302
see
30103203

670

2,629
908
982

92



5,281
sec
30103204

21,807

7,887

19,942

9,750



59,386
Total
541
42.203

14.998
908
31,165
6,807
29,117
40
1,756
1,171
128,706
       Natural gas production includes emissions from four SIC's: 1300, 1311, 1321, and 4922 and
were 59,498 tons of S02.  Petroleum Refining was represented by only one SIC and emissions were
29,117 tons of S02. In the NAPAP inventory, tire and chemical production were responsible for the
remaining 38,920 tons of SO2.  In addition, NAPAP reports 1,171 tons of SO2 emissions under Other
not classified.  The total NAPAP emissions sum to  128,706 tons of SO2.

2.3.5.6  Conclusion

       TRENDS reports SO2 emissions from sulfur recovery plants in two categories, natural eas
production and petroleum  refining.  As a result, it is not possible to directly assess whether the

                                             96

-------
published emission estimates were successfully recreated, although using the emission estimates does
allow the total natural gas production and total petroleum refinery estimates to match the published
values.  Errors were discovered in both the activity data and the emission factors that were used to
calculate the published 1985 TRENDS estimates.  As a result, the published TRENDS emission
estimate is too high.  The estimates using the erroneous information are 202,000 tons for petroleum
refineries and 163,000 tons for natural gas production. The corresponding NAPAP emission
estimates are 29,117 tons for petroleum refineries and 59,498 tons for natural gas production.

       In the TRENDS estimate, the activity data were erroneously left in metric units rather than
converted to english units.  The emission factor was not calculated from AIRS data, as the procedure
manual indicated, but rather was held constant.  Using the revised emission factor (106.8 Ibs/ton of
sulfur produced versus 137.5 Ibs/ton of sulfur produced) and corrected  activity data resulted in
TRENDS emission estimates of 172,696 tons for petroleum refineries and 139,374 tons for natural
gas production (a decrease of 53,198 tons of
       The estimates in NAPAP and TRENDS are still very different; 312,070 versus 88,615 tons of
SO2.  The TRENDS method, when applied correctly, probably overestimates emissions from sulfur
recovery plants.  NAPAP may underestimate emissions from sulfur recovery plants or the emissions
may be reported in other areas of the inventory (i.e., under petroleum refining or natural gas
production as opposed to sulfur production).

       The emission factors reported in the AIRS Facility Subsystem Source Classification Codes and
Emission Factor Listing for Criteria Air Pollutants1 are as follows.
 SCC         Description                                       Emission   Units
                                                                  Factor
 3-01-032-01  Chemical Manufacturing  Elemental Sulfur             280   Ib/tons
              Production: Mod. Claus: 2 stage w/o control (92-              100% sulfur
              95% removal)
 3-01-032-02  Chemical Manufacturing - Elemental Sulfur             189   Ib/tons
              Production: Mod. Claus: 3 stage w/o control (95-              100% sulfur
              96% removal)
 3-01-032-03  Chemical Manufacturing  Elemental Sulfur             145   Ib/tons
              Production: Mod. Claus: 4 stage w/o control (96-              100% sulfur
              97% removal)
                                              97

-------
 SCC         Description                                      Emission  Units
                                                                  Factor
 3-01-032-04  Chemical Manufacturing - Elemental Sulfur                4  Ib/tons
              Production: Sulfur removal process (99.9%                    100% sulfur
              removal)

       Multiplying the NAPAP sulfur recovery production values by these emission factors results in
total emissions of 103,348 tons of SO2.  There is a discrepancy between the production values.
emission factors, and reported emissions in the NAPAP inventory.  The emissions for 95-96 percent
recovery appear to be underestimated and  the emissions for 99.9 percent recovery appear to be
overestimated. Therefore, there are probably errors in the NAPAP values, either in the reported
production or in the reported emissions.

       Additional research should be expended on this category to try and determine what types of
sulfur recovery plants are in use in petroleum refineries and natural gas production fields.  Once
there is additional information, a new appropriately weighted emission  factor could be developed for
the TRENDS procedure.
2.3.6  Petroleum Refineries

       In TRENDS the SO2 emissions from petroleum refineries are reported for six categories:
Thermal Catalytic Cracking; Fluid Catalytic Cracking; Flares; Process Heaters  Oil;  Process Heater -
Gas; and Sulfur Recovery. The 1985 TRENDS value is 830,000 tons SO2.  The 1985 NAPAP value
for the same six  categories is 490,000 tons of SO2, although the total petroleum refining estimate is
640,000 tons of SO2. The difference between the two petroleum refining emission estimates is
190,000 tons of SO2 (30 percent).
                                             98

-------
2.3.6.1  TRENDS Activity

       The TRENDS activity data are presented below for the six categories.  The full discussion of
the derivation of sulfur recovery emissions was presented earlier but is summarized below. In all of
the categories except sulfur recovery, the activity data derived following the TRENDS procedure
manual were consistent with the published TRENDS activity spreadsheet.

Catalytic cracking

       In TRENDS,  the petroleum production activity data were obtained from the "Annual Refining
Survey" of the Oil & Gas Journal?  Prior to  1989, the survey also provided the split between fluid
catalytic cracking (FCC) and thermal catalytic cracking (TCC or thermofor).  Within the survey.
footnotes indicate whether the fresh feed for catalytic cracking is fluid or "other".  "Other" represents
thermal cracking. The TRENDS method has changed recently due to the discontinuation of the
reporting of fluid versus thermal catalytic cracking. The following discussion pertains to the
documented method for the development of the 1985 estimates.

       The total capacity of catalytic cracking fresh feed is obtained from the Annual Refining
Survey in the Oil and Gas Journal?5 The total fresh feed, catalytic cracking,  charge capacity as of
January 1, 1986 was  5,234,100 barrels per stream day.

       To convert from capacity to annual throughput, the ratio of production to capacity is obtained
from the Bureau of Economic Analysis, Refinery Operating Ratio, Crude Petroleum.26 For 1985 the
value is 78 percent. The conversion factor for stream day to year is 328.5 (365 calendar days per
year * 0.9 calendar day per stream day).

      Total 1985 catalytic  cracking throughput was:

      5,234,100 * 328.5 * 0.78 = 1,341,133,000 barrels/year.
                                              99

-------
       In the Annual Refining Survey, the Oil & Gas Journal25 differentiates between "Fluid" and
"Other" catalytic cracking.  Fluid cracking was 5,166,300 barrels per stream day (1,323,761,000
barrels/year). Other (thermal) was 67,800 barrels per stream day (17,372,394 barrels/year).

       FCC = 1,323,761 x  103 bbl/year
       TCC =   17,372 x 103 bbl/year
Flares
       The activity for flares is derived from the total refinery crude capacity. This capacity is
converted to an annual value and is multiplied by the percent control efficiency for blowdown
systems from the TRENDS procedure volatile organic compound (VOC) section.

       The TRENDS activity spreadsheet reports a 1985 value as 2,608 x 106 bbl.

Process heaters

       Process heater emissions are divided into oil and gas. Both categories use activity data from
Table 43 "Fuels  Consumed at Refineries by PAD District, 1985" of Petroleum Supply Annual I9855

       The quantity of oil consumed  at petroleum refineries is the sum of distillate, residual, and
crude oil.  The oil consumed in 1985 was reported as:

       758 + 84 +  13,326 = 14,168 x 103 barrels = 595,056 x 103 gallons.

       The quantity of gas consumed at petroleum refineries is the sum of natural gas and still
(process) gas.  Natural gas is reported in 106 ft3  (million cubic feet).  Still (process) gas is reported in
103 bbl and must be converted using the conversion factor of 6.3 x  106 ft3/103 barrel equivalent  The
gas consumed  in  1985 was reported as:

       487,830 + 212,443 * 6.3 =  1,826,220 x 106 ft3.

                                              100

-------
Sulfur recovery



       The quantity of sulfur recovered by petroleum refineries in 1985 is reported in Table 4

"Recovered Sulfur Produced and Shipped in the United States" of Minerals Yearbook 1989 "Sulfur."4

The production from petroleum refineries was 2,940,000 metric tons (3,234,000 short tons).  The

TRENDS activity spreadsheet erroneously left this activity rate in metric units.



2.3.6.2  TRENDS Emission Factors



       The emission factors for this category are rather complex.  The emission factors cited in the

TRENDS procedures manual or in the TRENDS spreadsheets are as follows.


  SCC         Description                     Emission Factor  Units

  3-06-001-05   Petroleum Industry - Process                 0.6  lbs/106 ft3 burned
               Heaters: Natural Gas-fired

  3-06-001-06   Petroleum Industry - Process             950.0S  lbs/106 ft3 burned
               Heaters: Process Gas-fired

  3-06-002-01   Petroleum Industry - Fluid                  493  lbs/103 BBL  fresh feed
               Catalytic Cracking Units

  3-06-002-02   Petroleum Industry - Thermal                  60  lbs/103 BBL  fresh feed
  [sic]          Catalytic Cracking Units

  3-06-004-01   Petroleum Industry -                        26.9  lbs/103 BBL  refinery
               Blowdown systems: w/ vapor                      feed
               recovery sys. w/ flaring

  1-02-004-01   External Combustion Boilers -            158.6S  lbs/103 gallon burned
               Industrial - Residual Oil:
               Grade 6 oil

       The SCC for thermal catalytic cracking (3-06-002-02) appears to be a typographical error.

The correct SCC for thermal catalytic cracking units is 3-06-003-01.  The SCC 3-06-003-01 will be

used throughout the rest of this discussion.



       These emission factors match the document AIRS Facility Subsystem Source Classification

Codes and Emission Factor Listing for Criteria Air Pollutants.1  The TRENDS procedure manual

instructs the user to use the TRENDS emission factor for industrial - residual  oil  boilers and also to

estimate the sulfur  content from a standard AIRS report for SCC 3-06-001-03 (Petroleum  Industry -

Process Heaters: Oil fired).  These two instructions are inconsistent because the TRENDS procedure


                                             101

-------
for industrial-residual oil does not use the AIRS sulfur content.  The average sulfur content as
reported in NAPAP for SCC 3-06-001-03 is 1.09 percent, resulting in an emission factor of:

       158.6 * 1.09 = 172.8 lb/103 gallon burned.

The  emission factor for the TRENDS industrial residual oil category is based on a sulfur content
from Heating Oils, 79S56 and for 1985 this was calculated as 1.63 percent.  This results in an
emission factor of:

       158.6 * 1.63 = 258.5 lb/103 gallon burned.

Finally, the TRENDS spreadsheets for both the 1990 and  1991 estimates used an emission factor of
6248 lbs/103 bbl burned (148.76 lbs/103 gallon burned). This translates to a sulfur content of 0.94
percent.

       The emission factor for natural gas and refinery gas is to be weighted based on the
throughput listed in Table 43 "Fuels Consumed at Refineries by PAD District, 1985" of Perroleum
Supply Annual 1985.5 The emission factor for refinery gas is listed in the TRENDS procedures
manual as  356.25 lbs/106 ft3.  This translates into an assumed sulfur content of 0.375 percent.  The
emission factor for combustion of natural gas is 0.6 lbs/106 ft3. The total gas combustion  is
1,826,220 x 106 ft3.  Natural gas is 26.7 percent of the total gas burned and process gas is 73.3
percent of the total gas burned.  Therefore the weighted emission factor is:

       0.267 * 0.6 + 0.733 * 356.25 = 261  lb/106 ft3 burned.

       The weighted average emission  factor in the TRENDS spreadsheet for process heaters gas
combustion was 253.1 for both  1990 and  1991.
                                              102

-------
Sulfur Recovery

       According to the TRENDS procedure manual, the emission factor is derived annually from
the emissions and throughput reported to AIRS/FS in SCC 3-01-032-01 through 3-01-032-04. In
1985 the throughput values were reported as 2,388,968 tons of sulfur.  In 1985 the emissions are
reported as 127,534 tons of SO2.  The emission factor is obtained by dividing the total emissions by
the sum of the operating rates.  The emission factor for 1985  should be:

       127,534 / 2,388,968 * 2000 = 106.8 Ibs/ton of sulfur.

       The emission factor reported through the TRENDS spreadsheets for both 1990 and 1991  is
137.5 Ib/ton of sulfur produced.

2.3.6.3  TRENDS Emissions

       Using the TRENDS emission factors and activity data published in the TRENDS activity
spreadsheet and adding the emissions from sulfur recovery described in the preceding section, results
in the following  1985 emissions.
 Description                                             Tons SCX Emitted
 Fluid  Catalytic Cracking                                          326,317
 Thermal Catalytic Cracking                                            522
 Flares (Blowdown System)                                         35,078
 Process Heaters: Oil                                               44,360
 Process Heaters: Gas                                             231,106
 Sulfur Recovery Units                                            202.125
 Total                                                            839,508
      The published TRENDS value is 830,000 tons. The discrepancy  may be due to independent
rounding or an incorrect assumption in  the recreation of the TRENDS  emission estimates.
                                            103

-------
2.3.6.4  NAPAP Activity



       NAPAP activity data are presented below for comparative purposes.  These data were not

priority elements for all sources in that inventory.

                                                    Value   Units
sec
Description
  1-02-007-01   External Combustion Boilers
               Industrial: Process Gas,
               Petroleum Refining Gas

  3-06-001-01   Petroleum Industry - Process
               heaters: Oil fired

  3-06-001-02   Petroleum Industry - Process
               heaters: Gas fired

  3-06-001-03   Petroleum Industry - Process
               heaters: Oil fired

  3-06-001-04   Petroleum Industry - Process
               heaters: Gas fired

  3-06-001-05   Petroleum Industry - Process
               heaters: Natural gas fired

  3-06-001-06   Petroleum Industry - Process
               heaters: Process gas fired

  3-06-002-01   Petroleum Industry - Fluid
               catalytic cracking unit

  3-06-003-01   Petroleum Industry - Thermal
               catalytic cracking unit

  3-06-004-01   Petroleum Industry -
               Slowdown systems: w/ vapor
               recovery sys. w/ flaring

  3-06-004-02   Petroleum Industry -
               Blowdown systems: w/o
               controls

  1-02-004-01   External Combustion
               Industrial: Residual oil, grade
               6 oil
                                                  17,345  106 ft3 gas burned



                                                  62,040  103 bbls oil burned


                                             116,387,121  103 ft3 gas burned


                                              11,747,943  103 gallons oil burned


                                              38,503,093  106 ft3 gas burned


                                                 311,504  106 ft3 gas burned


                                                 155,477  106 ft3 gas burned


                                               1,584,926  103 bbls fresh feed


                                                 171,954  103 bbls fresh feed


                                              11,603,678  103 bbls refinery feed



                                                 519,999  103 bbls refinery feed



                                               3,679,030  103 gallons oil burned
       Additional NAPAP production was added for two categories of process heaters burning oil

(3-06-001-01 and 03), and two categories of process heaters burning gas (3-06-001-02 and 04).
                                              104

-------
2.3.6.5  NAPAP Emissions
       The following emission estimates were reported through NAPAP for the categories of
petroleum refining emissions that are estimated in the TRENDS method:
 SCC             Description
 1-02-007-01      External Combustion Boilers  Industrial: Process
                  Gas, Petroleum Refining Gas
 3-01-032-01      Chemical Manufacturing   Elemental Sulfur
                  Production: Mod. Glaus: 2 stage w/o control (92-
                  95% removal) (SIC 2911  only)
 3-01-032-02      Chemical Manufacturing   Elemental Sulfur
                  Production: Mod. Glaus: 2 stage w/o control (95-
                  96% removal) (SIC 2911  only)
 3-01-032-03      Chemical Manufacturing - Elemental Sulfur
                  Production: Mod. Claus: 2 stage w/o control (96-
                  97% removal) (SIC 2911  only)
 3-01-032-04      Chemical Manufacturing - Elemental Sulfur
                  Production: Mod. Claus: 2 stage w/o control (99.9%
                  removal) (SIC 2911 only)
 3-06-001-01      Petroleum Industry - Process heaters: Oil fired
 3-06-001-02      Petroleum Industry - Process heaters: Gas fired
 3-06-001-03      Petroleum Industry - Process heaters: Oil fired
 3-06-001-04      Petroleum Industry - Process heaters: Gas fired
 3-06-001-05      Petroleum Industry - Process heaters: Natural gas
                  fired
 3-06-001-06      Petroleum Industry - Process heaters: Process gas
                  fired
 3-06-002-01      Petroleum Industry - Fluid catalytic cracking unit
 3-06-003-01      Petroleum Industry - Thermal catalytic cracking  unit
 3-06-004-01      Petroleum Industry - Blowdown systems: w/ vapor
                  recovery sys. w/ flaring
 3-06-009-02      Petroleum Industry - Hares: Residual oil
 3-06-009-03      Petroleum Industry - Hares: Natural gas
 3-06-009-04      Petroleum Industry - Hares: Process gas
 3-06-009-99      Petroleum Industry - Hares: Other not  classified
 1-02-004-01      External Combustion - Industrial: Residual  oil,
                  grade 6 oil (SIC 2911 only)
 Total
Tons SO. Emitted
          41.647

           7.175

          12,100

              92

           9,750

           4,762
           3,368
          75,174
          57,923
             549

          13,750

         204,647
           7.273
          15,671
          520.445
                                             105

-------
       Additional emissions are listed under Petroleum Refining in NAPAP for the following SCC
categories.
 SCC              Description
 3-06-001-07       Petroleum Industry - Process heaters: LPG fired
 3-06-001-99       Petroleum Industry  Process heaters: Other not
                   classified
 3-06-002-02       Petroleum Industry  Fluid Catalytic Cracking units:
                   Catalyst handling system
 3-06-004-02       Petroleum Industry  Blowdown systems: w/o
                   control
 3-06-005-03       Petroleum Industry - Fugitive emissions: Process
                   drains and wastewater separators
 3-06-005-04       Petroleum Industry - Fugitive emissions: Process
                   drains and wastewater separators
 3-06-006-03       Petroleum Industry - Vacuum distillate column
                   condenser
 3-06-008-01       Petroleum Industry - Fugitive emissions: Pipeline
                   valves and flanges
 3-06-008-05       Petroleum Industry - Fugitive emissions: Misc,
                   Sampling/ Non-Asphalt Blowing/ Purging/etc.
 3-06-010-01       Petroleum Industry - Sludge converter: general
 3-06-011-01       Petroleum Industry - Asphalt blowing: general
 3-06-012-01       Petroleum Industry - Fluid coking units: general
 3-06-014-01       Petroleum Industry - Petroleum coke calcining
 3-06-099-02       Petroleum Industry - Incinerators: Residual oil
 3-06-099-03       Petroleum Industry - Incinerators: Natural  gas
 3-06-099-04       Petroleum Industry - Incinerators: Process gas
 3-06-888          Petroleum Industry - Fugitive emissions: Not
                   classified
 3-06-999          Petroleum Industry  Miscellaneous: Not classified
 Total
Tons SO. Emitted
              60
            7,558

             327

              66

              20

            1,060

             110

             220

          12,091

             135
              91
          26,920
          12,560
            8,927
          16,183
            4,208
            1,739

          17.097
          109,372
       Finally, the distillate oil, residual oil, and natural gas combustion sections specifically
excluded petroleum refining emissions. The emissions excluded from these three combustion
categories add an additional 11,496 tons of SO2 to this section, bringing the total reported NAPAP
emissions for the petroleum industry to 641,313 tons of SO2.
                                              106

-------
2.3.6.6  Conclusion

       The petroleum refining emission estimates in TRENDS and NAPAP are quite different.  The
TRENDS method estimates emissions for six categories for a combined estimate of 830,000 tons of
SO2.  NAPAP estimates emissions for many more categories. For the six categories that correspond
to the TRENDS estimate, NAPAP estimates 520,445 tons of SO2.  NAPAP has a total estimate of
about 640,000 tons for petroleum refining.

       Fluid catalytic cracking dominates the TREND estimate with 326,317 tons. The  NAPAP
estimate is significantly lower, 204,647 tons of SO2.  It is unclear why the NAPAP emission estimate
is so  much lower, the reported NAPAP activity is actually higher than the TRENDS activity (1,585
versus 1,324 x 106 bbl/year fresh feed).

       Thermal catalytic cracking activity relies on an annual survey conducted by the Oil & Gas
Journal.25 The NAPAP emission estimate and the NAPAP activity are both an order of magnitude
higher than the TRENDS  values. The emissions are 7,273 tons versus 522 tons and the activity is 12
versus 17 x 106 bbl/year fresh feed.

       NAPAP reports significantly higher emissions for oil combustion at petroleum refineries.
NAPAP has an emission estimate of 117,512 tons versus the TRENDS estimate of 44,360 tons of
SO2.  The emission factor for SCC 1-02-004-01 is listed in TRENDS for process heaters - oil.  This
SCC is for grade 6 residual oil burned in external combustion boilers - industrial.  The petroleum
refining section also has a general SCC for oil-fired process heaters (3-06-001-03) with the same
emission factor and has added an SCC (3-06-001-11) for large grade 6 oil-fired process  heaters
(MOO MMBTU).  (This new SCC has a slightly higher emission factor, 159.3S lbs/103 gallons
burned versus 158.6S lbs/103 gallons burned).  The TRENDS method  should be rewritten to utilize
these  SCC codes.

      The sulfur content of the  oil burned and consequently the emission factor used by TRENDS
for oil-fired process heaters appears to be too low.  The majority of the oil reported burned is
actually crude oil (94 percent) and the distillate oil (5.4 percent) is far more significant than the

                                            107

-------
residual oil (0.6 percent). Therefore, the use of a residual oil emission factor is not very accurate.
There is no emission factor for combustion of crude oil in process heaters at a petroleum refinery.
Emission factors used in the industrial oil combustion section are 42.3 lbs/103 gallon burned for
distillate oil and 258.5 lbs/103 gallon burned for residual oil.  Using the residual oil emission factor,
which appears to be the intent of the TRENDS procedure manual, would result in process heater oil
emissions of:

       258.5 * 595,056 / 2000 = 76.911  tons of SO;.

       The emission estimates for gas-fired process heaters are also very different in TRENDS and
NAPAP.  The NAPAP estimate is 117,237 tons and the TRENDS estimate is 231,106 tons of S0:.
For gas-fired process heater emissions, the AIRS Facility Subsystem  Source Classification Codes and
Emission Factor Listing for Criteria Air Pollutants1 lists the emission factors as 0.6 lbs/106 ft3 for
natural-gas fired versus 950.0S lbs/106 ft3 for process-gas fired.  The emission factor of 253.1 lbs/106
ft3 used in TRENDS is between the  emission factors for the two  fuels.  In TRENDS, the quantity of
process gas burned as a fuel is three times as high as the amount of natural gas burned. It is unclear
why the two fuels are added together in the activity section. An emission factor weighted by
quantity of fuel burned would be 261.2 lbs/106 ft3, which would result in an emission estimate of
238,547 tons of SO2.  It is unclear if the  natural gas emissions bear  calculating since they  only
contribute approximately 146 tons of SO2 to the total estimate.

       Finally, the emission estimates  for sulfur recovery at petroleum  refineries is also very
different.  The NAPAP inventory reports emissions of 29,117 tons of SO: and the TRENDS estimate
is 202,125 tons of SO2.  Errors were discovered in the execution of the TRENDS method  (as
discussed  in Section 2.3.3).  The TRENDS emission estimate should be 172,696 tons of SO,,
however, the numbers are still very dissimilar.  Research into the types of sulfur recovery units
employed  at petroleum refineries (and  their emissions) is warranted.

       The data required to determine thermal catalytic  cracking versus the fluid catalytic cracking
are no longer available.  The thermal catalytic cracking  contributes 0.16 percent of the cracking
emissions  in the TRENDS estimate but contributes 3.4 percent in the NAPAP inventory.  Sicnificant

                                             108

-------
effort to identify a replacement source of data for thermal cracking may not be warranted, however,
additional research into why the NAPAP activity rates for thermal catalytic cracking are so different
from TRENDS may be warranted. Effort to determine why the fluid catalytic cracking emission
estimate is so much  lower in NAPAP is definitely warranted.

2.3.7 Natural Gas  Production

       The TRENDS procedure has two components of emissions from the production of natural gas,
combustion and sulfur recovery.  The combustion of natural gas during production contributes a
small amount of emissions (441 tons of SO2) as was reported in Section 2.1.5.  Emissions from
sulfur recovery are far more significant (163,000 tons) as was reported in Section 2.3.3.  This section
presents the results of the combustion and the sulfur recovery emission estimates that TRENDS
publishes  as Natural Gas Production. There is no section in the TRENDS procedure manual
specifically for natural gas production, rather the estimate originates in the two sections for natural
gas combustion and  sulfur recovery.

       The NAPAP estimate is 7,660 tons for combustion of natural gas during production and
59,500 tons of SO2 from sulfur recovery units at natural gas production facilities.  In addition,
NAPAP reports 265,000 tons of SO2 from other emission sources in natural gas production, resulting
in a  total estimate of 332,000 tons of SO2.  The difference between the two estimates is  172,000 tons
of S0: (52 percent).

2.3.7.1 TRENDS Activity

       The natural gas consumption for gas pipelines and plants is the sum of pipelines  fuel and
lease and plant fuel.  These values are obtained from Table 13 "Consumption of Natural Gas" of
Natural Gas Annual, 1985."  The value for 1985 for lease and plant fuel is 966,047 x 106 ft3 and the
                                             109

-------
1985 value for pipelines fuel is 503,766 x  106 ft3.  Therefore the total gas pipelines and plants natural
gas combustion rate is:
       966,047 + 503,766 = 1,469,813 x 10b ft3.

       The quantity of sulfur recovered by natural gas plants is reported in Table 4 "Recovered
Sulfur Produced and Shipped in the United States" of Minerals Yearbook 1989 "Sulfur.'*  The
production from natural gas plants  was 2,373,000 metric tons (2,610,000 short tons).  The TRENDS
activity spreadsheet erroneously left these activity rates in metric units.

2.3.7.2 TRENDS Emission Factor

       The TRENDS emissions factor is 0.6 Ib SCyiO6 ft3 burned for natural gas combustion in the
production of natural gas. This emission factor is consistent with all of the natural gas combustion
emission factors listed in AIRS Facility Subsystem Source Classification Codes and Emission Factor
Listing for Criteria Air Pollutants.1

       According to the TRENDS  procedure manual, the emission factor for sulfur recovery at
natural gas plants is derived annually from the emissions and throughput reported to AIRS/FS  in
SCC 3-01-032-01 through 3-01-032-04. In  1985 the throughput values were reported as 2,388,968
tons of sulfur.  In 1985 the emissions were reported as  127,534 tons of SO2.  The emission factor is
obtained by dividing the total emissions by the sum of the operating rates.  The emission factor for
1985 should be:

       127,534 / 2,388,968 * 2000 = 106.8  Ibs/ton of sulfur.

       The emission factor reported through  the TRENDS  spreadsheets  for both 1990 and 1991 is
137.5  Ib/ton of sulfur produced.
                                             110

-------
2.3.7.3  TRENDS Emissions

       TRENDS combustion emissions for natural gas pipelines are calculated as:

       0.6 * 1,469,813 / 2000 = 441 tons of SO;.

       The TRENDS emission estimates are calculated for sulfur recovery during natural gas
production using the emission factor of 137.5 Ib/ton of sulfur produced and the production rates in
the TRENDS activity spreadsheet. The emissions for natural gas fields are:

       2,373,000 *  137.5 / 2000 = 163,143 tons of S0:.

       Using the emission factor of  106.8 Ibs/ton of sulfur produced derived by following the
TRENDS procedure manual and the  correct activity data would result  in the following emission
estimates for natural gas fields.

       2,610,000 *  106.8 / 2000 = 139,374 tons of S02

2.3.7.4  NAPAP Activity

       The following activity data for natural gas production were reported through the NAPAP
inventory.
 SCC             Description                                            106 ft3 Burned
 3-10-004-04      Oil and Gas Production - Process Heaters: Natural              12,187
                  gas
 3-10-004-14      Oil and Gas Production - Steam generators:  Natural               2.852
                  gas
 Total                                                                          15,039
      The sulfur recovered at natural gas plants, as reported in  the NAPAP inventory,  has not been
determined.  The following activity data are for the other categories of emissions included in the
NAPAP inventory.
                                             Ill

-------
 SCC             Description                                              106 ft3 Gas
                                                                            Produced
 3-10-002-01       Natural Gas Production: Gas sweeting: Amine                5,061,719
                  Process
 3-10-002-02       Natural Gas Production: Gas stripping operations               747,129
 3-10-002-03       Natural Gas Production: Compressors                          199,002
 3-10-002-04       Natural Gas Production: Wells                                254.156
 3-10-002-05       Natural Gas Production: Hares                                 64.826
 3-10-002-06       Natural Gas Production: Gas Lift                               25.915
 3-10-002-99       Natural Gas Production: Other not classified                  2,099.517

2.3.7.5 NAPAP Emissions
       Emissions are listed for combustion of natural gas during natural gas production, sulfur
recovery, and other NAPAP natural gas production emission sources.  The emissions at the 8 digit
SCC level are as follows.

 SCC             Description                                       Tons SO, Emitted
 3-01-032-01      Chemical Manufacturing  Elemental Sulfur                     14,311
                  Production: Mod. Claus: 2 stage w/o control (92-
                  95% removal) (SIC 1311, and 1321  only)
 3-01-032-02      Chemical Manufacturing - Elemental Sulfur                     12,194
                  Production: Mod. Claus: 2 stage w/o control (95-
                  96% removal) (SIC 1300, 1311, and 4922 only)
 3-01-032-03      Chemical Manufacturing - Elemental Sulfur                      3.299
                  Production: Mod. Claus: 2 stage w/o control (96-
                  97% removal) (SIC 1311, and 1321  only)
 3-01-032-04      Chemical Manufacturing - Elemental Sulfur                     29,694
                  Production: Mod. Claus: 2 stage w/o control (99.9%
                  removal) (SIC 1311, and 1321 only)
 3-10-002-01      Natural Gas Production: Gas sweeting: Amine                  190,241
                  Process
 3-10-002-02      Natural Gas Production: Gas stripping operations                  1.427
 3-10-002-03      Natural Gas Production: Compressors                              31
 3-10-002-04      Natural Gas Production: Wells                                     17
 3-10-002-05      Natural Gas Production: Hares                                 63,055
 3-10-002-06      Natural Gas Production: Gas Lift                                    0
 3-10-002-99      Natural Gas Production: Other not classified                     10,140
 3-10-004-04      Oil and Gas Production  Process heaters: Natural                   442
                  gas

                                            112

-------
 SCC             Description                                       Tons SO. Emitted
 3-10-004-14      Oil and Gas Production - Steam generators: Natural              	18.
                  gas
 Total                                                                      324,869

       An additional 7,200 tons  of SO: were excluded from the natural gas combustion - boilers
estimate because the emissions were reported with an SIC for oil  and natural gas production.  This
brings the total NAPAP natural gas production emission estimate  to 332,069 tons of SO:.

2.3.7.6  Conclusion

       The NAPAP and TRENDS estimates for this category are  very different. The TRENDS
estimate is made up of two numbers, emissions from combustion  of natural gas during  natural gas
production and emissions from sulfur recovery units at natural gas plants.

       Both NAPAP and TRENDS have small combustion estimates (460 versus 7,660 tons of SO2).
The estimates for sulfur recovery are very different (163,143 versus 59,498 tons of SO2) and no
explanation for the difference is  available.  As stated earlier, errors were discovered in  the TRENDS
estimate and the  emissions from sulfur recovery should be  139,374 according to the TRENDS
procedure manual.  It is unclear  why these two values are so different.  Research into the type  of
sulfur recovery units that are utilized in natural gas production should be  conducted. Finally,
NAPAP reports an  additional 264,911 tons of SO, from standard  natural gas production processes.  It
is unclear why these processes are not accounted for in the TRENDS method.

2.3.8  Iron and Steel

       The 1985 TRENDS emission value is 360,000 tons SO2.  The 1985 NAPAP emission value is
212,000 for iron  and steel manufacturing (including byproduct and beehive coke manufacturing).
The discrepancy  between the two inventories is 148,000 tons of S02 (70 percent).
                                            113

-------
2.3.8.1  TRENDS Activity

       In TRENDS, there are four categories of SO2 emissions for the iron and steel industry: coke,
sintering, open hearth, and roll and finish. Activity for coke is obtained from Survey of Current
Business21 and activity for sintering, open hearth, and roll and  finish are obtained from the Minerals
Yearbook 1989 "Iron and Steel."4 All of the activity values derived following the TRENDS
procedure manual matched the TRENDS activity spreadsheet.

Coke
       Activity for Coke emissions is the total beehive and oven (byproduct) production figure. The
value is obtained from the U.S. Department of Commerce, Economics and Statistics Administration,
Bureau of Economic Analysis, Survey of Current Business, "Petroleum, Coal, and Products"21
Beehive  and Oven Coke (Byproduct) production.  For  1985 the value is 28,651,000 tons. Coke
production  is also available through Table A5 "Coke and Breeze Production at  Coke Plants" of
Quarterly Coal Report1 and  the 1985 value was 28,651,000 tons. This is consistent with the value
of 28,700,000 tons b'sted in the TRENDS activity  spreadsheet.

Sintering
       Activity for Sintering  is total pig iron production and is obtained from Table 3 "Materials
Consumed in Blast Furnaces  and Pig Iron Produced" of Minerals Yearbook 1989 "Iron and Steel"* or
from the  U.S. Department of Commerce, Economics and Statistics Administration, Bureau of
Economic Analysis, Survey of Current Business, "Metals and Manufactures" Pig iron production.28
The numbers from these two  references  are close (49,963,000 versus 50,446,000 tons).  The Minerals
Yearbook 1989 value was found in the Iron and Steel chapter rather than in the Iron Ore chapter.
The value used in TRENDS was from the Survey of Current Business.  The TRENDS method
requires dividing the total into three equal portions among the  three sintering components (windbox,
discharge, and sinter-fugitive), resulting  in a value of 16,815,000 tons for sintering. This is
consistent with the value of 16,800,000 tons listed in the TRENDS activity spreadsheet.
                                             114

-------
Open hearth
       TRENDS uses a complex process for determining the fraction of steel production in open
hearth furnaces (as opposed to an oxygen furnace or an electric arc furnace).  Open hearth is
obtained by multiplying the fraction of scrap and pig iron consumed in steel production (by type of
steelmaking furnace) by the total U.S. raw steel production value.

       The data to determine the fraction that is open hearth is obtained from Table 6  "U.S.
Consumption of Scrap and Pig Iron in Steel Production, by Type of Steelmaking Furnace"  of
Mineral Yearbook  1989 "Iron and Steel."*  The values for basic open hearth scrap and iron were
summed and were  7,148,000 short tons.  The values for basic oxygen scrap and iron were summed
and were 59,854,000 short tons.  The values for basic electric arc scrap and iron were  summed and
were 32,755,000 short tons. The total for the three furnace types, basic oxygen, basic  open hearth,
and basic electric arc, was 99,757,000 short tons. Therefore, the fraction that is basic open hearth is
7.2 percent.

       The TRENDS  procedure is to obtain total raw steel production from the U.S.  Department of
Commerce, Economics and Statistics Administration, Bureau of Economic  Analysis, Survey of
Current Business, "Metals and Manufactures'™ raw steel production. Total raw steel production is
also available from Table 1 "Salient Iron and Steel Statistics" of Minerals Yearbook 1989 "Iron and
Steel" and the value is 88,259,000 short tons in both references.4

       The open hearth activity is then calculated in TRENDS as:

       0.07 * 88,259,000 = 6,178,130 tons.

This is consistent with the TRENDS spreadsheet activity value  of 6,200,000 tons.

       The fraction of production that is basic open hearth is also available in Table 5 "U.S. Steel
Production, By type of Furnace Process" of Minerals Yearbook 1989 "Iron and Steel."* The value
for basic open hearth in 1985 was 7.3 percent.  Table 5 also reports that basic open hearth
                                             115

-------
production in 1985 was 6,428,000 short tons.  Open hearth production is decreasing and only

represents 4.5 percent of 1989 production.



Roll and finish

       The activity value for roll and finish operations is total raw steel production. That value was

88,259,000 short tons for 1985 and is consistent with the TRENDS spreadsheet.



2.3.8.2  TRENDS Emission Factors



       The emission  factors are also divided into the four categories.



Coke

       The emission  factors for coking cited in the TRENDS method are as follows.


  SCC             Description                    Emission Factor  Units

  3-03-003-02      Primary metal production -                 0.02   Ibs/ton of coal
                  By-Product Coke                                charged
                  Manufacturing:  Oven charging

  3-03-003-03      Primary metal production -                 3.3    Ibs/ton of coal
                  By-Product Coke                                charged
                  Manufacturing:  Oven pushing

  3-03-003-04      Primary metal production -                 0.4    Ibs/ton of coal
                  By-Product Coke                                charged
                  Manufacturing:  Quenching

  3-03-003-06      Primary metal production -                 4.0    Ibs/ton of coal
                  By-Product Coke                                charged
                  Manufacturing:  Oven
                  underfiring

  3-03-003-08      Primary metal production -                 0.1    Ibs/ton of coal
                  By-Product Coke                                charged
                  Manufacturing:  Oven/door
                  leaks

  3-03-003-14      Primary metal production -                 Q. l    Ibs/ton of coal
                  By-Product Coke                                charged
                  Manufacturing: Topside leaks

 Total                                                      7.92   Ibs/ton of coal
                                                                  charged


       These emission factors match the AIRS Facility Subsystem Source Classification Codes and

Emission Factor Listing for Criteria  Air Pollutants1 document.  There is a  0.0 emission factor for


                                             116

-------
beehive production-general.  An overall emission factor is obtained by adding these factors and
dividing by 0.7 to account for 0.7 tons of coke produced per ton coal consumed.  This percentage
coal to coke is substantiated in Table 2 "Statistical Summary of the Coke Industry in the United
States" of Coke and Coal Chemicals in J980.21 The emission factor for coking is:

       7.92 / 0.7 = 11.3 Ibs/ton coke produced.

Sintering
       The emission factor for sintering was obtained by  dividing  the 1980 NEDS emissions by the
 1980 production.  The emission factor is 2.5 Ibs/ton produced.

Open hearth
       The emission factor for open hearth was obtained by dividing the 1980 NEDS emissions by
the 1980 production. The emission factor is 1.5 Ibs/ton produced.

Roll and finish
       The emission factor for roll and finish is backcalculated from process  equipment combustion
emissions burning coke oven gas and residual  oil.  The coke oven  gas discussion was originally
presented in Section 2.1.1.2. The residual oil discussion was originally presented  in Section 2.2.1.
The derivation of the coke oven gas component and the residual oil component are repeated below.

Coke oven gas
       The TRENDS procedure is to obtain coke oven gas production from Quarterly Coal Report"
however, coke oven gas production is not provided in that report.  Table 23 "Coal Consumption by
End-Use Sector" of Quarterly Coal Report provides coal consumption by coke plants in 1985 of
41,056,000 short tons.  Figure 15 "Production  of Coke and Coal Chemicals"  of Coal Data: A
Reference1* indicates  that 11,000 ft3 of coke oven gas  are produced per ton of metallurgical (coking)
coal.  This  results in  1985 coke oven gas production of:

      41,056,000 *  11,000 / 1,000,000 = 451,616 x 106 ft3.
                                             117

-------
       The TRENDS method assumes that 40 percent of the coke oven gas is consumed in the iron
and steel industry. Therefore the coke oven  gas consumption is:
       451,616 * 0.40 = 180,646 x  106 ftj.

       The SO: emission factor for  coke oven gas is 1,091 lbs/106 ft1 as listed in AIRS.7  Therefore,
coke oven gas emissions are:

       180,646 * 1,091  / 2000 = 98,542 tons of SO:.

Residual oil
       The quantity of residual oil consumed by steels mills was calculated by taking the quantity in
tons of raw steel produced in 1985 and multiplying by a conversion factor for the value of residual
oil consumed per ton  of raw steel produced. The conversion factor used is 7.38 gal/ton raw steel.
The  quantity of raw steel produced is obtained from Table 1 "Salient Iron  and Steel Statistics" of
Minerals Yearbook 1986 "Iron and Steel".  The value for 1985 is 88,259,000 short tons.  The
quantity of residual oil consumed by steel mills is calculated as:

       88,259,000 * 7.38 gal/ton raw steel = 651,350,000 gallons.

       The emission factor for residual oil combustion is cited in the TRENDS procedure manual as
1,595 lbs/103 gal.  The emission factor  1,595 is probably a typographical error and  should read 159S
lb/103 gallon burned.  The TRENDS procedure manual also states that the emission factor used in the
residual oil combustion  section  should be used. The emission factor cited in Section 2.1.1.2  of this
report is  158.6S lb/103 gallon burned.

       The 158.6S lb/103 gallon burned emission factor matches the AIRS Facility  Subsystem Source
Classification Codes and Emission Factor Listing for Criteria Air Pollutants^ document.  The
average sulfur content of grade 6 fuel oil is obtained from Heating Oils, J9856   Again, this reference
provides  averages for  each of five regions.  The average sulfur contents range from 1.20  to 1.75
                                              118

-------
percent in a total of 44 samples. The average national figure based on the number of samples is 1.63
percent.  The emission factor for residual oil is:

       158.6 * 1.63 = 258.5 lb/103 gallons burned.

Residual oil emissions at steel mills are then calculated as:

       651,350 * 258.5 / 2000 = 84,187 tons of SO:.

Roll and finish emission factor
       The emissions from residual oil combustion  are added to coke oven gas emissions and the
quantity of emissions from open hearth  furnaces are subtracted from the total to yield:

       98,542 + 84,187 - 4,650 = 178,079 tons of SO2.

The emission factor for roll and finish operations is then calculated as:

       EF = 178,079 / 88,259,000 * 2,000 = 4.04 Ibs/ton of raw steel.

       The overall roll and finish emission factor used in the TRENDS spreadsheets is 3.8 Ibs/ton
raw steel produced for both 1990 and 1991.

2.3.8.3  TRENDS Emissions

       TRENDS emissions are calculated using the activity data and emission factors from the
TRENDS spreadsheet for the  four categories as follows.

Coking
       28,700,000  *  11.3 / 2000 = 162,000 tons SO,
                                             119

-------
Open hearth

      6,200,000 *  1.5 / 2000 = 4,650 tons of SO2



Sintering

       16,800,000 * 2.5 / 2000 = 21,000 tons of SO:



Roll and finish

       88,259,000 * 3.8 / 2000 = 168,000 tons of SO:



Total TRENDS

       162,000 + 4,650 + 21,000 + 168,000 = 355,650 tons of SO2



2.3.8.4  NAPAP Activity



       The NAPAP production for the four processes is broken out below. The activity data were

not high priority  elements for all sources in that inventory and are presented for comparison only.
Coking


 SCC             Description                                           Tons of Coal
                                                                           Charged

 3-03-003-02      Primary Metal Production - By-Product Coke               35,832,171
                  Manufacturing: Oven charging

 3-03-003-03      Primary Metal Production - By-Product Coke               27,288,497
                  Manufacturing:  Oven pushing

 3-03-003-04      Primary Metal Production - By-Product Coke               25,181,912
                  Manufacturing: Quenching

 3-03-003-06      Primary Metal Production - By-Product Coke               20,155,913
                  Manufacturing: Oven underfiring

 3-03-003-08      Primary Metal Production   By-Product Coke               21,489,908
                  Manufacturing: Oven/door leaks

 3-03-003-14      Primary Metal Production - By-Product Coke                7,764,158
                  Manufacturing: Topside leaks

 3-03-004-01      Primary Metal Production   Coke Manufacturing:              405,008
                  Beehive Process, general
                                            120

-------
Open hearth
 SCC             Description

 3-03-009-01      Primary Metal Production - Steel Production: Open
                  hearth furnace, stack

 3-03-009-18      Primary Metal Production - Steel Production:
                  Charging, open hearth

 3-03-009-19      Primary Metal Production - Steel Production:
                  Tapping, open hearth
                                                    Tons Produced

                                                        5,746,973


                                                           62,002


                                                          122,002
Sintering


 SCC


 3-03-008-01


 3-03-008-11


 3-03-008-13


 3-03-008-17


Roll and finish


 SCC

 3-03-009-31


 3-03-009-33


 3-03-009-34


 3-03-009-35
Description


Primary Metal Production - Iron Production - Blast
Furnaces: Ore charging

Primary Metal Production - Iron Production -
Sintering: Raw Mat. st'kpiles, coke breeze,
limestone, ore fines

Primary Metal Production - Iron Production -
Sintering: Windbox

Primary Metal Production - Iron Production -
Sintering: Cooler
Description

Primary Metal Production - Steel Production: Hot
rolling

Primary Metal Production - Steel Production:
Reheat furnaces

Primary Metal Production - Steel Production: Heat
treating furnaces: Annealing

Primary Metal Production - Steel Production: Cold
rolling
    Tons Iron
     Produced

    16,178.407


    38,736,563



    19,693,102


    7,139,166
Tons Produced

     5,510,147


    15,101,751


     4,972.078


      990.995
2.3.8.5 NAPAP Emissions
      The NAPAP emissions for the four processes are:
                                            121

-------
Coking

 sec

 3-03-003-02

 3-03-003-03

 3-03-003-04


 3-03-003-06

 3-03-003-08

 3-03-003-14

 Total
                  Description

                  Primary Metal Production - By-Product Coke
                  Manufacturing: Oven charging

                  Primary Metal Production  By-Product Coke
                  Manufacturing: Oven pushing

                  Primary Metal Production  By-Product Coke
                  Manufacturing: Quenching

                  Primary Metal Production  By-Product Coke
                  Manufacturing: Oven underfiring

                  Primary Metal Production  By-Product Coke
                  Manufacturing: Oven/door leaks

                  Primary Metal Production  By-Product Coke
                  Manufacturing: Topside leaks
Tons SO. Emitted

            7,305


           13,494


            3,307


           40.816


             329


             116
                                                                              65,367

       Additional coke manufacturing sources include: coke crushing/screening and handling (41

tons); coal preheater (3,243 tons); gas by-product plant (2,142 tons); coal storage pile (175 tons);

other/not classified (2,062 tons); and  beehive process general (1,599 tons) for a total additional

emissions of 9,262 tons of SO2.  This brings the total coke manufacture emissions to 74,629

tons SO,.
Open hearth

 sec
 3-03-009-01

 3-03-009-18

 3-03-009-19

 Total

Sintering


 sec
 3-03-008-11
                  Description

                  Primary Metal Production - Steel Production: Open
                  hearth furnace, stack

                  Primary Metal Production - Steel Production:
                  Charging, open hearth

                  Primary Metal Production - Steel Production:
                  Tapping, open hearth
                  Description

                  Primary Metal Production - Iron Production -
                  Sintering: Raw Mat. st'kpiles, coke breeze,
                  limestone, ore fines
Tons SO. Emitted

            1,159


               5


            	5


            1,169
Tons SO. Emitted

             470
                                            122

-------
 SCC             Description                                         Tons SO. Emitted
 3-03-008-13       Primary Metal Production - Iron Production                     33,756
                   Sintering: Windbox
 3-03-008-17       Primary Metal Production - Iron Production -                       280
                   Sintering: Cooler
 Total                                                                          34.506

Roll and finish

 SCC             Description                                         Tons SO. Emitted
 3-03-009-31       Primary Metal Production   Steel Production: Hot                 2,162
                   rolling
 3-03-009-33       Primary Metal Production - Steel Production:                     6,484
                   Reheat furnaces
 3-03-009-34       Primary Metal Production - Steel Production: Heat                  390
                   treating furnaces: Annealing
 3-03-009-35       Primary Metal Production - Steel Production: Cold                  201
                   rolling
 3-04-003-XX      Primary Metal Production - Gray Iron Foundries
 3-04-007-XX      Primary Metal Production - Steel Foundries
 Total

       Additional iron production sources include: ore charging (552 tons); blast heating  stoves
(2,785 tons); cast house (5,374 tons); other not classified (630 tons,) for total additional emissions of
9,341 tons of SO2.
       Additional steel production sources include: electric arc furnace (alloy steel) (1,749 tons):
electric arc furnace (carbon steel) (281 tons); soaking pits (9,353 tons); basic oxygen furnace (EOF) -
open hood stack (23  tons); charging and tapping EOF (47 tons); teeming (unleaded steel) and
continuous casting (533 tons);  and other not classified (4,316 tons) for total additional emissions of
16,302 tons of SO:.

       Finally, several of the combustion categories specifically exclude iron and steel facilities
during the development of their emission estimates.  The residual oil section excluded  14,318 tons of
SO,.  The natural gas - boilers section excluded 10,044 tons of S02. The coke and coke oven gas
portions of the miscellaneous fuels section specifically excluded iron and steel facilities.  The coke
                                              123

-------
and coke oven gas emissions are 747 and 11,252 tons of SO2 respectively.  In addition, 5,981 tons
from blast furnace gas combustion are added to the total iron and steel estimate.

       This brings the total NAPAP emission estimate for iron and steel production to 204,000  tons
of SO,.

2.3.8.6  Conclusion

       All four of the iron and steel categories have significantly different emission estimates in
TRENDS and NAPAP. The four categories are discussed separately.

       For coking emissions, TRENDS lists the six SCC categories that are used in their estimate.
Using these six SCC categories provides emission estimates of 65,367 tons (NAPAP)  versus 162,000
tons (TRENDS). Even if all of the NAPAP coke emissions are counted, the total NAPAP estimate
for coke is only 74,629 tons of SO:.  One possible cause for over estimation in the TRENDS
procedure is the inclusion of beehive process coke manufacturing in the activity number. It is not
clear whether the beehive process is still used and SO2 emissions may be  much lower. (There is also
a 0.0 as an emission factor in the SCC book.)  Another possible  disconnect is the application of SO,
control technologies in the NAPAP coke production.  Finally, the NAPAP inventory may have
emissions from coke production reported in other iron and steel processing steps.

       The TRENDS emission factor for sintering (2.5 Ibs/ton steel produced) was derived from
1980 NEDS  statistics and is therefore  quite dated. If the 1985 emissions  for sintering (SCC 3-03-
008-11 through 20)  were divided by one-third of the 1985 production of pig iron (from U.S. Bureau
of Mines) the emission factor would be significantly higher.  Using the production value for pig iron
would yield the following emission factor:

       (34,506 tons SO2 * 2,000)7(16,800,000 tons sintered pig iron) = 4.11 Ibs SO2/ton pig iron

       The NAPAP numbers appear to underestimate the open hearth emissions and the TRENDS
method  may overestimate the open hearth emissions.  Open hearth is a process that is declining.
                                             124

-------
Open hearth emissions are 1,169 tons in NAPAP and 4,650 tons in TRENDS.  Using the 1985
NAPAP open hearth emission estimate would change the TRENDS emission factor from 1.5 to 0.38
Ibs/ton produced.  The difference could also be due to the application of SO2 control devices in
NAPAP.

       The 1985 emission estimates underwent more  review than the 1980 estimates and therefore
the 1985 values are more suitable for use to calculate sintering and open hearth emissions in the
TRENDS method.  Revising the emission factor based on the 1985 emission estimate would make
the TRENDS and NAPAP emission estimates equivalent for these two categories of emissions.

       The TRENDS roll and finish estimate of 168,000 tons of SO, is substantially higher than the
NAPAP estimate of 25,304 tons of SO:. The TRENDS category may be misnamed as it is  really a
sum of emissions from combustion of coke oven gas  and residual oil.  A long and complex
procedure is put forth to calculate the roll and finish emission factor. This procedure, if followed,
did not change the emission factor used from 1985 to 1990 and 1991. Therefore, it is assumed  that
the factor has not been recently updated.

       The TRENDS procedure assumes that 40 percent of the coke oven gas is burned in iron  and
steel facilities. As discussed in the miscellaneous fuels section, this estimate may be too low.  Table
12 "Production and Disposal of Coke Oven Gas in the United States by Producing State: 1980" of
Coke and Coal Chemicals in 1980 reports that in  1980 coke gas use was 39 percent used by
producers in heating ovens, 58 percent was for other use by producers,  1.4 percent commercial sales,
and 1.5 percent was wasted.20 The NAPAP activity data are consistent with these statistics  indicating
that 99 percent of coke oven gas burned was at  iron and steel facilities.  In addition, TRENDS uses a
coke oven gas average sulfur value of 1.605 percent.  The NAPAP inventory lists an average  sulfur
content for coke oven gas of 0.5 percent.    Using  the lower sulfur content would result in an
emission  factor of 340 lb/106 ft3 burned. Assuming 98.6 percent of the coke oven gas is consumed
by the  iron and steel industry and that the sulfur content of coke oven gas is much  lower, would
result in 75,700 tons of SO2 from the coke  oven gas portion of the roll  and finish estimate.
Following the procedure, a roll and finish emission factor of 4.04 Ibs/ton was developed versus 3.8
Ibs/ton of raw steel in the TRENDS spreadsheet.  Both of these emission factors are too high.

                                             125

-------
       The reason the TRENDS roll and finish emission factor was calculated so high is based on
the ratio of residual oil used to produce steel.  The TRENDS procedures manual uses a factor of
0.00738 x 106 gal/103 ton of raw steel produced (7.38 gal/ton of steel produced).  This factor is too
high. Based on the data in Table 3 "Total Inputs of Energy for Heat, Power, and Electricity
Generation by Census Region, Industry  Group and Selected Industries, 1985" of Manufacturing
Energy Consumption Survey: Consumption of Energy, 1985* the total residual oil used in blast
furnaces and steel mills was 5,458,000 barrels in  1985. This corresponds to a new factor of:

       5,458,000  * 42 / 88,259,000 = 2.597 gal/ton of steel.

Using the new ratio or residual oil burned/ton of raw steel produced results in emissions of 11.248
tons of SO2 from  the residual oil portion of the roll and finish estimate.

       It is  unclear why the Survey of Current Business is introduced as  a reference in the TRENDS
procedure.  The coke production is available through the Quarterly Coal  Report, the pig iron and raw
steel production data are available through the Minerals Yearbook "Iron and Steel."

       Finally, the TRENDS procedure  does not account for any  SO2 controls.  This is probably the
most significant difference between the  TRENDS and NAPAP estimates.

2.3.9 Cement Manufacturing

       The 1985 TRENDS  emission value is 620,000 tons S(X The  1985 NAPAP value is 290.700
tons for cement manufacturing.  The discrepancy between the two inventories is 329,300  tons of  SO.
(113 percent). The discrepancy  is due to the difference in the way the fuel emissions are reported.

2.3.9.1  TRENDS Activity

       In TRENDS, the cement  production is obtained from Table 1 "Salient Cement Statistics" of
Minerals Yearbook 1989 "Cement.'*  The value for 1985 was 77,895,000 short tons.
                                             126

-------
       The activity section of the TRENDS procedure manual does not direct the user to obtain fuel
consumption values, however, the emission factor section implies that the user will need fuel
consumed in cement manufacture.  Therefore, fuel consumed in cement manufacture is obtained from
Table 8 "Clinker produced and fuel consumed by the Portland cement industry in the United States
by process" of Minerals Yearbook, 1986 "Cement."4  The 1985 values are as follows.
 Description                                                 Value
 Coal consumed (103 tons)                                   11,606
 Oil consumed (103 barrels)                                     755
       The oil combustion section of the TRENDS procedure document states that two thirds of the
oil consumed by the cement industry is residual oil and the remaining one third is distillate oil
(Sections 2.1.1.1 and  2.1.1.2).

2.3.9.2  TRENDS Emission Factors

       The emission factors cited in the TRENDS method account for the sulfur in the mineral
source, the sulfur in coal,  the sulfur in residual oil, and the sulfur in distillate oil. TRENDS cites
emission factors that are inconsistent with  the AIRS Facility Subsystem Source Classification Codes
and Emission Factor Listing for Criteria Air Pollutants.1  The emission factors cited in the TRENDS
procedure are as follows.
     Description       Emission Factor            Units
 Mineral source                    10.2   Ibs/ton cement produced
 Coal                             30.45   Ibs/ton coal burned
 Residual Oil                     124.5   lbs/103 gal burned
 Distillate Oil                    112.35   lbs/103 gal burned
      The emission factor for cement manufacturing kilns in both  the wet and dry process matches
the AIRS Facility Subsystem Source Classification Codes and Emission Factor Listing for Criteria
Air Pollutants.1  The procedures  manual states that the emission factors for coal  and residual oil
utilize the sulfur value derived in the industrial boilers section.   The average sulfur content for
bituminous coal  was calculated as 1.4 percent (Section 2.1.4). The average sulfur content cited in
                                              127

-------
the Residual Oil (Section 2.1.1.2) is 1.63 percent.  The emission factor for distillate oil utilizes an
emission factor of 0.3 percent sulfur.

       The TRENDS  method states that the uncontrolled emission factor is determined by adding the
emissions and dividing by the total cement production rate.  The TRENDS method does not cite
where the fuel consumption values  are obtained, however, as stated above, they are available from
Table 8 "Clinker produced and fuel consumed by the Portland  cement industry in the United  States
by process" of Minerals Yearbook,  1986 "Cement."4  The uncontrolled emission factor would then be
calculated as  follows (assuming two-thirds of the oil consumed is residual oil and the remaining one
third is distillate oil).

       F10.2  * 77.895 + 30.45 * 11.606 + (125.5 * 0.666 +  112.35 * 0.333)  * 31.7101
                                 77,895
       = 64.04 Ibs SCX/ton of cement produced

       The TRENDS  manual then directs the user to estimate the control efficiency using linear
interpolation  and the particulate control efficiency. The TRENDS procedure manual states that the
baseline value of 13.75 percent SO2 control corresponds to a 99 percent particulate  control and  12
percent SO2 control corresponds to  92 percent particulate control. For this analysis the full 13.75
percent SO2 control will be assumed. This would result in an overall TRENDS  emission factor of:

       64.04  * (1 -  0.1375) = 55.23 tons of SO^ton of cement produced.

       The overall emission factor for cement manufacturing listed in the TRENDS spreadsheets is
15.95 Ibs/ton  of cement produced.

2.3.9.3 TRENDS Emissions

       Using  the emission factor in the TRENDS spreadsheet results in an emission estimate of:

       77,895,000 * 15.95 / 2000 = 621,000 tons of SO,.

                                             128

-------
2.3.9.4  NAPAP Activity

       The NAPAP cement production is reported as 44,124,892 tons of cement produced dry
process and  16,041,286 tons of cement produced wet process.

2.3.9.5  NAPAP Emissions
       The NAPAP reported emissions for cement manufacturing are:
sec
3-05-006-06
3-05-007-06
3-90-002-01
3-90-004-02
3-90-005-02
Total
Description

Mineral Products Cement Manufacturing - Dry
Process: Kilns
Mineral Products Cement Manufacturing Wet
Process: Kilns
In-process Fuel Use -
Kiln/Dryer
In-process Fuel Use -
Kiln/Dryer
In-process Fuel Use -
Kiln/Dryer

Bituminous Coal: Cement
Residual Oil: Cement
Distillate Oil: Cement

                                                                   Tons SO. Emitted
                                                                             109,150
                                                                              76.614
                                                                              77,859
                                                                               3,574
                                                                               2.404
                                                                             269,601
       Additional emissions of 6,040 tons are listed in NAPAP for grinding and drying and primary
crushing and other not classified, and an additional 15,012 tons of SO2 were excluded from the
combustion categories of residual oil, distillate oil, bituminous coal, and natural gas, bringing the
total NAPAP emission estimate to 290,653 tons of SO:.

2.3.9.6 Conclusion

       Since the TRENDS method was last revised and the NAPAP inventory was completed, the
Portland cement section of AP-42 has been updated.   AP-42 currently lists the uncontrolled SO:
emission factor for the dry process as 7.0 Ibs/ton of clinker produced and for the wet process as 6.0
Ibs/ton of clinker produced when coal is the fuel.  Coal dominates as the fuel of choice providing 93
percent of kiln fuel consumption. The dry production is overtaking wet production (probably due to
the lower energy requirements of dry).  Statistics for manufacture of both (using all types of fuel) are
                                            129

-------
available in Minerals Yearbook "Cement."4  Table 8  "Clinker produced and fuel consumed by the
Portland cement industry in the United States by process" of Minerals Yearbook, 1986 "Cement"
provides 1985 statistics is as follows.
 Description                                                 Value
 Wet process (103 tons of clinker                             26,066
 produced)
 Dry process (103 tons of clinker                             37,797
 produced)
 Coal consumed (103 tons)                                    11,606
 Oil consumed (barrels)                                    755,000

       Assuming that the AP-42 emission factors (which are for coal burned) apply, emissions can
be calculated with these statistics as:

       (26,066,000 * 6.0 + 37,797,000 * 7.0) / 2,000 = 210,488  tons of SO2.

       These are uncontrolled  emissions.  AP-42  states that the use of a baghouse (for paniculate
control) would result in approximately 75 percent reduction in SO; due to the basic nature of the
paniculate (calcium).

       Assuming 75 percent control would result in emissions of:

       210,488 * (1 - 0.75) = 52,621 tons of SO2.

       The TRENDS estimate  apparently double counts the fuel sulfur emissions. The emission
factor that is applied in the  TRENDS method is out of date and already accounts for the fuel sulfur.
Double counting the fuel sulfur emissions is a significant error in the TRENDS method.
                                             130

-------
2.3.10 Glass Manufacturing

       The 1985 TRENDS emission value is 30,000 tons SO2. The 1985 NAPAP value is 23,000
for glass manufacturing.  The discrepancy between the two inventories is 7,000 tons of SCX. (30
percent).

2.3.10.1  TRENDS Activity

       In TRENDS, the glass production is obtained from the Current Industrial Repons. The SO:
emissions are from three categories: container glass - melting furnace; flat glass   melting furnace;
and pressed and blown glass - melting furnace.

       For TRENDS the activity is the sum of total production of flat glass, net packed weight of
glass containers, and miscellaneous glass products (pressed and blown glass). Flat glass is obtained
from Table 1A  "Summary of Flat Glass Production, Shipments, and Inventories:  1986 and 1985" of
Current Industrial  Reports: Flat Glass Summary for 7956.15 The value for 1985 was 3,670,719 tons.
The value for container glass is obtained from Table 5 "Shipments,  Production, and Stocks of Glass
Containers: 1985"  of Current Industrial Reports: Glass Containers Summary for 1986.IS The value
for 1985 was 22,196,448,000 pounds or 11,098,224 tons.  The value for miscellaneous glass products
is an  additional  10  percent of the combined flat and container glass production.  Total glass
production is then  calculated as:

       (3,670,719 + 11,098,224)  *  1.1  = 16,245,837 tons  of glass.

       This value matches the TRENDS  activity spreadsheet.

2.3.10.2  TRENDS Emission Factors

       The emission factors cited in the TRENDS method are as follows.
                                             131

-------
 SCC              Description                     Emission Factor   Units
 3-05-014-02       Mineral Products  Glass                     3.4   Ibs/ton of glass
                   Manufacture: Container glass                       produced
                   Melting furnace
 3-05-014-03       Mineral Products  Glass                     3.0   Ibs/ton of glass
                   Manufacture: Flat glass -                           produced
                   Melting furnace
 3-05-014-04       Mineral Products  Glass                     5.6   Ibs/ton of glass
                   Manufacture:  Pressed and                         produced
                   blown glass  Melting furnace

       These emission factors match the AJRS Facility Subsystem Source Classification Codes and
Emission Factor Listing for Criteria Air Pollutants1 document.  An overall emission factor of 3.56
Ibs/ton of glass produced is applied to all glass production in the TRENDS method.  The  3.56
emission factor is obtained  by assuming  that the distribution of glass manufacturing is 75  percent
container glass, 15 percent flat glass and  10 percent blown glass.

       3.4(.75) + 3.0(.15) + 5.6(.l) = 3.56 Ibs SCyton of glass  produced

2.3.10.3  TRENDS Emissions

       Using the TRENDS emission factor results in  1985 emissions of:

       16,245,837 * 3.56 /  2000 = 28,918 tons of SO2.

2.3.10.4 NAPAP Activity

       The reported NAPAP production  is as  follows.
 SCC              Description                                            Tons of Glass
                                                                             Produced
 3-05-014-02       Mineral Products - Glass Manufacture: Container             7,219,485
                   glass - Melting furnace
 3-05-014-03       Mineral Products  Glass Manufacture: Flat Glass -           2,397,263
                   Melting furnace
                                             132

-------
 SCC             Description                                            Tons of Glass
                                                                            Produced
 3-05-014-04      Mineral Products - Glass Manufacture:                       787.572
                  Pressed/Blown glass   Melting furnace
 Total                                                                     10,404320
       These reported production figures translate into a distribution of 69 percent container glass,
23 percent flat glass,  and 8 percent pressed and blown glass.  The NAPAP production figures are not
complete enough to be used as a basis for distribution of emissions.

2.3.10.5  NAPAP Emissions

       The reported 1985 NAPAP emission are not calculated solely using the  SCC emission factors.
The NAPAP emissions for the three  SCC codes are as follows.
 SCC             Description                                        Tons SO. Emitted
 3-05-014-02      Mineral Products - Glass Manufacture: Container                11,776
                  glass - Melting furnace
 3-05-014-03      Mineral Products - Glass Manufacture: Flat Glass -               3,616
                  Melting furnace
 3-05-014-04      Mineral Products - Glass Manufacture:                          3.110
                  Pressed/Blown glass - Melting furnace
 Total                                                                         18,502
       Additional emissions of 2,215 tons are listed in NAPAP for the  SCC category soda lime and
722 tons of SO2 are included for raw material handling and other not classified. This brings the total
NAPAP emission estimate  for glass manufacture to 21,439 tons of SO;. The TRENDS procedure
specifically excludes natural gas combustion in the  glass industry from  the industrial natural gas
combustion section. As a result the 1,582 tons of SO2 that are reported in NAPAP with industrial
combustion SCCs and glass manufacture SIC's should also be reported in this section. This brings
the NAPAP value to 23,021 tons of S02.
                                             133

-------
2.3.10.6  Conclusion

       The absolute difference between the TRENDS and NAPAP estimates is fairly small and there
is no evidence that either is in error.  However, the TRENDS method may benefit from two
comments.

       The purpose of averaging the production numbers and the emission factors in the TRENDS
methodology is unclear.  If pressed and blown glass represent 10 percent of the industry (both
production  and emission factor derivations assume this) the production and corresponding emission
factors could be applied directly.

       (3,670,719 * 3.0 + 11,098,224 * 3.4 +  1,476,894 * 5.6) A2000 = 28,508 tons

       Based on the NAPAP production numbers, the 10 percent pressed and blown glass
assumption may be a small overestimate.  Because this type of production has the highest SO:
emission factor, it would also cause the TRENDS estimate to be an overestimation.

2.3.11  Lime Manufacturing

       The 1985 TRENDS value is 30,000 tons SO2. The 1985 NAPAP value is 32,000 tons for
lime.  The discrepancy between the two inventories is 2,000  tons of SO2 (6 percent).

2.3.11.1  TRENDS Activity

       In TRENDS, the lime manufacturing estimate is obtained from "Facts and Figures for the
Chemical Industry" published in Chemical & Engineering News.r" In 1985 the U.S primary
production was reported as 15,800,000 tons. Annual lime production is also published in Table 1
"Salient Lime Statistics" of Minerals Yearbook, 1989 "Lime".* The 1985 value was reported as
15,690,000  tons of lime sold or used by producers.
                                            134

-------
2.3.11.2  TRENDS Emission Factor



       The SO2 emissions are calculated using an emission factor of 3.4 Ibs/ton of lime produced.

This emission factor was developed  from the actual S0: emissions reported in NEDS divided by the

NEDS lime production rate as reported in February 1980.



2.3.11.3  TRENDS Emissions



       Emissions are then calculated as:



       3.4 * 15,800,000 / 2,000 = 26,860 tons SO:.



2.3.11.4  NAPAP Activity



       The reported lime manufacturing production in NAPAP for the three main SCCs is as

follows.


 SCC             Description                                             Tons Lime
                                                                          Produced

 3-05-016-03      Mineral Products - Lime Manufacture: Calcining,             1,020,000
                  Vertical Kiln

 3-05-016-04      Mineral Products - Lime Manufacture: Calcining,            15,479,000
                  Rotary  Kiln

 3-05-016-17      Mineral Products - Lime Manufacture: Multiple                 135.000
                  Hearth  Calciner

 Total                                                                   16,634.000


2.3.11.5 NAPAP Emissions



       The reported NAPAP S02 emissions  for lime manufacture are as follows.
                                            135

-------
 SCC             Description                                        Tons SO. Emitted
 3-05-016-03      Mineral Products - Lime Manufacture: Calcining,                 3,085
                  Vertical Kiln
 3-05-016-04      Mineral Products   Lime Manufacture: Calcining,                18,647
                  Rotary Kiln
 3-05-016-17      Mineral Products   Lime Manufacture: Multiple                    572
                  Hearth Calciner
 3-90-002-03      In-process Fuel Use: Bituminous Coal, Lime kiln                 6,384
 3-90-004-03      In-process Fuel Use: In-process fuel: Residual oil,                  738
                  Lime kiln
 3-90-005-03      In-process Fuel Use: Distillate  Oil, Lime kiln                    	12.
 Total                                                                         29,438

       In addition, NAPAP reports emissions of 2,263 tons of SO: from the Other Not Classified
SCC for lime production, which brings the NAPAP total to 31,701 tons of SO,.
2.3.11.6  Conclusion

       The emission estimates in TRENDS and NAPAP are practically identical.  Nevertheless, the
TRENDS emission factor of 3.4 Ibs SO^ton of lime produced may be too low.  There are the
following three emission factors in the AIRS Facility Subsystem Source Classification Codes and
Emission Factor Listing for Criteria Air Pollutants1 document with units of Ibs  SO2 emissions/ton of
lime produced.

 SCC              Description                      Emission Factor   Units
 3-05-016-03       Mineral Products  - Lime                    8.2   Ibs SO:/ton of lime
                   Manufacture: Calcining,                           produced
                   Vertical kiln
 3-05-016-04       Mineral Products  - Lime                    5.1   Ibs SO2/ton of lime
                   Manufacture: Calcining,                           produced
                   Rotary kiln
 3-05-016-17       Mineral Products  - Lime                    8.2   Ibs SOVton of lime
                   Manufacture: Multiple Hearth                      produced
                   Calciner
       The distribution of the three types of lime calcining operations and how  they dominate the
industry should be investigated if an average emission factor is going to be used.  In addition,
investigation into the use  of control devices  for the lime manufacturing industry should be
                                             136

-------
investigated.  As discussed in cement, any paniculate control device for this industry will have very

good SO2 control due to the properties of the lime paniculate being captured.



       In the NAPAP inventory, calcining with a rotary kiln is the dominant method.  A weighted

emission factor based on the production figures used in the 1985 NAPAP inventory would lead to

the following results.  Total NAPAP production is:



       1,020,000  + 15,479.000 + 135,000 = 16,634,000 tons.



By percent, the production is distributed across the three categories.


  SCC             Description                                                 Percent
                                                                           production

  3-05-016-03       Mineral Products - Lime Manufacture: Calcining,                     6
                   Vertical kiln

  3-05-016-04       Mineral Products - Lime Manufacture: Calcining,                   93
                   Rotary kiln

  3-05-016-17       Mineral Products - Lime Manufacture: Multiple                       1
                   Hearth Calciner

       Using these production percents to weight an emission factor results in a factor in excess of

the 3.4 TRENDS  figure:



       8.2(.07) +  5.1 (.93) = 5.3 Ibs/ton of lime produced.



Using this emission factor results in a revised uncontrolled TRENDS emission rate:



       15,800,000 * 5.3 / 2000 = 42,000 tons of SO2.



2.3.12 Additional Industrial Process Emission Categories in the NAPAP Inventory



       The NAPAP inventory includes additional emission categories beyond the categories covered

by the TRENDS SO: emission estimation method. As stated earlier,  the TRENDS method  was

developed to reflect changes in emissions from large (> 10,000 tons of SCX/year) source categories.
                                             137

-------
Within the chemical manufacturing group of emissions, TRENDS only reports emissions for carbon
black, sulfuric acid and elemental sulfur production. Additional sources within the chemical
manufacturing source categories are listed in Table 2-12.
           TABLE 2-12. EMISSIONS FROM CHEMICAL MANUFACTURING
                        SOURCES INCLUDED IN NAPAP BUT NOT IN
                        TRENDS
Chemical Manufacturing
Source Category (SIC)
Ammonia Production (2873)
Charcoal Manufacture (2861)
Hydrochloric Acid (2819)
Plastics Production Specific Products (2821)
Synthetic Rubber (Manufacturing Only) (2822)
Ammonium Phosphates (2874)
Sulfur Recovery Plants" (2819)
Inorganic Pigments (2816)
Propylene, Butylene, Ethylene, and Olefm Production (2869)
Nitriles, Acrylonitrile, Adiponitrile Production (2869)
General Processes - Fugitive Leaks (2865, 2869)
Waste Gas Flares
No SCC Descriptor
Other
Total
SCC
3-01-003
3-01-006
3-01-011
3-01-018
3-01-026
3-01-030
3-01-032
3-01-035
3-01-197
3-01-254
3-01-800
3-01-900-99
3-01-999


1985 NAPAP
(tons)a
1.094
4.643
1,540
1.881
4,663
1.554
38,920
5,682
4,184
3,281
1,945
39,620
48,091
4.499
161.597
'The 1985 NAPAP Emissions Inventoiy (version 2):  Development of the Annual Data and
 Modelers' Tapes, EPA-600/7-89-012a, November 1989.
"Excluding emissions from sulfur recovery at petroleum refineries and oil and natural gas production
plants.
      In the mineral products group, TRENDS includes emissions from cement, lime and alass
manufacturing in the SO: emission estimation method. Mineral product source categories included in
the NAPAP SO: inventory but not addressed in the TRENDS estimate are listed in Table 2-13.
                                          138

-------
             TABLE 2-13.  EMISSIONS FROM MINERAL PRODUCTS
                          SOURCES INCLUDED IN NAPAP BUT NOT
                          IN TRENDS
Mineral Products
Source Category (SIC)
Asphaltic Concrete (2951)
Brick Manufacture (3251)
Ceramic Clay Manufacture (3261)
Clay & Fly Ash Sintering (3295)
Coal Cleaning & Surface Mining Operations (1111)
Fiberglass Manufacture - Wool & Textile Type Fiber (3229,
3296)
Gypsum Manufacture (3275)
Phosphate Rock (1475)
Stone Quarrying/Processing (1411, 1422, J423, 1429, 1499)
No SCC Descriptor
Other
Total
SCC
3-05-002
3-05-003
3-05-008
3-05-009
3-05-010
3-05-012
3-05-015
3-05-019
3-05-020
3-05-999

1985 NAPAP
(tons)3
1,959
2.193
2.362
1.389
12.481
2,487
9,394
4,651
6,190
5,676
1.918
50,700
The 1985 NAPAP Emissions Inventory (version 2): Development of the Annual Data and
 Modelers' Tapes, EPA-600/7-89-012a, November 1989.
      Finally, the NAPAP inventory includes additional sources of SO2 emissions from other

industrial processes as listed in Table 2-14.
                                      139

-------
            TABLE 2-14. EMISSIONS FROM OTHER INDUSTRIAL
                         PROCESS SOURCES INCLUDED IN NAPAP BUT
                         NOT IN TRENDS

                 Source Category (SIC)                      SCC       1985 NAPAP
                                                                         (tons)"

 Sugar Beet Processing (2063)                              3-02-016              1.918

 Other Food and Agriculture (0100, 0200, 0700, 2000, 2100,     3-02-XXX             1.449
 4200, 4400, and 5100)

 Other Pulp & Paper and Wood Products (2400, 2500. 2600, and  3-07-XXX              241
 2700)b
Rubber and Miscellaneous Plastics Products (3000, and 7500)
Fabricated Metal products (3400, 5000)
Electrical Equipments (7600)
Machinery, Miscellaneous Leather, and Leather Products
Organic Solvent/Petroleum Product Evaporation
Total
3-08-XXX
3-09-XXX
3-13-XXX
3-12-XXX
3-20-XXX
4-XX-XXX
773
258
463
2
2.374
7,478
"The 1985 NAPAP Emissions Inventory (version 2): Development of the Annual Data and
 Modelers' Tapes, EPA-600/7-89-012a, November 1989.
"Does not include sulfate (kraft) pulping, sulfite pulping, or neutral sulfite semichemical
 pulping.
                                        140

-------
                                       SECTION 3
                                     CONCLUSIONS

       This in-depth analysis of the 1985 TRENDS and NAPAP emission estimates for industrial
SO2 sources has led to the following conclusions.

       First, the TRENDS method is outdated relative to data that are currently available and the
1985 TRENDS estimates, as currently published, contain numerous mathematical errors.  In addition,
the TRENDS method relies on average fuel consumption values for some industrial processes. These
values  should be periodically revised and updated.

       Second, the TRENDS method does not account for any control measures in the majority of
the industrial SO2 emission estimates.

       Third, the NAPAP activity data are not complete enough to provide a reliable estimate of
industrial production.  Therefore, the NAPAP activity data do not support the generation of global
emission factors for future use in the TRENDS method.

       Fourth, for  the majority of source categories, the NAPAP emission estimates appear more
reasonable than the TRENDS  estimates.

       Due to findings in this report as well as other factors, the TRENDS methodology has  been
revised as of 1993; thus, references to TRENDS in this report will no longer be valid for years 1985
and beyond, effective  with the 1993 edition of the TRENDS  report. The new TRENDS methodology
uses the 1985 NAPAP Emission Inventory as a base.  Further changes will be seen in the TRENDS
reports  published in 1994 and  thereafter.  Thus, the reader is cautioned that comments on the EPA
TRENDS report in this document are valid for editions prior to 1993, but are not valid for the
editions for 1993 and  thereafter.

      The following  discussion has been excerpted from the previous section.
                                           141

-------
3.1  COMBUSTION SOURCES

       The combustion estimates in TRENDS and NAPAP are very similar.  This is due in part to
the  method utilized by NAPAP to estimate area source emissions.  In the NAPAP inventory, the
industrial point source fuel usage is subtracted from national fuel use estimates provided by the
Department of Energy, and emissions from the unaccounted for fuel are allocated to the area source
inventory. This method of determining area  source emissions or alternatively of accounting for
emissions from unaccounted for fuel use has several drawbacks.  First,  many sources do not report
fuel use. Second, the State of Texas does not report any individual fuel use, instead, all industrial
fuel use is reported at the county level to ensure confidentiality.  As a result, the initial fuel use
estimate, reported through NAPAP, is believed to be an underestimate.  Consequently, the NAPAP
inventory may allocate too much fuel and therefore too many fuel-related emissions to the area
source inventory.

       In both the NAPAP and the TRENDS method, the majority of the emissions are from the
combustion of bituminous coal.  The average sulfur content that is used in the computation of  the
emission factor is extremely important. The  TRENDS method has a complicated procedure to
determine an average fuel sulfur content based  on statistics for the coal-producing regions.  The
complex method has not been performed recently.  Additional research  into average sulfur contents
for  all  of the fuels (bituminous coal, anthracite, lignite, residual oil, distillate oil, crude oil and
process gas) is warranted.

3.2  COMBUSTION OF OIL

       The TRENDS estimate is 72,480 tons for distillate oil excluding cement plants and  petroleum
refineries. The NAPAP estimate for the same category is 107,358 tons of SO2.  The NAPAP
estimate includes 55,000 tons of SO2 from area sources.  The NAPAP and TRENDS estimates are
much closer if the NAPAP estimate does not include area source emissions.

       The TRENDS estimate is 459,510 tons for residual oil excluding cement plants,  petroleum
refineries and steel mills.  The NAPAP estimate for the same category  of emissions  is  605,200 tons
                                            142

-------
of S02. The NAPAP estimate includes 242,000 tons of S02 from area sources.  Again, the two
estimates would be much closer if the area source component of the NAPAP total were not included.

       There are large discrepancies in the oil activity data (quantity of reported combusted) used in
the TRENDS method versus the NAPAP inventory for both categories. TRENDS reports 3.426
million gallons and NAPAP reports approximately 2,000 million gallons of distillate oil consumed.
Both values appear to be in error.  Table 3 "Total Inputs of Energy for Heat, Power, and Electricity
Generation by Census Region, Industry Group and Selected Industries, 1985" of Manufacturing
Energy Consumption Survey: Consumption of Energy, 1985 reports 31,684,000 bbls consumed
(1,330.7 million gallons). This survey reports an activity that is approximately one-third the value
reported in TRENDS.8

       TRENDS reports 3,555 million gallons and NAPAP reports approximately 6,000 million
gallons of residual oil consumed.  Table 3 "Total Inputs of Energy for Heat, Power, and Electricity
Generation by Census Region, Industry Group and Selected Industries, 1985" of Manufacturing
Energy Consumption Survey: Consumption of Energy, 1985s reports 80,252,000 bbls consumed
(3,370.6 million gallons).

       The activity data used in TRENDS were derived from Fuel Oil and Kerosene Sales.3  If these
data were also used to compute area source activity in NAPAP, this could have resulted in an
overestimation of  both residual  oil and distillate consumed  and resulting emissions reported in
NAPAP through the area source category. It is likely that the NAPAP point source inventory
underreported both distillate and residual oil consumption and that the NAPAP area source inventory
overestimated emissions for these two categories.  The discrepancy between the distillate oil
consumption reported though Fuel Oil and Kerosene Sales  and  Manufacturing Energy Consumption
Survey: Consumption of Energy, 1985 should be investigated.

       The development of average sulfur contents for distillate and residual oil is not well
documented in the TRENDS procedure.  The sulfur content assumptions have a large impact on the
overall emission factor.  Additional effort should be expended to determine reasonable average sulfur
                                            143

-------
contents and to determine if industrial oil consumers are electing to use lower sulfur content oil. and
if so, what the overall effects are on emissions from this category.

       The NAPAP inventory also includes, as residual oil burned, approximately 932,000 x 10
gallons of crude oil which is burned during crude oil production. This oil use is not counted by
TRENDS  and was placed in the residual oil category because no other category existed in NAPAP

3.3 COAL  COMBUSTION

       As stated above, industrial coal combustion, specifically  bituminous coal combustion is  by far
the largest single category of industrial SO2 emissions.  In the TRENDS  method, emissions from
bituminous,  subbituminous and lignite are combined in one estimate and coal burned in cement and
lime kilns is subtracted.  TRENDS assumes that all of the remaining industrial coal combustion
occurs in boilers.  An average emission factor is developed based on an average sulfur content.

       The initial bituminous, subbituminous, and lignite activity value, including coal combusted in
lime manufacturing and cement plants, is 75.3 million tons in TRENDS versus 74.5 million tons in
NAPAP.  Both of these values are fairly consistent with Table 3 "Total Inputs of Energy for Heat,
Power, and Electricity Generation by Census Region, Industry Group and Selected Industries, 1985"
of Manufacturing Energy Consumption Survey: Consumption of Energy,  1985 which reports 59.195
million tons  of coal  burned in the industrial sector.  The anthracite activity value that is published in
the TRENDS activity spreadsheet could not be replicated. Following the published TRENDS
procedure manual resulted in an activity  value of 800,000 tons of anthracite versus 658,800 tons of
anthracite that is currently in the TRENDS activity spreadsheet.

       The published TRENDS emission estimate  for coal combustion is a sum of the anthracite
value and the bituminous, subbituminous, and lignite value.  The published TRENDS value of
1,840,000 tons of SO2 for coal combustion could not be replicated by following the published
TRENDS  procedure. The bituminous, subbituminous, and lignite emission estimate in TRENDS
excluding lime and cement emissions, is 1,670,000 tons of SO:  versus 1,710,000 tons of SO,
reported in NAPAP   The anthracite emission estimate is 11,000 tons of SO: in both TRENDS and
                                            144

-------
NAPAP.  Again, the NAPAP inventory relies on a total fuel balance to determine the area source

emissions (356,000 tons of SOj) for this category. For this category the addition of the area source

emissions brings the  NAPAP and TRENDS estimate closer together. The sulfur contents and

resulting emission  factors that were actually used in the 1985 estimate are not documented and

assumptions made  about an average sulfur content for the bituminous, subbituminous, and lignite

category may be the  reason for  the difference.


       The four types of coal that constitute the coal combustion category each have slightly

different emission  factors.  The  general emission factors for external combustion in industrial boilers

are as follows.
       Anthracite           1-02-001-01,07             39.0S
       Bituminous          1-02-002-01,19             39.0S
       Subbituminous       1-02-002-21,29             35.0S
       Lignite              1-02-003                  30.0S
       Although there are a couple of smaller emission factors for some types of bituminous coal

combustion (for example fluidized bed), these constitute a very small amount of the coal combustion

activity.


       Sulfur contents reported in the NAPAP inventory for these four general types of coal range as

follows.
       1 -02-001 -01,07             0.7 to  1.2 percent
       1 -02-002-01,19             1.0 to  1.9 percent
       1-02-002-21,29             0.4 to  1.5 percent
       1-02-003                   0.5 to 0.9 percent
      Average sulfur contents, based on emissions and reported coal consumption, are as follows.
       1-02-001-01,07             0.7 percent
       1-02-002-01,19             1.4 percent
       1-02-002-21,29             0.7 percent
       1-02-003                   0.7 percent

                                             145

-------
      Use of these sulfur contents results in the following average emission factors.
      Anthracite           1-02-001-01,07             27.3 Ib/ton burned
      Bituminous          1-02-002-01,19             54.6 Ib/ton burned
      Subbituminous       1-02-002-21,29             24.5 Ib/ton burned
      Lignite              1-02-003                   21.0 Ib/ton burned
      The TRENDS emission factor of 27.3 for anthracite combustion appears reasonable.  The
TRENDS emission factor of 54.3 for bituminous, subbituminous, and lignite is probably an
overestimation.

      The TRENDS procedure manual refers to the use of control assumptions as documented in
the EIA-767 data.  If control assumptions have been applied (EIA-767 data pertain to utilities and
should not be used to estimate controls on the industrial sector) they are not documented. There are
probably some industrial emission controls that are applied to the NAPAP estimates and are not
reflected in the TRENDS method.

3.4 NATURAL GAS COMBUSTION

      The TRENDS method excludes natural gas combustion from cement manufacturing,
petroleum refineries, the iron and steel industry, glass manufacture, and oil and natural  gas
production.  Once these emissions are excluded, the resulting TRENDS emission estimate is 1.400
tons of SO2. Because TRENDS rounds their estimates to 10,000 tons/year, the emission estimate for
natural gas combustion in  boilers is  0.0 tons.  The NAPAP emission estimate is 32,800 tons of S0:
excluding natural gas combustion from cement manufacturing, petroleum refineries, the iron and steel
industry, glass manufacture, and oil  and natural gas production.

      The activity rates for natural gas combustion in boilers reported in the NAPAP inventory are
not consistent with the emission estimates.  In addition, the total (unadjusted) natural eas reportedly
consumed is over 715,188 billion ft3 in NAPAP versus 5,901 billion ft3 in TRENDS. This mav be
due in part to the reporting of  natural gas consumption as a feedstock versus as a fuel in the NAPAP

                                            146

-------
inventory.  The TRENDS value is fairly consistent with the 4,512 billion ft3 reported through the
Manufacturing Energy Consumption Survey: Consumption of Energy, 79S5.8

       The high consumption of natural gas in the NAPAP inventory may correlate with an
overestimation of natural gas emissions. In the NAPAP inventory, the  user had the option of
entering an activity rate and allowing the system to calculate emissions based on an emission factor.
If the user incorrectly coded the activity for natural gas combustion  (usually through a
misunderstanding of the appropriate units)  and allowed the  system to estimate the emissions, the
emissions would be overestimated.

       Conversion factors are used to correlate natural gas combustion  with both the iron and steel
industry and the glass manufacturing industry.  These factors are suppose to  be periodically updated
based on information available through the Department of Energy.  The value for the steel industry
that is currently used is 4.25 million ft3 of  natural gas/103 tons of raw steel.  The value for the iron
and steel industry can be updated with information provided in Table 3 "Total Inputs of Energy for
Heat, Power, and Electricity Generation by Census Region, Industry Group, and Selected Industries,
1985" of Manufacturing Energy Consumption Survey:  Consumption of Energy, 1985* Natural gas
consumed by blast furnaces and steel mills was 400 billion  cubic feet (TRENDS calculated 375
billion cubic feet for raw steel).  Based on  a 1985  raw steel production of 88,259,000, a revised
factor for iron and steel would be:

       400,000 / 88,259 = 4.53 106 ft3/103  ton raw steel.

       The value for glass cannot be recalculated at this time because the reference cited  combines
stone, clay, and glass products.

3.5 MISCELLANEOUS FUELS

       The TRENDS method includes four categories of fuel in  the industrial SO2  miscellaneous
fuels category. The emissions from the four categories are  published together as one value of 80.000
tons of SO2.  Several apparent errors were  discovered in the TRENDS  method.  Correcting the errors

                                             147

-------
would result in a revised TRENDS estimate of 30,000 tons of SO2 for miscellaneous fuels.  The four
fuels are coke, coke oven gas, kerosene, and liquified petroleum gas (LPG) and are discussed
separately below.

3.5.1  Coke

       The coke emissions are 11,300 tons of SO: in NAPAP and 36,000 tons of SO: in TRENDS.

       There are several problems in the TRENDS activity data.  The activity value in the TRENDS
spreadsheet could not be reproduced by following the procedure.  TRENDS lists 1,343,000 tons in
the activity spreadsheet: following .the TRENDS  procedure resulted in a value of 1,621,000 tons.
The Manufacturing Energy Consumption Survey: Consumption of Energy, 1985* lists 1,952,000 tons
of coke and breeze for industries other than blast furnaces and  steel mills. Therefore, the activity
value for coke combustion is probably too low in the TRENDS procedure. Also,  it is unclear why
the TRENDS procedure includes petroleum coke delivered to electric utilities in the industrial coke
activity value.

       The coke combustion emission factor used in the TRENDS method probably overestimates
the emissions from this category.  TRENDS uses an emission factor of 30.0 Ib/ton burned.  The
AIRS Facility Subsystem SCC and Emission Factor Listing for Criteria Pollutants7 has an emission
factor of 39.0S Ib/ton burned. The NAPAP inventory lists an average coke sulfur content of 0.7
percent which results in an overall emission factor of 27.3 Ib/ton  burned.

       A revised emission  estimate using the smaller emission factor and the coke consumption
value derived following the published TRENDS  procedure would result in a TRENDS coke
combustion emission  estimate of 22,000 tons of  SO..
                                            148

-------
3.5.2 Coke Oven Gas

       The coke oven gas emissions are 2,700 tons of SO2 in NAPAP and 43,300 tons of SO; in
TRENDS.  It appears as though TRENDS overestimates SO2 emissions for combustion of coke oven
gas outside the iron and steel industry.

       TRENDS assumes 18.8 percent of the coke oven gas produced is burned in boilers outside the
iron and steel industry. The 18.8 percent is not documented. In addition, the TRENDS iron and
steel section assumes that 40 percent of coke oven gas is used in the iron and steel process
equipment (see Roll and Finish subsection of iron and steel). The  TRENDS procedure apparently
does not account for the remaining 40 percent of coke oven gas produced. Table  12 "Production and
Disposal of Coke Oven Gas in the  United States by Producing State: 1980" of Coke and Coal
Chemicals in 798020 reports that in 1980 coke gas use was 39 percent used by producers in heating
ovens, 58 percent was for other use by producers, 1.4 percent commercial sales, and  1.5 percent was
wasted.

       TRENDS lists a coke oven  gas average sulfur value of 1.605 percent.  The NAPAP inventory
lists an average sulfur content for coke oven gas of 0.5 percent  Using a factor of 1.4 percent of
coke oven gas bumed in industrial  boilers outside the iron  and steel industry and using the NAPAP
average sulfur content results in emissions of:

       451,616 * 0.014 * 680 * 0.5 / 2000 = 1,075 tons of SO2.

       Two issues in the TRENDS method need to be addressed.  First, the amount of coke oven gas
consumed outside of the iron and steel industry must be examined. Second, a reasonable sulfur
content for coke oven gas should be determined.

3.5.3  Kerosene

      The kerosene emissions are  421  tons of S02 in NAPAP and 2,491 tons of  SO2 in TRENDS.
The TRENDS method overestimates kerosene emissions. The emission factor used in the TRENDS

                                            149

-------
procedure to estimate kerosene emissions is actually an emission factor for distillate oil. The
emission factor cited is 10.77 lb/103 gallon burned.  Using a kerosene emission factor of 6.2 lb/10
gallon burned results in emissions of:

       462,630 * 6.2 / 2000 = 1,434 tons of SO2.

3.5.4 LPG

       The LPG emissions are 52 tons in NAPAP and 109 tons in TRENDS.  Both values are
insignificant.  Following the TRENDS procedure manual did not result in the same activity value for
LPG combustion as is published in the TRENDS activity spreadsheet.  The LPG activity value used
in the  1985 TRENDS  estimate is higher (1;979 million gallons versus  1,116 million gallons) than the
value reported through the Manufacturing Energy Consumption Survey:  Consumption of Energy,
79S5.8  Using  the value reported through the survey results in emissions  of:

        1,116,000 * 86.5 * 0.0013 / 2000 = 63 tons of SO2.

       The TRENDS  procedure for determining LPG combustion activity is difficult to understand.
A preferred  approach  may be to hold the value constant and update it every three years with a new
Manufacturing Energy Consumption Survey.

3.6  WOOD

       The TRENDS  emission estimate for wood combustion is  10,000 tons of SO,.  The NAPAP
estimate is 41,700 tons of SO2.  The area source component of the NAPAP estimate is rather
substantial (17,000 tons of SO:).  Nearly half of the NAPAP point source emissions are from a
general in-process wood combustion category.  The emission factor for this category (3-90-008-89) is
38.0S Ibs/ton burned.  NAPAP also reports an average sulfur content of  1.5 percent for the SCC.
The resulting emission factor (57 Ib/ton of wood burned) is substantially higher than the emission
factor used in  TRENDS and the rest of the  NAPAP categories (0.15 Ib/ton burned).  As a result this
SCC category  is responsible for a disproportionate  share of the NAPAP  wood combustion point

                                             150

-------
source emissions.  Due to the high emissions for this one point source category and the high area
source estimate, the NAPAP inventory probably overestimates the wood combustion emissions.

      Following the TRENDS procedure did not recreate the activity value that was used in the
calculation of the 1985 emission estimate for wood combustion.  However, once the emission
estimate is rounded to the nearest 10,000 tons, the difference  is insubstantial.

3.7 NON-FERROUS SMELTING SOURCES

      Non-ferrous smelting emissions are an important component of industrial SO2 emissions
because sulfur is present in  the ores.  Consequently, sulfur recovery is a imponant component of the
emission estimate. The TRENDS method determines sulfur recovery  through statistics reported
through the Minerals Yearbook.  The values cited in the reference (501,5000 tons of SO2 recovered
as H2SO4) were  apparently not used in the development of the published 1985 TRENDS emission
estimate (327,900 tons of SO2 recovered as H2SO4).

      The  TRENDS method includes emission estimates for primary copper, primary lead, primary
zinc, primary  aluminum, and secondary lead. The primary  copper estimates are obtained on a point
by point basis from the remaining domestic primary copper smelters.  As a result,  the TRENDS  and
NAPAP estimate are consistent for primary copper smelters.

      The primary lead and primary zinc estimates are combined and reported as one value.
Following the TRENDS procedure did not recreate the published primary lead and primary zinc
emission estimate.  The TRENDS method appears to overestimate the emissions from primary lead,
primary zinc, primary aluminum and secondary lead.

3.7.1  Primary Zinc

      The TRENDS published estimate for primary zinc production  is combined  with the primary
lead estimate and the published total of 240,000 tons of SO2 from both industries  could not be
recreated. This discussion is based on the estimate for primary zinc that was developed following

                                           151

-------
the TRENDS procedure.  The NAPAP and TRENDS estimates for emissions from primary zinc
production are very different. NAPAP reported emissions of 7,642 tons of SO2 and the TRENDS
method resulted in an emission estimate of 93,864 tons of SO2.  As stated above, the value used for
sulfur recovered as H:SO4 apparently was in error and this error could account for the inability to
recreate the published TRENDS value.

      The TRENDS method only accounts for emissions from multiple hearth roasters.  In addition.
TRENDS assumes all roasting is done in a multiple hearth roaster.  Two  additional SCCs for roasting
exist, flash  roaster (3-03-030-07) and fluid bed roaster (3-03-030-08). Both have a smaller emission
factor (404.4 and 223.5 Ibs/ton of concentrated ore processed respectively)  than the multiple hearth
roaster (1100 Ibs/ton of concentrated ore processed).

       The NAPAP inventory did not report the majority of emissions through the multiple hearth
roaster.  The NAPAP inventory may have overestimated SO2 emissions from some of the other
processes in zinc production (specifically the sinter strand and the vertical retort/electrothermal
furnace SCC 3-03-030-03,05). The  discrepancy in NAPAP where the majority of emissions were not
reported through the roasting process needs to be investigated.

3.7.2 Primary Lead

       As mentioned above, the published TRENDS value of 240,000 tons of  SO: for primary lead
and zinc could not be recreated.  The TRENDS estimate developed following the TRENDS
procedure was 34,500 tons of SO:, which differs markedly from the NAPAP estimate of 9S.775 tons
of SO,.

       The TRENDS method for  determining primary lead SO: emissions is very complex and
appears to be outdated relative to  the data that are currently provided in the Minerals Yearbook,
I989.4 The TRENDS method has four steps to determine lead  processing.  After folio win 2 the four
steps, the result was a lead processing value of 975,378 short tons, which did not match the value of
759,300 tons in the TRENDS activity spreadsheet.  After analyzing the four steps that currently
comprise the TRENDS emission estimation procedure for primary lead, it appears that the final
                                            152

-------
number is a simple sum of the emissions reported through NEDS (now AFS) and the sulfur
recovered as sulfuric acid.  The recovered sulfur is then subtracted from the emission estimate.
Therefore there should be complete agreement between NAPAP and TRENDS for this category.  If
the product of the TRENDS method is intended to be different from the simple sum, there are errors
in the TRENDS procedure manual that need to be addressed.

       The NAPAP activity for this category is fairly close to the activity published in the Minerals
Yearbook.* The Minerals Yearbook cites a  1985 production  of 543,403 short tons. TRENDS
assumes that there is a 2:1  ratio of concentrated ore processed to lead produced.  NAPAP reports
1,006,182 tons of concentrated ore processed in the blast furnace, which would correspond to a lead
production rate of 503,000 tons of lead.

3.7.3 Primary Aluminum

       The TRENDS estimate is 70,000 tons of SO: from primary aluminum smelters, the NAPAP
estimate is 58,400 tons of SO2.  The TRENDS estimate relies on an average emission factor derived
from the 1980 NEDS data for the State of Washington.  The validity of the TRENDS emission factor
for primary aluminum could not be confirmed and appears suspect for two reasons.  First, it relies on
one set of old emissions (not test) data. Second, there is no documentation of an adjustment due to
controls.

       There are three very different emission factors in the AIRS Facility Subsystem Source
Classification Codes and Emission Factor Listing for Criteria Air Pollutants document for primary
aluminum smelting, prebake (57.3 Ibs/ton), HSS (10.0 Ibs/ton), and VSS (17.0 Ibs/ton).7  An
investigation into the distribution of the three types of electro-reduction processes, their controls and
how they dominate the primary aluminum industry should be undertaken to develop an appropriately
weighted emission factor.  The  1985 NAPAP production estimates provide the following distribution
between the three process types: prebake (77.2 percent of production), HSS (7.9 percent of
production),  and VSS (14.8 percent of production). Using this weighing would result in an emission
factor of 47.5 Ibs/ton (versus the current TRENDS emission factor of 36.85 Ibs/ton).  If the 1985
NAPAP primary aluminum emission estimate is used to develop a revised emission factor, an overall

                                             153

-------
factor of 19.8 Ibs/ton Al produced results. The absolute factor is lower presumably because NAPAP
accounts for the effect of S02 controls.

       In the NAPAP inventory, only 38,063 tons of SO2 are attributable to the three processes
covered by the TRENDS method. Of the processes that are not  included in TRENDS, aluminum
hydroxide calcining is the most important. TRENDS includes the aluminum hydroxide calcining
process in the estimation of TSP and PM-10 emissions but not in the SO2 estimate.

3.7.4 Secondary Lead

       The TRENDS estimate for secondary lead smelters is 30,000 tons of SO2 compared to 20,700
tons of SO2 reported in the NAPAP inventory.  The TRENDS estimate has a slight error in that the
activity value for the blast furnace was not converted to english units. This will not  make a
significant difference in the published TRENDS estimate,  because the values that are published are
rounded to the nearest  10,000 tons. The TRENDS  estimate does not account for SO2 controls such
as baghouses and wet scrubbers.

       There is a new SCC,  with an SO2 emission factor of 144.0S Ibs SCyiO3 gallons burned, for
this category. The SCC is 3-04-004-07 for pot furnace  heater burning distillate oil.  This SCC is not
in the TRENDS method and not in the NAPAP inventory.

3.7.5  Other Non-ferrous Emissions Reported in  NAPAP

       Additional emissions  of 41,511  tons of SO2  are reported for other non-ferrous emission
categories in the NAPAP inventory.  The largest source categories include ferroalloy manufacture,
furnace electrode manufacture, secondary aluminum and secondary zinc.  Of these categories only
the ferroalloy source category has emissions of more than 10,000 tons of SO2 in  1985.
                                            154

-------
3.8  OTHER INDUSTRIAL PROCESS EMISSION SOURCES

       Significant differences exist in the 1985 TRENDS and NAPAP SO2 emission estimates for
other industrial process emissions. In general, the TRENDS estimates depend on national production
figures developed by the Department of Energy and the Department of Commerce and an emission
factor. The emission factor often represents the largest sources of emissions  within the category. To
compare the NAPAP and TRENDS estimate, the  entire NAPAP estimate for  the category is used.
This would include in-process fuel, fugitive emissions, and processes that are not specifically cited in
the TRENDS method.

       The TRENDS method, with very few exceptions, does not include effects of air pollution
control devices unless the effects are inherent in the process (e.g., sulfur recovery) or in the emission
factor. The TRENDS estimates for most categories are significantly higher than the corresponding
NAPAP estimates.  A major exception to this statement is the category Oil and Natural Gas
Production.

       The NAPAP inventory includes many additional source categories that are not included in the
TRENDS industrial SO2 emission estimation method.  The TRENDS method includes only three
categories  in the chemical manufacturing group; sulfur, sulfuric acid, and carbon black. The NAPAP
inventory includes ten additional chemical manufacturing source categories with combined additional
emissions of 127,000 tons of SO2. The TRENDS method includes three categories in the mineral
products manufacturing group; cement, glass, and lime.  The NAPAP inventory includes nine
additional mineral products manufacturing source categories with combined additional emissions of
50,800 tons of S0:.

3.8.1  Kraft Pulp Production

      The pubb'shed TRENDS  value for pulp and paper production is 250,000 tons of SO2 and is
probably an overestimate. The NAPAP value is 130,400 tons of SO2.
                                           155

-------
      The TRENDS method apparently used an emission factor of 11.3 Ibs/ton of air-dry
unbleached pulp to calculate kraft emissions, when a more appropriate value would have been
approximately 7 Ibs/ton of air-dry unbleached pulp.  It is unclear how the 11.3 emission factor was
derived since it is significantly higher (nearly 50 percent) than one calculated following the TRENDS
procedure.  Based solely on the discrepancy, TRENDS emissions may be overstated by nearly 70,000
tons.  Also, TRENDS does not account for the effect of any controls.  These two issues could result
in an overesdmation of SO: emissions from wood pulping processes.

       The TRENDS procedure does not include a third type of paper pulping process, semi-
chemical. Activity data for semi-chemical pulping are available and emission factors exist in
AP-42.19 Published statistics indicate that semi-chemical production has recently overtaken sulfite
(3.9 x 106 vs. 1.6 x 106 tons) production. If the semi-chemical  production is increasing with fewer
associated SO2 emissions, this industry trend should be reflected in the emission estimates.

3.8.2 Carbon Black Manufacture

       The carbon black production emission estimates are 28,031  tons of SO2 in NAPAP versus
14,585 tons of SO2 in TRENDS.  The TRENDS method appears to underestimate the emissions from
carbon black manufacture.

       The total  NAPAP production for the oil furnace of 1,013,232 is very similar to the value
TRENDS references (90 percent of total production) 1,156,500. The NAPAP value for the gas
furnace 98,179 is less similar to the value TRENDS references  (10 percent of total  production)
128,500 tons.

       Two questionable  items need to be addressed regarding  the NAPAP  emission estimates and
the TRENDS emission  factor. The NAPAP emission estimates by SCC show only a minority of
emissions from the oil furnace (3,958 out of 28,031 tons), although logically this would be the
source of most emissions.  However, the pellet dryer combustion furnace (with emissions  of 15 183
tons) is, in  essence, a thermal incinerator and emissions associated with the  furnace itself  are emitted
                                             156

-------
here. It is likely that engineers coding the NAPAP inventory indicated the vents as discrete emission
points in addition to the oil furnace emission source.

       Second, the TRENDS emission factor appears to be too small.  The TRENDS  procedure
initially uses a fairly high emission factor of 50 Ibs/ton for the flare from an oil furnace process (this
emission factor is supported by AP-42).  If a CO boiler and incinerator exist, primarily to control  CO
emissions, the AP-42 SO: emission factor drops to 35.2 Ibs/ton.  The TRENDS number is 22.7
Ibs/ton,  and is unlikely that the emission factor would drop that low, even if all sources had a CO
boiler and  incinerator.

       As  noted, the largest source of emissions in NAPAP is for the pellet dryer. There is no
corresponding category in TRENDS for the pellet dryer, although it is  likely that emissions have
been accounted for.  As stated  earlier, there is no NAPAP category- specifically for the flare,
however, the flare actually represents otherwise uncontrolled oil furnace emissions.

      The TRENDS documentation probably needs to be modified to ensure that all  emission points
and  sources are included. Further investigation of the NAPAP value is warranted to determine why
emissions associated with the oil furnace were distributed to other emission points (vents) if possible.
Finally,  the assumption that flares represent otherwise uncontrolled emissions could be confirmed  by
looking  at the control equipment for these sources coded in NAPAP.

3.8.3 Sulfuric Acid

      The NAPAP estimate of 217,166 tons of SO: is extremely close to the TRENDS estimate  of
215,405 tons of SO2. One concern about this category is the production of H2SO4 from recovered
sulfur. The NSPS does not apply to sulfuric acid production in conjunction with SO2 controls.  It is
unclear if the NAPAP data reflect only the chemical companies producing sulfuric acid or if NAPAP
estimates also reflect byproduct H,SO4 production.

      There are several minor errors in  the development of the published TRENDS estimate. The
errors included use of an incorrect production value for 1984 from which the 1985 emission factor is

                                             157

-------
derived.  Because the TRENDS estimates are rounded to 10,000 tons of SO2, the error did not make
a significant difference in the published estimate.

       This is the only category of industrial SO2 emissions where the TRENDS method addressed
the implementation of  NSPS.  The NSPS emission factor of 4 Ib SO:/ton of 100 percent sulfuric
acid produced is consistent with the emission factor for sulfuric acid contact process, 99.9 percent
conversion. In the NAPAP inventory, the activity  for 99.9 percent conversion dominates the
category. An analysis  of the production data provided in Current Industrial Reports, Inorganic
Chemicals22 reveals that production had a low value of 33,233,000 tons of sulfuric  acid in 1982 and
a high of 44,336,818 in 1990.  Because the NSPS  was promulgated in the 1970's, production over
33,233,000 (and at least 25 percent of production)  should be at the. NSPS level. The NAPAP data
indicate  that approximately 50 percent of 1985 production was at the NSPS level.

3.8.4  Sulfur Recovery Plants

       The TRENDS  procedure manual has a separate section for estimating SO2 emissions from
sulfur recovery plants.  The resulting emission estimates are published in two other categories,
natural gas production  and petroleum refining.  As a result, it is not possible to directly assess
whether  the published emission estimates were successfully recreated although using the emission
estimates does allow the total natural gas production and total petroleum refinery estimates to match
the published values.

       Errors were discovered in both the activity  data and the emission factors that were used to
calculate the published 1985 estimates.  As a result, the published TRENDS emission estimate is too
high.  The estimates using the erroneous information are 202,000 tons for petroleum refineries and
163,000  tons for natural gas production.  The corresponding NAPAP emission estimates  are 29.117
tons for  petroleum refineries and 59,498 tons for natural gas production.

       The activity data were  erroneously left in metric units rather than convened to Enslish units.
The emission factor was not calculated from AIRS data, as the procedure manual indicated  but
rather was held constant.  Using the revised emission factor (106.8  Ibs/ton of sulfur produced  versus
                                             158

-------
 137.5 Ibs/ton of sulfur produced) and corrected activity data resulted in TRENDS emission estimates
 of 172,696 tons for petroleum refineries and  139,374 tons for natural gas production (a decrease of
 53,198 tons of SO,).

       The emission factors reported in the AIRS Facility Subsystem Source Classification Codes and
 Emission Factor Listing for Criteria Air Pollutants1 range from 280 to 4 Ibs/ton  100 percent sulfur
 recovered depending on the sulfur removal efficiency.  Multiplying the NAPAP sulfur recovery
 production values by these emission factors results in a higher NAPAP total emission estimate of
 103,348 tons of SO2.  There is a discrepancy between the production values, emission  factors, and
 reported emissions in the NAPAP inventory.  The emissions for 95-96  percent recovery appear to be
 underestimated and the emissions for 99.9 percent recovery appear to be overestimated. Therefore
 there are probably errors in the NAPAP values, either in  the reported production  or in  the reported
 emissions.

       Additional research should be expended on this category to try and determine what types of
 sulfur recovery plants are in use in petroleum refineries and natural gas production fields.   Once
 there is additional information, a new appropriately weighted emission  factor could be  developed for
 the TRENDS procedure.

 3.8.5 Petroleum Refineries

       The petroleum refining emission estimates in TRENDS and  NAPAP are quite different.  The
 TRENDS method estimates emissions for six categories for a  combined estimate  of 830,000 tons of
 S0:.  NAPAP estimates emissions for more than six categories of emissions. For the six  categories
 that correspond to the TRENDS estimate, NAPAP estimates 520,445 tons  of SO2, however, NAPAP
 has a total estimate of 640,000 tons for petroleum refining.

       Fluid catalytic cracking dominates the TRENDS estimate with 326,317 tons. The  NAPAP
estimate is significantly lower, 204,647  tons of SO;.  It is unclear why the NAPAP emission  estimate
is so  much lower, the reported NAPAP activity is actually higher than  the TRENDS activity  (1,585
versus  1,324 million bbl/year fresh feed).

                                            159

-------
       Thermal catalytic cracking activity relies on an annual survey conducted by the Oil & Gas
Journal™ The NAPAP emission estimate and the NAPAP activity are both an order of magnitude
higher than the TRENDS values, although the emissions are relatively insignificant relative to fluid
catalytic cracking.  The emissions are 7,273 tons versus 522 tons and the activity is 12 versus 17
million bbl/year fresh feed.

       NAPAP reports significantly higher emissions for  oil combustion at petroleum refineries.
NAPAP has an emission estimate of 117,512 tons versus  the TRENDS estimate of 44,360 tons of
SO2.  The TRENDS procedure should be rewritten to reference the oil fired process heater SCC's.
The sulfur content of the oil burned and consequently the emission factor used by TRENDS for oil-
fired process heaters appears to be too low. The majority of the oil reported burned is actually crude
oil (94 percent) and the distillate oil (5.4 percent) is far more significant than the residual oil (0.6
percent). Therefore the use of a residual oil emission factor is not very accurate.  There  is no
emission factor for combustion of crude oil in process heaters at a petroleum refinery. Emission
factors used in the industrial oil combustion section are 42.3 lbs/103 gallon burned for distillate oil
and 258.5 lbs/103 gallon burned for residual oil. Using the residual  oil emission factor, which
appears to be  the intent of the TRENDS procedure manual, would result in 76,911  tons of SO, from
process heaters burning oil.

       The emission estimates for gas-fired process heaters are 117,237 tons of SO2 in NAPAP
versus 231,106 tons of SO2 in TRENDS.  For gas-fired process heater emissions, the AIRS Facility
Subsystem Source Classification Codes  and Emission Factor Listing for Criteria Air Pollutants1
breaks out the emission factors for natural-gas fired 0.6 lbs/106 ft3 versus 950.0S lbs/106  ft3 process-
gas fired. The TRENDS emission factor of 253.1  lbs/106 ft3 appears to be somewhere in between the
emission factors for the two fuels.  The quantity of process gas combusted as a fuel is three times as
high  as the amount of natural gas combusted. The natural gas emissions are 146 tons of SO,.

       Finally, the emission estimates for sulfur recovery at petroleum refineries is also very
different. The NAPAP inventory reports emissions of 29,117 tons of SO2 and the TRENDS  estimate
is 202,125 tons of SO:. Errors were discovered in the execution of the TRENDS method (as
discussed in Section 2.3.3) and the TRENDS emission estimate should be 172,696 tons of SO,
                                             160

-------
however the numbers are still very dissimilar.  Research into the types of sulfur recovery units
employed at petroleum refineries (and their consequent emissions) is warranted.

       The data required to determine thermal  catalytic cracking versus the fluid catalytic cracking
are no longer available.  The thermal catalytic cracking contributes 0.16 percent of the cracking
emissions in the TRENDS estimate but contributes 3.4 percent in the NAPAP inventory.  Significant
effort to find a replacement source of data for thermal cracking may not be warranted, however,
additional research into why the NAPAP activity rates for thermal catalytic cracking are so different
from TRENDS  may be warranted.  Effort to determine why the fluid catalytic cracking estimate is so
much lower in NAPAP is definitely warranted.

3.8.6 Natural Gas Production

       The NAPAP and TRENDS estimates for this category are very different. The TRENDS
estimate is made up of two numbers, emissions from combustion of natural gas during natural gas
production and emissions from sulfur recovery units at natural  gas plants.
       Both NAPAP and TRENDS have small combustion estimates (460 versus 7,660 tons of
The estimates for sulfur recovery are very different (163,143 tons of SO2 in TRENDS versus 59,498
tons of SO2 in NAPAP). As stated earlier, errors were discovered in the TRENDS estimate and the
value  should be 139,374 according to the procedure manual.  Research into the type of sulfur
recovery units that are utilized in natural gas production should be conducted.

       NAPAP reports an additional  264,911 tons of SO2 from standard natural gas production
processes including gas-sweetening amine process, gas stripping operations and flares.  It is unclear
why these processes are not accounted for in the TRENDS method.
                                            161

-------
3.8.7  Iron and Steel

       All four of the iron and steel categories, coke manufacture, sintering, open hearth furnace, and
roll and finish  operations, have significantly different emission estimates in TRENDS and NAPAP.
The four categories are discussed separately.

       For coking emissions, TRENDS lists the six SCC categories that are used in their estimate.
Using these six SCC categories provides coking emission estimates of 65,367  tons (NAPAP) versus
 162,000 tons (TRENDS).  Even if all of the NAPAP coke emissions are counted, the total NAPAP
estimate for coke is only 74,629 tons of SO2.

       The TRENDS method appears to overestimate coking emissions. One  possible cause for over
estimation in the procedure is the inclusion of beehive process coke manufacturing in the activity
number when  beehive process coke production may  not have any associated SO2 emissions (there is
a 0.0 as an emission factor in the SCC book and the NAPAP emission estimate is  1,599 tons of
SOj).  Another possible disconnect is the accounting of SO2 control techniques in the NAPAP
inventory.  Finally, the NAPAP inventory may have emissions from coke production reported in
other iron and steel processing steps.

       The sintering emission estimate is 21,000 tons of SO2 in TRENDS versus 34,506 tons of S0:
in NAPAP.  The open hearth emission estimate is 4,650 tons of SO2 in TRENDS versus 1,169 tons
of SO2 in NAPAP.  For both these categories, the TRENDS method utilizes an emission factor that
was derived from 1980 NEDS statistics.

       The 1985 NAPAP inventory underwent significantly more review than the  1980  NEDS
estimates. In addition, the 1985 numbers are more current.  Therefore the 1985 values are more
suitable for use to calculate sintering and open hearth emissions in the TRENDS method, although
their absolute reliably for this purpose is unclear. Revising  the emission factor based on the 1985
NAPAP emission estimate  and utilizing the activity  data that are available through the Minerals
Yearbook would make the TRENDS and NAPAP emission estimates equivalent for these two
categories of emissions.  The overall sintering emission factor would increase from 2.5  Ibs/ton steel
                                            162

-------
produced to 4.11 Ibs/ton of pig iron sintered.  The overall emission factor for open hearth furnaces
would decrease from 1.5 Ibs/ton produced to 0.4 Ibs/ton produced. These changes in the emission
factors are so drastic that they cast doubt on this method of obtaining an emission factor.

       The TRENDS roll and finish emission estimate of 168,000 tons of SO2 is substantially higher
than the NAPAP estimate of 25,304 tons of SO2. The TRENDS category may be misnamed as it is
really a sum of emissions from combustion of coke oven gas and residual oil. Although a long
involved procedure is put forth to calculate the roll and finish emission factor, the factor has not
changed from  1985 though 1991 and therefore the procedure has probably not been used.  After
following  the procedure, a roll and finish emission factor of 4.04 was developed versus  3.8 Ibs/ton of
raw steel in the current TRENDS spreadsheets.  Both of these emission factors are too high.

       The reason the TRENDS roll and finish emission  factor is so high is based on the ratio  of
residual oil used to produce steel. The TRENDS Procedures Manual uses a factor of 7.38 gal/ton of
steel produced. Based on the data in Table 3 "Total Inputs of Energy for Heat, Power, and
Electricity Generation by Census Region, Industry Group and Selected Industries, 1985" of
Manufacturing Energy Consumption Survey:  Consumption of Energy,  1985,  the total residual oil
used in blast furnaces and steel mills was 5,458,000 barrels in 1985.8  Together, the changes in the
coke oven gas and residual oil emission estimate would decrease the TRENDS roll and  finish
estimate from  168,000 tons to 87,000 tons SO2.

3.8.8  Cement Manufacturing

       There is a significant difference in the 620,000 tons of SO, estimated in the TRENDS method
versus the 290,700 tons of SO2 estimated in the NAPAP  inventory for cement manufacture.  The
TRENDS method apparently double counts the fuel sulfur. The AP-42 emission factor  of 10.2
Ibs/ton cement produced that was used  in the TRENDS method accounts for the fuel sulfur. The
TRENDS method adds coal, residual oil, and distillate oil combustion  emission estimates to the
estimate made with the 10.2 Ibs/ton cement produced emission factor.  This is a significant error in
the TRENDS method.
                                             163

-------
      In the time since the TRENDS method was last revised and the NAPAP inventory was
completed, the Portland cement section of AP-42 has been updated.  AP-42 currently lists the
uncontrolled SO2 emission factor for the  dry process as 7.0 Ibs/ton of clinker produced and for the
wet process as 6.0 Ibs/ton of clinker produced when coal is the fuel.  Coal dominates as the fuel of
choice providing 93 percent of kiln fuel consumption. The dry production is overtaking wet
production with a corresponding lower energy requirement per ton of clinker produced.

       Statistics for manufacture of both (using all types of fuel) are available in Minerals Yearbook
"Cement".4  Total 1985 production using the wet process was 26,066 10" tons of clinker and using
the dry process was 37,797 103 tons of clinker. Assuming that the AP-42 emission factors (which
are for coal  burned) apply, emissions can be calculated as 210,500 tons of SO2. These are
uncontrolled emissions.  AP-42 states that the use of a baghouse  (for paniculate control) would result
in approximately 75 percent reduction in SO2 due to the basic nature of the paniculate (calcium).
Assuming 75 percent control would result in emissions of 52,600 tons of SO2.

3.8.9 Glass Manufacturing

       The TRENDS estimate for glass manufacture is 30,000 tons of SO2 and the NAPAP estimate
is 23,000 tons of SO2. The absolute difference between the TRENDS and NAPAP estimates is  fairly
small and there is no  evidence that either is in error. However, the TRENDS method may benefit
from two comments.

       The purpose of averaging the  production numbers and the emission factors in the TRENDS
methodology is unclear.  If pressed and blown glass represent 10 percent  of the industry (both
production and emission factor derivations assume this) the production and corresponding emission
factors could be applied  directly.

       Based on the NAPAP production numbers, the 10 percent pressed and blown glass
assumption may be a small overestimate. Because this type of production has the highest SO,
emission factor, it would also skew the TRENDS estimate to an  overestimation.
                                             164

-------
3.8.10 Lime Manufacturing

       The TRENDS estimate  for lime manufacture is 30,000 tons of SO, and the NAPAP estimate
is 32,000 tons of SO2.  Statistically there is no difference in the emission estimate in TRENDS and
NAPAP. Nevertheless, the TRENDS emission factor  of 3.4 Ibs SO2/ton of lime produced may be too
low.  There are three emission  factors  in the AIRS Facility Subsystem Source Classification Codes
and Emission Factor Listing for Criteria Air Pollutants7 document with units of Ibs S02
emissions/ton of lime produced.  Calcining in a vertical kiln and multiple hearth calcining both have
an emission factor of 8.2 Ibs SO2/ton of lime produced.  Calcining in a  rotary kiln has an emission
factor of 5.1 Ibs SO^ton of lime produced. In the NAPAP inventory, calcining with a rotary kiln is
the dominant method 93 percent of production).  Calcining in a vertical kiln or in a multiple hearth
calciner is 7 percent of production.  Using these production statistics results in a weighted average
emission factor of 5.3  Ibs/ton lime produced.  Using this emission factor results in a revised
uncontrolled TRENDS emission estimate of 42,000 tons of SO2.

       An investigation into the distribution of the three types of lime calcining operations and how
they dominate the industry should be undertaken if an average emission factor is going to  be used.
In addition, an investigation into the use of control devices for the lime manufacturing industry
should be undertaken.  As discussed in cement, any paniculate control device for this industry will
have very good SO2 control due to the properties of the lime paniculate being captured.
                                             165

-------
                                      SECTION 4
                                     REFERENCES
1.     Saeger, M. et al.  The 1985 NAPAP Emissions Inventory (Version 2): Development of
      the Annual Data and Modelers1 Tapes. EPA-600/7-89-012a (NTIS PB91-119669).
      U.S. Environmental Protection Agency, Air and Energy Engineering  Research
      Laboratory.  Research Triangle Park, NC.  November 1989. pages 3-32 through 3-34.

2.     U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
      National Air Pollutant Emission Estimates, 1900-1991. EPA-454/R-92-013.  Research
      Triangle Park, NC.  October 1992.

3.     U.S. Department of Energy, Energy Information Administration. Fuel Oil and
      Kerosene Sales 1989. DOE/EIA-0535 (89).  Washington, DC.  January 1991.

4.     U.S. Department of the Interior, Bureau of Mines. Minerals Yearbook 1986.  Volume
      I: Metals and Minerals.  Washington, DC.  1988.

5.     U.S. Department of Energy, Energy Information Administration. Petroleum Supply
      Annual 1985. Volume 1. DOE/EIA-0340  (85)/l. Washington, DC.  May 1986.

6.     Shelton, E.M., and C.L.  Dickson, Heating Oils, 1985, NIPER-141 PPS 85/4, Prepared
      by the National Institute for Petroleum and Energy Research, Bartlesville, OK. July
      1985.

7.     U.S. Environmental Protection Agency, National Air Data Branch. AIRS Facility
      Subsystem Source Classification Codes and Emission Factor Listing for Criteria Air
      Pollutants.  EPA-450/4-90-003 (NTIS PB90-207242). Research Triangle Park, NC.
      March 1990.

8.     U.S. Department of Energy, Energy Information Administration Manufacturing Energy
      Consumption Survey: Consumption of Energy, 1985.  DOE/EIA-0512 (85).
      Washington, DC. November 1988.

9.     U.S. Department of Energy, Energy Information Administration.  Cost and Quality  of
      Fuels for Electric Utility Plants 1985. DOE/EIA-0191(85). Washington DC  July
      1986.

10.    U.S. Department of Energy, Energy Information Administration.  Coal Distribution
      January-December 1985. DOE/EIA-0125  (85/4Q).  Washington, DC.  April 1986.

11.    U.S. Department of Energy, Energy Information Administration.  Quarterly Coal
      Report, October-December 1985.  DOE/EIA-0121 (85/4Q). Washington, DC.  April
      1986.
                                           166

-------
 12.    Facts and Figures for the Chemical Industry, Chemical and Engineering News, Volume
       64, Number 23, Washington, DC. June 9, 1986.

 13.    U.S. Department of Energy, Energy Information Administration.  Natural Gas Annual
       1985.  DOE/EIA-0131 (85). Washington, DC.  November 1986.

 14.    U.S. Department of Commerce, Bureau of the Census. Current Industrial Reports:
       Flat Glass Summary for 1986. MO32A(86)-5.  Washington, DC.  June 1987.

 15.    U.S. Department of Commerce, Bureau of the Census. Current Industrial Reports:
       Glass Containers Summary for 1986.  M32G(86)-13.  Washington, DC. May 1987.

 16.    U.S. Department of Energy, Energy Information Administration.  Manufacturing
       Energy Consumption Survey: Consumption of Energy, 1988.  DOE/EIA-0512(85).
       Washington, DC.  November 1988.

 17.    U.S. Department of Energy, Energy Information Administration.  Electric Power
       Annual 1985.  DOE/EIA-0348(85). Washington, DC. July 1986.

 18.    U.S. Department of Energy, Energy Information Administration.  Coal Data: A
       Reference. DOE/EIA-0064(87).  Washington, DC.  May  1989.

 19.    U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.
       Compilation of Air Pollutant Emission Factors.  Volume I:  Stationary Point and Area
       Sources.  Fourth Edition.  AP-42 (GPO 055-000-00251-7).  Research  Triangle Park.
       NC. September 1985.

 20.    U.S. Department of Energy, Energy Information Administration. Coke and Coal
       Chemicals in 1980. DOE/EIA 0120 (80).  Washington, DC. November 1981.

 21.    U.S. Department of Energy, Energy Information Administration.  Estimates of U.S.
       Biofuels Consumption 1990. DOE/EIA-0548(90).  Washington, DC.   October 1991.

 22.    U.S. Department of Commerce, Bureau of the Census. Current Industrial Reports:
       Pulp, Paper, and Board 1986.  MA26A(86)-1.  Washington, DC. September 1987.

 23.    U.S. Department of Commerce, Bureau of the Census. Current Industrial Reports:
       Lumber Production and Mill Stocks 1991.  MA24T(91)-1. Washington, DC.
       September 1992.
24.    U.S. Department of Commerce, Bureau of the Census.  Current Industrial Reports:
      Inorganic Chemicals 1986. MA28A(86)-1. Washington, DC.  October  1987.

25.    Annual Refining Survey.  Oil & Gas Journal.  Volume 84. March 24, 1986
                                           167

-------
26.    Symes, R., Refinery Operating Ratio, Crude Petroleum. Communication from United
      States Department of Commerce, Economics and Statistics Administration, Bureau of
      Economic Analysis, Washington D.C.  April 27, 1993.

27.    Symes, R., Beehive and oven coke (byproduct), production. Communication from
      United States Department of Commerce, Economics and Statistics Administration,
      Bureau of Economic Analysis. Washington, DC. April 23, 1993.

28.    Becker, B. Pig iron production.  Communication from United States Department of
      Commerce, Economics and Statistics Administration, Bureau of Economics Analysis,
      Washington, DC.  April 29,  1993.

29.    Becker, B. Raw steel production. Communication from United States Department of
      Commerce, Economics and Statistics Administration, Bureau of Economic Analysis,
      Washington, DC.  April 29,  1993.
                                         168

-------
                                       APPENDIX A

             EPA TRENDS PROCEDURE FOR INDUSTRIAL S02 EMISSIONS
       The TRENDS Procedures Document was developed to support the TRENDS emission
estimation method using spreadsheets to perform the calculations.  As a result, the TRENDS
Procedure as published, is somewhat disjointed and the user needs to jump around the document to
obtain the values that are needed.  During the development of the annual TRENDS estimate, the
document works because once a number is developed, the  spreadsheet automatically uses the number.

       The following procedure was developed by compiling the appropriate sections into a
sequential series of steps, thus avoiding jumping around the document.  In addition, sections that do
not pertain to the industrial SO2 emission estimates are not included. This would include whole
source categories such as highway vehicles as well as sources within a source category that are not
part of the TRENDS SO2 method, generally due to the emission cutoff level that was imposed on the
TRENDS procedure (10,000 short tons per year).

       The order of the source categories within this document follows the published TRENDS
procedure manual.  The procedures described below have not been edited under this effort.
Interpretation of the procedure is described within  the body of the report and this section has been
left intact in order to allow the reader to develop a different interpretation if appropriate.

       Two spreadsheets accompany the TRENDS procedure manual.  One spreadsheet contains the
emission factors for the source categories and is used to calculate current TRENDS estimates. The
second spreadsheet contains historical activity data and is used in the projection of TRENDS
estimates. On occasion, the emission factors reported in the text of the Procedures manual did not
correspond to the emission  factors located in the spreadsheets.  The emission factors as they appear
in the spreadsheet (in metric units) are included at the end of each section.
Anthracite Coal

Activity

       From Coal Distribution, table entitled "Distribution of U.S. Coal by Origin and Consumer,"
obtain the distribution of anthracite from Pennsylvania to industrial less coke plants.  State data is
provided.

Emission Factors

Table 1.2-1, AP-42 Fourth Edition, Volume I

   SCC      SO,
1-02-001-01  39.0 S
1-02-001-04  39.0 S

                                            A-l

-------
       Weight the emission factors based on the AIRS/FS AFP650 report (emissions by SCC report).
Fuel totals and actual emissions are reported in this printout. Use an ash content of 11  percent and a
sulfur content of 0.7 percent.

EF from print-out:  24.8 MLbs/tons

Bituminous Coal and Lignite

Activity

       Obtain the total national consumption by "other industrial" (not including coke plants) from
the Quarterly Coal Report.  From this total, subtract the sum of the following three  values:

  1. Consumption by cement plants in Mineral Industry Survey, Cement.

  2. Consumption by lime plants.  Calculated from C&E News. Estimated coal consumption is lime
       production, multiplied by 0.1 ton coal/ton lime produced.

Emission Factors

       Obtain a weighted average EF from AIRS/FS AFP650, for SCC 102002.  AFP650 is used to
obtain amounts of coal burned for each SCC.  These are used as weighting factors with  the emission
factors obtained from the AIRS/FS SCC listing to calculate the weighted average EF.  The value for
SO2 is 38.1(S) Ibs SOJ ton.  None of the emission factors need to be changed unless AP-42 changes.
However, ash and sulfur content numbers may change. Therefore it may be  necessary to change the
emission factors.

       Obtain the average percentage sulfur content from Coal Production (ref. 30) for shipments
from each coal-producing state to other industrial consumers. Use the sulfur content data for the
latest available year. As  a first step it is necessary to determine an  average sulfur content value for
each coal production district.  The reported sulfur contents in ref. 30 are for each state.  Reference
30 also contains a description of the coal production districts.  This information can be used to match
the states to coal production districts. For those districts that represent only parts  of states  ref 30
also gives coal production by  county  for each state, which can be used to estimate the shipments
from each district from component states.  Compute a weighted average sulfur content value for each
production  district, as necessary. Weight these district averages by the shipments data from each
district to destination contained in the Coal Distribution report.  In this case the destination  is "other
industrial' consumers.

       NOTE: This procedure is quite cumbersome, and thus has not been applied  The latest
       available data for sulfur content of coal by production district is for 1978  These data have
       been used for all TRENDS calculations for  1979-1985.  (Assumes sulfur contents of coal
       from each production district have not changed since then.)  In the future DOE/EIA mav
       produce new reports that give  information on sulfur content in industrial coal shipment  The
       reader should make use of any new DOE reports to improve this procedure
                                             A-2

-------
EF from print-out:   49.34 MLbs/tons (1990)
                    46.85 MLbs/tons (1991)

Residual Oil

Activity

       Residual oil and distillate oil source categories can be done together for fuel consumption
only.

       Obtain the "adjusted" quantity of residual oil sales for industrial and oil company use from
Fuel Oil and Kerosene Sales 19xx. Subtract the total of the following three statistics:

  1. Quantity of oil consumed by cement plants reported in Mineral Industry Survey, Cement.
       Assume that 2/3 of the oil reported is residual oil; convert to gallons.

  2. Quantity of residual oil consumed by petroleum refineries reported in Petroleum Supply
       Annual, table entitled, "Fuel Consumed at Refineries by PAD District." Convert to
       gallons.

  3. Quantity of residual oil consumed by steel mills. From the Survey of Current Business,
       table containing information on Metals and Manufactures, obtain the quantity of raw
       steel production in short tons and multiply by 0.00738 x 106 gal/103 ton steel.  (This
       value should be updated for 1982 and later years based on the 1982 Census of
       Manufactures. Fuels and Electric Energy  Consumed).

Emission Factors

Table 1.3-1, AP-42 Fourth Edition, Volume I

   SCC       SO,
1-02-004-01   158.6S

S = 1.82

       For  SO2, obtain a  sulfur content value for No. 6  fuel oil from Heating Oils. An average value
can be interpreted from the graphs presented in this report.

EF from print-out:   261.9 MLbs/10*3 Gal. (1990)
                    277.2 MLbs/10*3 Gal. (1991)
                                             A-3

-------
Control

S02:  From ref. 33c.

NOTE: Ash  and sulfur content numbers may change.  Therefore, it may be necessary to change the
emission factors

Distillate Oil

Activity

       These values were derived simultaneously with  residual oil consumption.
       Obtain the "adjusted" quantity of distillate oil sales to industrial and to oil companies from
Fuel Oil and  Kerosene Sales 19xx. Subtract  the total of the following statistics:

  1. Quantity of oil  consumed by cement plants reported in Mineral Industry Survey. Cement.
       Assume that 1/3 of the oil reported is distillate oil.  Convert to gallons.

  2. Quantity of distillate oil consumed by petroleum refineries reported in Petroleum Supply
       Annual, table entitled, "Fuel Consumed at Refineries by PAD District." Convert to
       gallons.

Emission Factors

Table  1.3-1, AP-42  Fourth Edition, Volume I

   SCC       SO,
1-02-005-01   143.6S
1-02-005-04   150.0S

       Weight the factors above based on the AIRS/FS AFP650  report (see Table 3.9-1).

       For SO:, obtain average sulfur content values for No.  1, No.  2 and No. 4 oils reported in
Heating Oils.  Weight these values by the corresponding distribution  to industrial reported in Fue.
Oil and Kerosene Sales 19xx. to obtain  a weighted average sulfur content value.

EF from print-out:    35.6 MLbs/10*3 Gal. (1990)
                    35.0 MLbs/10*3 Gal. (1991)

Control

SO2:  From ref. 33c.

NOTE: Ash  and sulfur content numbers may change.  Therefore, it may be necessary to chan°e the
emission factors.                                                                        c
                                             A-4

-------
          Table 3.9-1. Weighted Average Emission Factors for Industrial Oil Combustion
Boiler Type
Boilers burning No.
2
Boilers burning No.
4
Turbines
1C engines"
Weighted Average
(lbs/1000 gal)
AIRS/FS
Consumption
32758
3640
30193
6536

AP-42 Emission Factors
(lbs/1000 gal)
CO





NOX





SO2





voc
*





                    * Use the nonmethane VOC EF.
                    ** Internal combustion engines.

Natural Gas

Activity

Boilers:  Obtain the total industrial consumption figure for natural gas from Natural Gas Annual.
Subtract from this  figure the sum of the following:

 1.  Total natural gas consumption by cement plants obtained from Mineral Industry Survey. Cement.

 2.  Total natural gas consumption by petroleum refineries obtained from Petroleum Supply Annual,
(Table:  Fuels Consumed  at Refineries).

 3.  Total natural gas consumption by iron and steel industry.  Obtain the raw steel production
       from Mineral Industry Survey, Iron and Steel. Multiply the production figure by 4.25
       x 106 cu. ft. natural gas/1000 tons steel*

 4.  Total natural gas consumption by glass manufacture industry. Take the total production
       as computed for glass production from Table 3.15.35, "Particulate Emissions from the
       Mineral Products Industry."  Multiply this figure by 10.8 x 106 cu. ft. natural gas/1000
       tons glass produced.*

"This value should  be updated for 1982 and later years based on data from the 1982 Census of
Manufactures and the Annual Survey of Manufactures. Fuel and Electricity Energy Consumed.
                                            A-5

-------
For glass activity data, see Glass below.

Glass: Table 8.13-1, AP-42 Fourth Edition
                                                                   IbsATon
SCC                Description                        JSP         PM-10
3-05-014-02  Container Glass:  Melting Furnace         1.4           1.32
3-05-014-03  Flat Glass:  Melting Furnace               2.0           1.9
3-05-014-04  Pressed & Blown: Melting Furnace        17.4          16.5

Weight the EF's shown  above by the factors shown in Table 3.15-4 [see Glass below].

Gas Pipelines and Plants:  Obtain the total natural gas consumption for lease and plant fuel plus
pipeline fuel from Natural Gas Annual.

Emission Factors

Table 1.4-1, AP-42 Fourth Edition, Volume I

    SCC       SO,
 1-02-006-02  .6

EF from print-out:   Boilers                    0.5 MLbs/10*6 CF
                    Gas Pipelines & Plants      0.5 MLbs/10*6 CF
Control

SO,:  From ref. 33c.

NOTE:  Ash and sulfur content numbers may change. Therefore, it may be necessary to change the
emission factors.

Miscellaneous Fuel

Activity

Coke: The objective is to estimate coke consumption, in tons, outside the iron and steel industry.
From the Quarterly Coal Report, obtain the following data:

  1. Total breeze production at coke plants. Assume  24 percent is sold  for use as boiler fuel.
       Multiply total breeze production by 0.24.

 2. Coke sales to "other industrial plants;" if data for foundries and other industrial plants are
       combined, assume that 49 percent of the total is for other industrial plants.

       Add 1  and 2 to obtain total coke produced from coal. Alternatively, if  1 and 2 are not
available, assume 5.75 percent of total coke production represents coke consumption outside the iron
and steel industry.

                                              A-6

-------
       From the Cost & Quality, or Electric Power Annual, obtain the total quantity of petroleum
 coke consumed  or received by power plants.

       Add the  values obtained in  1 and 2 above and petroleum coke receipts together to obtain the
 total coke consumption.

 Coke Oven Gas:  Obtain the total coke-oven gas production, in cubic feet from Quarterly Coal
 Report.  Multiply this total by 0.188. It is assumed that 18.8 percent of total coke gas produced is
 consumed outside of the iron and steel industry.  If  not published, call National Energy Information
 Center (202) 586-8800.  If not available, use previous year number.

 Bagasse: Use the number from  the previous year.

 Kerosene:  Obtain the "adjusted" quantity of kerosene sales in gallons, from Fuel Oil and Kerosene
 Sales 19xx.  Add the "adjusted"  sales figures reported for industrial and !'all other."

 LPG:  Use  LPG supplied to industrial use.  American Petroleum Institute, Jim Tsiderdaos (202) 682-
 8498, table entitled, "LPG Supply and Disposition." The objective is to project the 1982 consumption
 figures, in gallons, to the update year based on the quantity of products supplied.  The following
 equations can be used:

 Industrial  = 5,397 * 106 gal.M, * LPG Supplied
                 1,499 * 103 bbl/day                                                    (eq. 48)
 where,

 i = year

 5,397 x 106 gal = total industrial sales in 1982.

 1,499 x 103  bbl/day = products supplied in 1982 obtained from  Petroleum Supply Annual.

 LPG Supplied = products supplied  obtained from Petroleum Supply Annual

 Wood:  Obtain the consumption  figures, in tons, from Estimates of U.S. Wood  Energy Consumption,
 1980-1983.  This reference gives consumption in terms  of oven-dried equivalent weight for the
 previous  year. For example, for  the 1984 update, the reference was available for 1983.  Assume  that
 15 percent of the heating value is lost to moisture on a  typical basis.  Therefore, multiply the
reported consumption figures in tons by 0.85. Do this for industrial and residential, separately.  As
of 1990,  wood consumption  was published in therms of Btu's and an average Btu content per oven-
dried short ton was provided for  both residential and industrial sectors. The ratio is:

                                       77.2 million Btu
                                     oven-dried short  ton

No adjustment to the calculated tonnage is necessary.
                                             A-7

-------
Next, project the converted consumption figure to the update year.  Assume that 75 percent of the
industrial wood is consumed by the oulp and paper industry and 25 percent is used in lumber and
wood products. Refer to Section 3.15.3.7 [below] to obtain the production figures for the base and
update year. Then project as before (see LPG), but for pulp and paper and lumber and  wood
products, separately.  Add the projected consumption figures.

Pulp and Paper

Kraft:  Use production found in Current Industrial Reports, Pulp, Paper and Board; use production
value reported for "sulfate."

Sulfite:  Same reference  as for  Kraft; use production value reported for "sulfite."

Lumber.

       Obtain the total lumber production expressed in  million board  feet from Current  Industrial
Reports, Lumber Production and Mill Stocks.  If not available, use  Survey of Current Business.

Emission Factors

Coke:
Coke Type
Petroleum Coke
Coal Coke
LbsS/Ton
SO,
38.8
30.3
       Calculate weighted average EF's based on the total coke produced from petroleum and the
 total coke produced from coal (breeze production plus coke sales to boilers) as shown in Table 3.11-
 1.

       Use a sulfur content value of 3.25 percent for petroleum coke.

                   Table 3.11-1.  Weighted  Average Emission Factors for Coke
Coke Type
Petroleum Coke
Coal Coke"
Consumption (from
Part 3A)
554
1384
Weighted Average (metric Ibs/ton)
EF (metric Ibs/ton)
Part
1.4
4.2

SO,
114.3*
27.5

NOX
19
12.7

PM-10
1.2
9

* Assumes a constant sulfur content value of 3.25 percent for petroleum coke.
** Total of breeze production plus coke industrial boilers.
                                              A-8

-------
Coke Oven Gas:
                   LbsS/106 Cubic Feet Burned
   SCC            SO/
1-02-007-07         680.0S

"Assume a sulfur content value of 1.605 percent.
Bagasse: Table 1.8-1, AP-42 Fourth Edition, Volume I

                   LbsS/Ton
   SCC            SO,
1-02-011-01         0

'Obtain percent control efficiency from AIRS AFP650, SCC 1-02-011-01.  Currently this percent
control = 69 percent.

Kerosene:
                   LbsS/103 Gal.
   SCC            SO.
1-02-005-01         143.6S

'Assume a sulfur content value of 0.075 percent

LPG: Table 1.5-1, AP-42 Fourth Edition, Volume I

   scc            so.;
1-02-010-02         86.5S

'Assume a sulfur content value of 0.0013 percent.

Wood:  Table 1.6-1, AP-42 Fourth Edition, Volume I

   SCC            SO,
1-02-009-01         .15
1-02-009-02         .15
1-02-009-03         .15

EF from print-out:   Coke              47.8 MLbs/ton
                   Coke oven Gas      990.0 10*9 cu ft.
                   Kerosene           9.8 million gal.
                   LPG               0.1 million gal.
                   Wood              2.2 10*3 tons
                                           A-9

-------
Primary Copper

Activity

Roasting-  Obtain the primary copper smelter production from domestic and foreign ores from the
Minerals Yearbook, Copper. The Table is entitled, "Copper:  World Smelter Production, by
country"  This figure is expressed in units of blister copper produced.  Convert to short tons and
multiply the reported number by 2. (This  multiplier assumes that there  are 4 tons  of copper
concentrate/ton of blister, but only half is roasted.)

Smelting, Converting:  Same as above but instead multiply the reported number  by 4.

Fugitive:  Use the total new copper smelter production figure obtained from Minerals Yearbook,
Copper. Total primary includes domestic and foreign ores.  Convert to short tons.

Regional Fractions:  Contact directly (i.e., telephone) the State or Regional Air Quality Bureaus for
copper smelter activity within their area.

Emission Factors
 Roasting: Table 7.3-2, AP-42 Fourth Edition
    SCC
 3-03-005-02
 3-03-005-09
Description
Multi-Hearth Roaster
Fluid-Bed Roaster
LbsS/Ton
  SO,
  280
  360
        Calculate a weighted average EF based on the data in Table 3.15-9.  Multiply each EF by the
 corresponding capacity.  Add the products and divide by the total capacity.  Then add 1 Ibs/ton to
 the weighted average EF to account for fugitive emissions.

                                  Table 3.15-9.  Capacity Data
Type of Roaster
Multihearth
Fluid Bed
EF
280
360
1981 Capacity
430
230
                                              A-10

-------
 Smelting:  Table 7.3-2, AP-42 Fourth Edition
    SCC
Description
 3-03-005-07   Reverb. Furnace + Convenors
 3-03-005-03   Multi-Hearth + Reverb. Furnace + Convenors
                Fluid Bed Roaster + Reverb. Fum. + Convenors
 3-03-005-10   Electric Furnace + Convenors
                Fluid Bed + Electric Arc + Convenors
 3-03-005-26   Flash Furnace + Cleaning Furnace + Convenor
                                         90
LbsS^Ton
  SO.
  320
  180
  160
  240

  820
       Calculate a weighted average EF based on the data in Table 3.15-10.  Multiply each EF by
 the corresponding capacity.  Add the products and divide by the total capacity.  Then add 4 Ibs/ton
 to the weighted average EF to account for fugitive emissions.
                         Table 3.15-10. Smelting Emission Factor Data
Type of Process
Reverb. Furnace + Convenors
Multihearth + Reverb. Furnace + Convenors
Fluidized Bed Roasters + Reverb. Furnace +
Convenors
Electric Furnace + Convenors
Fluidized Bed Roaster + Electric Arc + Convenors
Flash Furnace, Cleaning Furnace, Convenor
EF (Ibs/ton)
320
180
160
240
90
820
1981
Capacity
405
430
212
124
18
115
Converting:  Table 7.3-2, AP-42 Fourth Edition
   SCC             Description
3-03-005-23  Reverberatory Furnace + Convenor         740
3-03-005-24  Multi-Hearth + Reverb. + Converter
3-03-005-25  Fluid Bed Roaster + Reverb. + Converter    540
3-03-005-26  Electric Arc + Converter
3-03-005-27  Flash Furn. + Cleaning Furn. + Converter    240
3-03-005-28  'Noranda Reactor + Converter              600
                                       LbsS/Ton
                                        SO.

                                        600

                                        820
      'Assumed value used for Noranda Reactor emission factor.

      Same procedure as for Roasting and Smelting except use the data in Table 3.15-11.  Add  130
Ibs/ton to the weighted average EF to account for fugitive emissions.
                                           A-ll

-------
 NOTE: For copper smelting, calculation of new weighted average emission factors is needed only if
 an existing smelter ceases operation or if a new smelter begins operation or if an existing smelter is
 modified. See Minerals Yearbook, Copper for information on such changes in capacity.

                        Table 3.15-11.  Converting Emission Factor Data
Type of Process
Reverb. Furnace + Convenors
Multihearth + Reverb. Furnace + Convenors
Fluidized Bed Roasters + Reverb. Furnace +
Convenors
Electric Arc Furnace + Convenors
Flash Furnace, Cleaning Furnace, + Convenors
Noranda Reactor + Convenors
EF (Ibs/ton)
740
600
540
820
240
600
1981
Capacity
405
448
212
124
115
231
Primary Zinc

Activity


Roasting:  Obtain the total slab zinc production from the Minerals Yearbook. Zinc.  Convert the
units to short tons and multiply by 2 to account for the fact that there are 2 units of concentrate/ton
slab zinc.

Emission Factor


Zinc Roasting: Table 7.7-1, AP-42 Fourth Edition


                    LbsSAbn

                    1,100
   SCC
3-03-030-02


EF from print out: 998.0 MT Lbs/ton

Control
Control efficiency is derived from AIRS/FS (eq.  1) for all subcategories except Zinc-Fucitive  The
control efficiency for this subcategory is obtained by best guess.
                                            A-12

-------
Primary Lead

Activity

       For all subcategories, obtain the primary refined lead production from domestic and foreign
ores from the Minerals Yearbook, Lead.  Convert the units to short tons.

Lead Processing.

       In  order to calculate a production value for lead processing the following procedure should be
used:

       1.     The total copper and zinc SO2 emissions must be calculated first.

       2.     Then  calculate SO2 Removal in H2SO4 as follows:

Byproduct Sulfuric Acid - Copper:  Obtain the quantity of byproduct sulfuric acid produced from
Copper plants in the United States (Minerals Yearbook. Copper). Multiply this total by 0.6531.  This
is the ratio of the molecular weight of SO2 (64) to the molecular weight of H2SO4 (98). Enter the
result in the TRENDSXX.xls file "Sulfur Oxide Emissions from Non-Ferrous Smelters, SO2 Removal
in H2SO4," for Copper.

Byproduct Sulfuric Acid - Lead + Zinc (SO2 lead+zinc): Add the quantity of byproduct sulfuric
acid produced from  Lead plants from Minerals Yearbook, Lead, to the quantity of byproduct sulfuric
acid produced from  Zinc plants. Again multiply this total by 0.6531.  Enter the result in the
TRENDSXX.xls file "Sulfur Oxide Emissions from Non-Ferrous Smelters, SO2 Removal in H2SO4,"
for a total of Lead+Zinc byproduct sulfuric acid produced.

       In  the event the Minerals Yearbook, Copper is unavailable, one must estimate the byproduct
sulfuric acid production from copper, lead, and zinc  processing.  The Bureau of Mines may also be
able to supply preliminary numbers for these calculations.

       3.     Then  calculate total lead SO2 emission as follows:

       In  order to calculate Total S0; Lead Emissions' the following procedure should be used:
                                              AFS
Total SO2   = SO2 (lead+zinc) - SO2(zinc) + [    Lead  * .9072]                          (eq. 68)
Lead Emissions                                Emissions

where:

SO2(lead+zinc) is the total SO2 removed in byproduct sulfuric acid production from lead +
zinc processing, calculated  above.

SO2(zinc)  = Total SO2 removed in byproduct sulfuric acid production from zinc processing.
Calculated as follows:
                                        A-13

-------
Byproduct sulfuric acid prod, from zinc plants * 0.6351                                   (eQ- 69)

AIRS/FS Lead Emissions = the total S02 emissions from lead production, SCC 303 010 **.
This is obtained from a AIRS/FS AFP650 report for the latest year available.

       4.     Once this is  done, it is easy to back calculate a lead processing value.

Calculation is as follows:


Lead Processinc = Total SO. Lead Emission * 2000                                      (eq. 70)
                       540

The value for 'Lead Processing' may now be entered into TRENDSXX.xls.

Emission Factors

Lead Roasting: Table 7.6-1, AP-42 Fourth Edition
                                                      "LbsS/Ton Lead Produced
    SCC             Description                       SO,	
3-03-010-01         Sintering                               550
3-03-010-02         Blast Furnace                     45

       'AP-42 units are based on quantity of lead produced.
       'NEDS SCC file units are based on tons  of concentrated ore produced.

EF Lead Processing from print-out:  540.0 MT Ibs/ton

Control

Control efficiency is derived  from AIRS/FS (eq. 1) for all subcategories except Lead-
Fugitive.  The control efficiency for this subcategory is obtained by best guess.

Primary Aluminum

Activity

Material Handling:  Use the total primary production figure obtained from the  Minerals
Yearbook, Aluminum.

Emissions Factors

       Obtain an average EF based on NEDS, February  1980, for Washington State only.

EF from print-out:  33.5 MT Ibs/ton
                                         A-14

-------
 Control

       Control efficiency is derived from AIRS/FS (eq. 1) for all subcategories except
 Aluminum-Fugitive. The  control efficiencies for this subcategory is obtained by best guess.

 Secondary Lead

 Activity

 Reverberatory Furnaces:  Obtain the total consumption of lead scrap  and multiply by the
 following fraction:

 Lead recovered as soft lead/Total lead recovered from scrap
 Convert to short tons.

 Blast Furnaces:  Same as for Reverberatory Furnaces except that the fraction is calculated as
 follows:

 Lead recovered as antimonial lead/Total lead recovered from scrap
 Convert to short tons.

 Emission Factors

 Table 7.11-1, AP-42 Fourth Edition

                                                     LbsS/Ton
   SCC             Description                        SO.
 3-04-004-02         Reverberatory Furnace              80
 3-04-004-03         Blast Furnace                      53

 EF from print-out:   Reverberatory Furnace      72.6 MT Ibs/ton
                    Blast Furnace              48.0 MT Ibs/ton

 Control

       Control efficiency is derived from AIRS/FS (eq. 1) for all subcategories except Lead-
 Fugitive.  The control efficiency for this subcategory is obtained by best guess.

 Pulp and Paper

Activity

 Kraft Pulp Production.

       Obtain the production  of sulfate and sulfite combined from Current Industrial Reports.
           & Board. Kraft process and Sulfite process are reported individually in this
                                         A-15
report.

-------
Kraft:  Use production found in Current Industrial Reports. Pulp. Paper and Board;  use
production value reported for "sulfate."

Sulfite:  Same reference  as for Kraft; use production value reported for "sulfite."
Emission Factors

Kraft Pulp Production:  Table 10.1-1, AP-42 Fourth Edition

                     LbsS/Ton
    sec              sex
3-07-001-04         7

       The EF is obtained by adding the EF's for Kraft and Sulfite Mills.  In the case of
Kraft use the AP-42 EF of 7 Ibs/ton (see above).  In the case of Sulfite, the EF must be
calculated as described below.

Sulfite Uncontrolled EF = 52 Ibs/ton*
Sulfite Controlled EF  = 20 Ibs/ton"

       If the sulfite paniculate control efficiency is 90 percent (taken from Table 2.25), then
assume 90 percent of production is at the controlled emission rate and  10 percent at the
uncontrolled rate.  Calculate SO2 emissions from sulfite mills and add to the emissions for
Kraft Mills (5 Ibs/ton).  Calculate the  EF as follows:

EF = (Total Emissions, Kraft + Sulfite)/Production                                   (eq. 63)

'Obsolete version  of AP-42.
EF from print-out Kraft Pulp Prod. & Sulfite:  10.3 MT Ibs/ton

Sulfuric Acid

Activity

       Obtain the total production from Current Industrial Reports.  Inorganic Chemicals.

Emission Factor

       Calculate  the EF as follows:

    EF,=

    (0.95 *  EF. , Jl_P,.,) + (0.05 * EFvCPc *_P. .) + (fP. - P.,) * EFVW)
                            P'                                                              (eq. 64


                                                 A-16

-------
 where,

 i = Year
 EFNSPS = NSPS EF (4 Ibs/ton)
 p = Total Production

        If the current year production is less than the previous year production, the last term
 (Pj - P,.,) is zero.  Only assume new capacity for production above the previous record high
 production level.

 EF from print-out:  7.8  MT Ibs/ton

 Carbon Black Production

 Activity

       Obtain the total quantity of carbon black produced from C&E News.

 Oil Process:  Assume that 90 percent of total production is by oil process.

 Gas Process:   Assume that 10 percent of total production is by gas process.

 Emission Factors

 Table 5.3-3, AP-42 Fourth Edition
                                               LbsS/Ton
       Description                               SO.
 Flared Furnace Exhaust (Oil Process)              50

       Calculate the EF as follows:

 EF = (CO Control Efficiency / 0.913) * 50 Ibs/ton                                  (eq. 73)

 where,

 CO Control Efficiency = fraction.

 EF from print-out:  20.6 MT Ibs/ton

 Sulfur Recovery Plants

Activity

       Obtain the quantity of sulfur recovered by petroleum refineries and by natural gas
 plants, respectively, from the Minerals Industry Survey, Sulfur.  Convert to  short tons.
                                          A-17

-------
Emission Factors

       Add the actual emissions reported in AIRS/FS for SCC 301-032-01 through
301-032-04.  Divide the total by the sum of the operating rates.

Petroleum Refining

Activity

       From  the Oil and Gas Journal, obtain the total capacity of catalytic cracking fresh feed
in bbl/stream day.  Convert this number  to bbl/calendar year by multiplying by 328.5 (365
days/year * 0.9 calendar day/stream day).

       Prior to 1989, it was possible to obtain from Oil and Gas Journal, the  sum of the
catalytic cracking fresh feed capacity per plant for Thermofor and Houdriflow combined with
the "other" category, as opposed to the "fluid" category, as  designated by footnotes. This
"other"  category represented TCC.  The total capacity was converted to bbl/calendar year by
multiplying by 328.5.  Then subtract this number from the total capacity above as follows:

                    Total Catalytic
FCC Capacity =     Cracking Capacity - TCC Capacity                            (eq. 64)

       Then it was necessary to convert capacity  to throughput.  From the Survey of Current
Business, in table containing information on Petroleum, Coal, and Products, obtain the
refinery operating ratio.  Divide the ratio by 100 to convert it to percent, and  multiply the
estimated capacities for FCC and TCC by the relevant number to  get an estimate of
throughput

       Since  1989, the Oil and Gas Journal no  longer gives catalytic cracking fresh feed for
Thermofor and Houdriflow.  Therefore, multiply the  total capacity of catalytic cracking fresh
feed in bbl/stream day by 328.5 and by the refinery operating ration.  Also, add the total FCC
and TCC reported in TRENDSXX.xls for the previous year. Then calculate FCC and TCC
for the update year as  follows:

       FCC,  = FCC,, * cc^m                                                    (eq. 65)
                      cc(FF),.,

       TCC, = TCC,,  *  ccim                                                     (eq. 66)
                      cc(FF),.,

Process Heaters.

Oil: Obtain the quantity of oil consumed at petroleum refineries  by PAD District from
Petroleum Supply Annual.  Obtain the total of distillate, residual and crude oil  Divide hv
1,000.                                                                    '         y
                                         A-IX

-------
     :  Obtain the total of natural gas and still (process) gas consumed at petroleum refineries
      Petroleum Supply Annual.  The quantity reported for still gas is expressed as thousand
 bbl equivalents. Multiply the reported number by 6.3 to get 106 cu. ft.

 Emission Factors
                                                      LbsS/103 BBL Fresh Feed
 ___SCC__         Description                        SO,
 3-06-002-01          Fluid Catalytic Cracking           493
 3-06-002-02          Thermal Catalytic Cracking         60
 3-06-004-01          Flares (Blowdown  System)         26.9
 1-02-004-01          Process Heaters: Oil               158.6S
 3-06-001-05          Process Heaters: Natural Gas         .6
 3-06-001-06          Process Heaters: Process Gas       950.0S

       The EF's for FCC, TCC, and Flares may be entered directly into the spreadsheet.

 Process Heaters:

 Oil: Obtain the EF from AP-42 for industrial- residual oil boilers. Estimate sulfur content
 from AIRS/FS AFP650 report for SCC 30600103.

 Gas: Weight the EF's for Natural Gas (0.6 lbs/106 cu.ft.) and Refinery Gas (356.25 lbs/106
 cu.ft) by the natural gas and refinery gas consumption obtained from Petroleum Supply
 Annual.  Convert bbl's of Refinery (Still) Gas to 106 cu. ft. by multiplying by 6.3.

 EF from print-out:          MT Ibs/ton
             FCC          447.2
             TCC           54.4
             Flares         24.4
       Process Heaters:
             Oil           5680.0
             Gas          230.1

 Iron and Steel

Activity

       Use the same numbers as in  the Megacalc Table 3.15.3.3, "Paniculate Emissions from
 the Iron and Steel Industry", on coke, sintering, and open  hearth, convert to thousand short
 tons from million short tons.  For roll and finish, obtain the total raw steel production from
 the Survey of Current Business, in the table containing information on "metals and
 manufactures".
                                         A-19

-------
 Coke.

 Byproduct:  Obtain the beehive and oven (byproduct) production figure, expressed in
 thousand short tons, from Survey of Current Business, in the table showing information on
 "Petroleum, Coal, and Products." SCC=3-03-003

 Sintering.

       Obtain the total production of pig iron from Mineral Industry Survey. Iron Ore, ref. 25
 or use the total reported in the Survey of Current Business. Convert long tons to short tons:
 divide result by 3 and enter as 3 components of sintering (windbox, discharge, and sinter-
 fugitive).

 Open Hearth.

       From Mineral Industry Surveys. Iron and Steel Scrap, obtain the total scrap and pig
 iron consumed by open  hearth, furnaces by manufacturers of pig iron  and raw steel and
 castings.  Calculate the fraction of scrap and pig iron (combined) that is consumed by the
 furnace types.  Multiply the fraction by the total raw steel production, expressed in thousand
 short tons, obtained from the Survey of Current Business, in the table showing information on
 "Metals and Manufactures." Enter the calculated values  both the "stack" and "fugitive"
 components of each furnace type. (Note:  The fraction for open hearth furnace will be used
 in several tables of the OAQPS Data File.  Once the Minerals Yearbook. Iron and Steel
 chapter, is available, values calculated above should be revised to agree with Minerals
 Yearbook final data.)

 Emission Factors

 Coking:  "Table 7.2-1, AP-42 Fourth Edition
              NED SCC and Emission Factor File
   SCC             Description
3-03-003-02         Charging
3-03-003-03         Pushing^
3-03-003-04         Quenching
3-03-003-06         Underfiring
3-03-003-08         Oven/Door Leaks
3-03-003-14         Topside Leaks

       Add the factors and divide by 0.7 to convert units.  This is based on 0.7 tons coke
produced/ton  coal consumed.

Sintering:  Divide the actual emissions reported in NEDS, February 1980, by the production
TcltC.
                                         A-20

-------
   Pfin Hearth: Same procedure as for Sintering.
 **°" and Finish: The objective is to compute fuel use by process equipment  For SO2, the
 fuels are coke oven gas and residual oil.  Multiply the quantity of fuel use by the
 corresponding AP-42 EF.  For example:

 Quantity of Coke Oven Gas * 1,091  lbs/106 cu.ft.
 Quantity of Residual Oil * 1,595 lbs/1,000 gal.

       Coke oven gas  used in iron and steel  manufacturing is calculated as follows: (1)
Obtain  total annual coke oven gas production from Quarterly Coal  Report, (2) Assume 40
percent of production is used in iron  and steel process equipment.  (3) The  EF for coke oven
gas (1,091 lbs/106 cu.ft.) is obtained from AP-42. (4)  Multiply fuel quantity by the EF

       Residual Oil  used in iron and steel manufacturing is calculated as follows:  (1)  From
the survey of Current Business, containing information on Metals and Manufactures, obtain
the quantity of raw steel production in short tons and  multiply by 0.00738 * 106 gal/10J ton
steel.  (This value should be updated for 1982 and later years based on the  1982  Census of
Manufacturers, Fuels and Electric Energy Consumed).  (2) The EF for industrial boilers
(1,595 lbs/1,000 gal) is used as calculated for the TRENDSXX.XLS file "SO^Fmissions
from Residual Oil Combustion."  (3)  Multiply fuel quantity by the  EF.  (4) Assume percent
Sulfur for industrial-residential oil boilers calculated for the table on SO2 emissions applies.

       Add emissions together to obtain total SO2 emissions. Then subtract the quantity of
emissions from Open Hearth Furnaces shown in the TRENDSXX.xls file "SO2 Emissions
from Other  Industrial Processes."  Calculate the EF as follows:

EF = Emissions left over / Roll and Finish Operating  Rate                           (eq. 74)

      EF from print-out:                MT Ibs/ton

                          Coking              10.3
                          Sintering            2.3
                          Open Hearth          1 .4
                          Roll &  Finish       3.5

 Cement Manufacturing

Activity

       Obtain the total quantity of cement production from Mineral Industry Survey. Cement.
 Use the same figure for all subcategories.
                                         A-21

-------
Emission Factors

       The first objective is to calculate total uncontrolled SO2 emissions.  These emissions
are a function of mineral sources of SO2 and sulfur in fuel used to fire kilns.  Obtain total
cement production from Minerals Industry Survey. Cement. The EF's shown in  Table 3.15-
12 are used to calculate uncontrolled emissions. Add these emissions and divide by the total
cement production rate to get the uncontrolled EF.

       At this point, it is convenient to estimate control efficiency.  The baseline value of
13.75 percent S0: control corresponds to a cement kiln paniculate control efficiency of 99
percent.  The value of 12 percent SO: control corresponds to a cement kiln control efficiency
of 92 percent.  For other  percent SO: control values, estimate the corresponding cement kiln
SO: control efficiency value by linear interpolation/extrapolation from the paniculate control
efficiency.

                Table 3.15-12.  Emission Factors for Uncontrolled  Emissions
Fuel
Mineral Source
Coal
Residual Oil
Distillate Oil
Emission Factor
10.2 Ibs/ton cement produced
30.45 Ibs/ton coal consumed1
124.5 lbs/1,000 gal residual oil consumed2
112.35 lbs/1,000 gal distillate oil consumed3
1  S = value derived for industrial boilers, Section 3.7.
2  S = value derived for industrial boilers, Section 3.8.
3  S = 0.3
EF from print-out:  14.5 MT Ibs/ton


Glass

Activity

       Refer to Current Industrial Reports, Glass Containers and Current Industrial Reports
Rat Glass.  Add the following quantities, after converting to thousands of short tons, as
necessary:
                                          A-22

-------
  1.  Total production of flat glass (in short tons)
  2.  Net packed weight of glass containers (in thousands of pounds)

       Multiply the total by 1.10 to account for miscellaneous glass products.

Emission Factors

Table 8.13-1, AP-42  Fourth Edition
sec
3-05-014-02
3-05-014-03
3-05-014-04
Description
Container Glass: Furnace
Flat Glass: Furnace
Blown Glass: Furnace
LbsS/Ton
SO.
3.4
3.0
5.6
Weighting
Factor
.75
.15
.1
       Calculate a weighted average EF based on the weighting factors in the above table.

                           Table 3.15-13. Weighting Factors
Type of Glass
Container Glass
Flat Glass
Blown Glass
Weighting Factor
0.75
0.15
0.1
EF from print-out:  3.2 MT Ibs/ton

Lime Manufacturing

Activity

       Obtain the production figure from C & E News, for lime.  Enter the obtained value for
both kilns and fugitive.

Emission Factors

       Divide the total  actual SO2 emissions reported in NEDS, February 1980, by the NEDS
 lime production rate.

 EF from print-out:  3.1 MT Ibs/ton

 Control
     vve-  Obtain a value for control efficiency by best guess.

                                           A-23

-------
                                TECHNICAL REPORT DATA
                         (Please read Instructions on the reverse before completing)
1. REPORT NO.
 EPA-600/R-94-012
                           2.
                                                      3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Comparison of the 1985 NAPAP Emissions Inventory
 with the 1985 EPA TRENDS Estimate for Industrial
 SO2 Sources
            5. REPORT DATE
             January 1994
            6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
David Zimmerman and Rebecca Battye
                                                      8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
EC/R Incorporated
University Tower, Suite 404
3101 Petty Road
Durham, North Carolina 27707
                                                       10. PROGRAM ELEMENT NO.
            11. CONTRACT/GRANT NO.
             68-D2-0181,
              (TRC)
Tasks 2 and 8
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Air and Energy Engineering Research Laboratory
 Research Triangle Park, NC  27711
            13. TYPE OF REPORT AND PERIOD COVERED
             Task Final;  3-10/93
            14. SPONSORING AGENCY CODE
              EPA/600/13
 15.SUPPLEMENTARY NOTES AEERL    -ectofficer ig Charles c>  Masser,  Mail Drop 62, 9197
 541-7586.
 IB. ABSTRACT The report'g^gg results Qf analyses of 1985 industrial sulfur dioxide (SO2)
 emissions from two data sources: the National Acid Precipitation Assessment Pro-
 gram (NAPAP) inventory and the EPA TRENDS report. These analyses conclude that
 the two data sources estimate comparable emissions  in the aggregate, but that esti-
 mates for specific categories and for processes within those categories vary widely.
 The TRENDS method,  limited to source categories that emit 10,000 tonnes of SO2 per
 year, generally overestimates emissions from these  source categories,  due prima-
 rily to the absence  of SO2 control efficiency assumptions.  Overestimating emissions
 in the TRENDS data set is offset by including additional source categories in the NA-
 PAP inventory,  with the final aggregate estimates within <10% of each other. Due to
 these findings, the  TRENDS methodologies are being  revised for 1993 and thereafter,
 using the 1985 NAPAP inventory as a base. (NOTE: The  1990 Clean Air Act Amend-
 ments (CAAA) require that EPA report to Congress by 1995 a national inventory of
 annual SO2 emissions from industrial sources, and emission projections for the next
 20 years.  This stems from the  5.6 million tons of industrial SO2 emissions cited in
 CAAA Title IV,  and based on the estimated 1985 NAPAP emissions.)
 7.
                             KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
                                           b.IDENTIFIERS/OPEN ENDED TERMS
                         c. COSATI Field/Group
Pollution
Sulfur Dioxide
Emission
Inventories
Pollution Control
Stationary Sources
NAPAP
TRENDS
13 B
07B
14G
15E
 8. DISTRIBUTION STATEMENT
 Release to Public
                                           19. SECURITY CLASS (This Report)
                                           Unclassified
                                                                    21. NO. OF PAGES
                             202
20. SECURITY CLASS (Thispage)
Unclassified
                         22. PRICE
EPA Form 2220-1 (9-73)
                                         A-24

-------