Toxic Release Inventory Data Quality Report 1994-1995


HOI :n :•••;•
 i'.i
 1

 !»
 i1' •
ilP
          oixoj. 966 V P"B 17661    Vd3^
            (80W)
                   AouaBy
     soixoi pue
       uoiin||

-------
 fc                                  TABLE OF CONTENTS

rO
^                   OVERVIEW 	 i

        1.0          INTRODUCTION 	1-1
                     1.1    EPA's Overall Quality Assurance Program	1-2
                     1.2    Site Survey Objectives	1-2
                     1.3    EPA Site Surveys	1-3

        2.0          APPROACH	2-1
                     2.1    Survey Instrument  	2-3
                     2.2    Sample Selection	2-3
                           2.2.1   RY 1987 and RY 1988 Sample Selection	2-4
                           2.2.2  RY 1994 and RY 1995 Sample Selection	2-6
                     2.3    Site Surveyor Selection and Training 	2-6
                     2.4    Arranging Site Visits	2-7
                     2.5    Conducting Site Visits	2-8
                           2.5.1   Data Collection  	2-8
                           2.5.2  Threshold Determinations	2-9
                           2.5.3  Release and Other Waste Management Estimates	2-10
                     2.6    Data Management/Data Quality Assurance	2-10
                           2.6.1   Quality Review of Survey Instrument and Data Entry  	2-11
                           2.6.2  Data Weighting   	2-11
                           2.6.3  Limitations of the Analysis 	2-12
                     2.7    Data Analysis and Reporting	2-13

        3.0          THRESHOLD DETERMINATIONS  	3-1
                     3.1    Approaches Used for Determining Thresholds	3-1
                     3.2    Frequency of Errors Made When Determining Thresholds	3-7
                     3.3    Sources of Errors Made When Determining Thresholds	3-13
                           3.3.1   Reasons Why Facilities Failed to Submit Form Rs for
                                  Chemicals That Exceeded Thresholds	3-13
                           3.3.2  Reasons Why Facilities Submitted Form Rs for Chemicals
                                  That Did Not Exceed Thresholds	3-15
                           3.3.3  Chemical Activity Classification	3-18
                           3.3.4  Impact of Not  Calculating Thresholds	3-21
                     3.4    Lessons Learned	3-21

        4.0          SOURCES AND TYPES OF RELEASES, OFF-SITE TRANSFERS, AND ON-SITE
                     WASTE MANAGEMENT ACTIVITIES	4-1
                     4.1    Observed Release and Other Waste Management Activity Sources
                           and Types	4-1
                     4.2    Incorrectly Reported Release and Other Waste Management Activity
                           Types  	4-13
                     4.3    Overlooked Releases and Other Waste Management Activities	4-17

-------
                       TABLE OF CONTENTS (Continued)

             4.4   Calculation Methodologies 	4-21
             4.5   Releases to On-site Waste Management Activities (recycling,
                   treatment, and energy recovery)	4-45

5.0          RELEASES AND OTHER WASTE MANAGEMENT ACTIVITIES	5-1
             5.1   On-Site Release and Transfers Off-Site for Waste Management
                   Estimates  	5-1
                   5.1.1   Overview of On-Site Releases and Transfers Off-Site for Waste
                          Management as Reported by Facilities and by Site Surveyors  . 5-2
                   5.1.2   Comparison of RY 1994 and RY 1995 On-Site Releases and
                          Transfers Off-Site for Waste Management to RY 1987 and
                          RY 1988 On-Site Releases and Transfers Off-Site for
                          Waste Management	5-15
                   5.1.3   Analysis of Specific Releases  	5-24
             5.2   On-Site Waste Management Activities	5-44
                   5.2.1   On-Site Waste Management Activities by SIC Code 	5-44
                   5.2.2   On-Site Recycling 	5-48
                   5.2.3   On-Site Treatment	5-50
                   5.2.4   On-Site Energy Recovery 	5-50
             5.3   Production Ratio/Activity Index 	5-54
             5.4   Source Reduction Activities 	5-59
                   5.4.1   Errors in Classifying Source Reduction	5-59
                   5.4.2   Sources of Errors Made When Claiming Source Reduction  .. 5-60
                   5.4.3   Feedback from Facilities 	5-61
                   5.4.4   Overall Accuracy of Source Reduction Data	5-62

6.0          PREPARATION OF THE FORM R	6-1
             6.1   Facility Personnel and References	6-1
             6.2   Amount Of Time Needed To Prepare Form R Reports	6-7
             6.3   Use of the Hotline  	6-13
             6.4   Comments on the Form R Instructions 	6-13
             6.5   Comments on the Automated Form R (AFR) 	6-16
             6.6   Comments on Use of the Form A 	6-18

7.0          CONCLUSIONS	7-1

8.0          RECOMMENDATIONS	8-1
             8.1   Additional Guidance Concerning Form R Instruction and
                   Documentation	8-1
             8.2   Additional Guidance Concerning Threshold Determinations	8-3
             8.3   Additional Guidance Concerning Release Estimates	8-3

-------
                                LIST OF TABLES
Table 2-1


Table 3-1

Table 3-2

Table 3-3


Table 3-4


Table 3-5


Table 4-1


Table 4-2


Table 4-3


Table 4-4

Table 4-5


Table 4-6

Table 4-7

Table 4-8

Table 4-9

Table 4-10

Table 4-11
Distribution of Facilities Among the SIC Codes
For Each Year of the Data Quality Site Visits	2-5

Approaches Used by Facilities to Make Threshold Determinations	3-3

Accuracy of Threshold Determinations by Reporting Year and SIC Code	3-3

Reasons Why Facilities Failed to Submit Form Rs for Chemicals That
Exceeded Thresholds	3-14
Reasons Why Facilities Submitted Form Rs for Chemicals That
Did Not Exceed Thresholds	
3-16
Comparison of Chemical Activity Classifications Made by Reviewers
to Those Made by Facilities, by Reporting Year and SIC Code	3-19

Distribution of Release and Other Waste Management
Activity Sources, RY 1994 and RY 1995  	4-2

Incorrectly Identified Release or Other Waste Management Activity
Types, RY 1994 and RY 1995	4-15

Overlooked Release and Other Waste Management Activity
 Sources, RY 1994 and RY 1995	4-18

Distribution of Calculation Methodologies, RY 1994 and RY 1995	4-23

Types of Emission Factors Used for Fugitive and Stack Releases
RY 1994 and RY 1995	4-42

Observed On-Site Recycling Activities (SIC Code 281)	4-48

Observed On-Site Recycling Activities (SIC Code 285)	4-48

Observed On-Site Recycling Activities (SIC Code 25)	4-49

Observed On-Site Recycling Activities (SIC Code 30)	4-49

Observed On-Site Recycling Activities (SIC Code 286)	4-50

Chemicals For Which Recycling Was Claimed (SIC Codes 281, 285,
25, 30, and 286 Combined) 	4-51

-------
                         LIST OF TABLES (Continued)

Table 5-1     Summary of SIC Code 25 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for Reporting Year 1994 (millions of Ibs.)	5-3

Table 5-2     1994 Reported TRI On-Site Releases and Transfers Off-Site for Waste
             Management for SIC Code 25 (millions of Ibs.)	5-4

Table 5-3     Summary of SIC Code 281 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for RY 1994 (millions of Ibs.)	5-5

Table 5-4     1994 Reported TRI On-Site Releases and Transfers Off-Site for Waste
             Management for SIC Code 281 (millions of Ibs.)	5-6

Table 5-5     Summary of SIC Code 285 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for RY 1994 (millions of Ibs.)	5-7

Table 5-6     1994 Reported TRI On-Site Releases and Transfers Off-Site for Waste
             Management for SIC Code 285 (millions of Ibs.)	5-8

Table 5-7     Summary of SIC Code 30 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for RY 1994 (millions of Ibs.)	5-9

Table 5-8     1994 Reported TRI On-Site Releases and Transfers Off-Site for Waste
             Management for SIC Code 30 (millions of Ibs.)	5-10

Table 5-9     Summary of SIC Code 26 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for RY 1995 (millions of Ibs.)	5-11

Table 5-10    1995 Reported TRI On-Site Releases and Transfers Off-Site for Waste
             Management for SIC Code 26 (millions of Ibs.)	5-12

Table 5-11    Summary of SIC Code 286 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for RY 1995 (millions of Ibs.)	5-13

Table 5-12    1995 Reported TRI On-Site Releases and Transfers Off-Site for Waste
             Management for SIC Code 286  (millions of Ibs.)	5-14

Table 5-13    Summary of RY 1995 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for SIC Codes 26 and 286 (millions of Ibs.)	5-20

Table 5-14    Summary of RY 1994 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for SIC Codes 25, 281, 285, and 30 (millions of Ibs.) .. 5-20

Table 5-15    Summary of RY 1988 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for SIC Codes 28,291, and 34 Through 38
             (millions of Ibs.)	5-21

-------
                         LIST OF TABLES (Continued)

Table 5-16    Summary of RY 1987 TRI On-Site Releases and Transfers Off-Site for
             Waste Management for SIC Codes 20 Through 39 (millions of Ibs.)	5-21

Table 5-17    Percent Difference of Facility Estimated and Site Surveyor Estimated
             Total TRI On-Site Releases and Transfers Off-Site for Waste Management
             for RY 1995, RY 1994, RY 1988, and RY 1987 (millions of Ibs.)	5-24

Table 5-18    Comparison of the Percent Difference (PD) Between Facility and Site
             Surveyor Estimates for Fugitive Air Emissions	5-25

Table 5-19    Comparison of the Percent Difference (PD) Between Facility and Site
             Survey Estimates for Stack Air Emissions  	5-27

Table 5-20    Comparison of the Percent Difference (PD) Between Facility and Site
             Surveyor Estimates to Receiving Streams	5-29

Table 5-21    Comparison of the Percent Difference (PD) Between Facility and Site
             Surveyor Estimates for Underground Injection	5-31

Table 5-22    Comparison of the Percent Difference (PD) Between Facility and Site
             Surveyor Release Estimates to Land On-Site  	5-32

Table 5-23    Comparison of the Percent Difference (PD) Between Facility and Site
             Surveyor Estimates for Discharges to POTWs  	5-34

Table 5-24    Comparison of the Percent Difference (PD) Between Facility Transfer
             Estimates and Site Surveyor Transfer Estimates for Off-Site Disposal	5-36

Table 5-25    Comparison of the Percent Difference (PD) Between Facility Transfer
             Estimates and Site Surveyor Transfer Estimates for Off-Site Treatment .... 5-38

Table 5-26    Comparison of the Percent Difference (PD) Between Facility Transfer
             Estimates and Site Surveyor Transfer Estimates for Off-Site Recycling .... 5-40

Table 5-27    Comparison of the Percent Difference (PD) Between Facility Transfer
             Estimates and Site Surveyor Transfer Estimates for Off-Site Energy
             Recovery  	5-42

Table 5-28    Summary of SIC Code 25 TRI On-site Waste Management Activity
             Quantities for RY 1994 (millions of Ibs.) 	5-45

Table 5-29    Summary of SIC Code 281 TRI On-site Waste Management Activity
             Quantities for RY 1994 (millions of Ibs.) 	5-45

-------
                         LIST OF TABLES (Continued)

Table 5-30    Summary of SIC Code 285 TRI On-site Waste Management Activity
             Quantities for RY 1994 (millions of Ibs.) 	5-46

Table 5-31    Summary of SIC Code 30 TRI On-site Waste Management Activity
             Quantities for RY 1994 (millions of Ibs.) 	5-46

Table 5-32    Summary of SIC Code 26 TRI On-site Waste Management Activity
             Quantities for RY 1994 (millions of Ibs.) 	5-47

Table 5-33    Summary of SIC Code 27 TRI On-site Waste Management Activity
             Quantities for RY 1995 (millions of Ibs.) 	5-47

Table 5-34    Comparison of the Percent Difference (PD) Between Facility On-Site
             Recycling Estimates and Site Surveyor On-Site Recycling Estimates 	5-48

Table 5-35    Comparison of the Percent Difference (PD) Between Facility On-Site
             Treatment Estimates and Site Surveyor On-Site Treatment Estimates	5-50

Table 5-36    Comparison of the Percent Difference (PD) Between Facility On-Site
             Energy Recovery Estimates and Site Surveyor On-Site Energy Recovery
             Estimates  	5-52

Table 5-37    Method of Estimate Used by Facilities to Calculate Production Ratio	5-55

Table 5-38    Percent of Time Surveyor Agreed with Facility Basis of Production Ratio
             Estimate	5-57

Table 5-39    Percent of Time Surveyor Agreed with by Facilities to Calculate Production
             Ratio	5-58

Table 5-40    Errors in Source Reduction Activity Classifications	5-61

Table 5-41    Most Common Source Reduction Activities Claimed by the Selected
             Facilities	5-62

Table 6-1     Number of Employees at Visited Facilities	6-2

Table 6-2     Types of Personnel  Completing the Form R	6-3

Table 6-3     Common References Used to Compile the Form Rs	6-4

Table 6-4     Number of Hours Required to Complete all the Form Rs for RY 1988 	6-8

Table 6-5     Number of Hours Required to Complete all the Form Rs for RY 1994 	6-8

-------
                        LIST OF TABLES (Continued)

Table 6-6    Number of Hours Required to Complete all the Form Rs for RY 1995 	6-11

Table 6-7    Average Number of Hours Needed to Complete A Form R	6-11

Table 6-8    Comments on the Form R Chemical Specific Instructions for RY 1994  ....  6-14

Table 8-1    Recommendations for Avoiding Errors in Threshold Determinations 	8-4

Table 8-2    Recommendations for Avoiding Errors in Identifying Release and Other
            Waste Management Activity Types and Sources  	8-5

-------

-------
                               LIST OF FIGURES


Figure 2-1    Approach used to Perform the EPCRA Section 313 Site Visit Program	2-2

Figure 3-1    Approaches Used by Facilities to Make Threshold Determinations
             forRY 1987	3-4

Figure 3-2    Approaches Used by Facilities to Make Threshold Determinations
             forRY 1994	3-5

Figure 3-3    Approaches Used by Facilities to Make Threshold Determinations
             forRY 1995	3-6

Figure 3-4    Accuracy of Threshold Determinations by Reporting Year and SIC Code ... 3-10

Figure 3-5    Accuracy of Threshold Determinations for RY 1987 and RY 1988	3-11

Figure 3-6    Accuracy of Threshold Determinations for RY 1994 and RY 1995	3-12

Figure 3-7    Comparison of Chemical Activity Classifications Made by Facilities to
             Those Made by Reviewers	3-20

Figure 3-8    Frequency with which Facilities Calculated Thresholds for Section 313
             Chemicals	3-22

Figure 4-1 a   Distribution of Release Type or Other Waste Management Activity,
             RY 1994 and RY 1995	4-5

Figure 4-lb   Distribution of Sources for Fugitive Releases, RY 1994 and RY 1995  	4-6

Figure 4-lc   Distribution of Sources for Stack Releases, RY 1994 and RY 1995	4-7

Figure 4-1 d   Distribution of Sources for Receiving Stream Releases, RY 1994
             and RY 1995  	4-8

Figure 4-1 e   Distribution of Sources for Underground Inj ection, RY 1994
             and RY 1995  	4-9

Figure 4-If   Distribution of Sources for Land On-Site, RY 1994 and RY 1995	4-10

Figure 4-lg   Distribution of Sources for POTW Transfers, RY 1994 and RY 1995	4-11

Figure 4-lh   Distribution of Sources for Off-Site Transfer, RY 1994 and RY 1995	4-12

-------
                        LIST OF FIGURES (Continued)

Figure 4-2    Incorrectly Identified Releases and Other Waste Management Activity
             Types, RY 1994 and RY 1995	4-16

Figure 4-3    Overlooked Releases and Other Waste Management Activity Sources,
             RY 1994 and RY 1995	4-19

Figure 4-4a   Distribution of Calculation Methodologies (Fugitive), RY 1994
             and RY 1995  	4-26

Figure 4-4b   Distribution of Calculation Methodologies (Stack), RY 1994
             and RY 1995  	4-27

Figure 4-4c   Distribution of Calculation Methodologies (Receiving Stream), RY 1994
             and RY 1995  	4-28

Figure 4-4d   Distribution of Calculation Methodologies (Underground Injection), RY 1994
             and RY 1995  	4-29

Figure 4-4e   Distribution of Calculation Methodologies (Land On-Site), RY 1994
             and RY 1995  	4-30

Figure 4-4f   Distribution of Calculation Methodologies (POTW), RY 1994
             and RY 1995  	4-31

Figure 4-4g   Distribution of Calculation Methodologies (To Off-Site Disposal), RY 1994
             and RY 1995  	4-32

Figure 4-4h   Distribution of Calculation Methodologies (To Off-Site Treatment), RY 1994
             and RY  1995  	4-33

Figure 4-4i   Distribution of Calculation Methodologies (To Off-Site Recycle), RY 1994
             and RY  1995  	4-34

Figure 4-4j   Distribution of Calculation Methodologies (To Off-Site Recovery), RY 1994
             and RY 1995  	4-35

Figure 4-4k   Distribution of Calculation Methodologies (To On-Site Treatment), RY 1994
             and RY 1995  	4-36

Figure 4-41   Distribution of Calculation Methodologies (To On-Site Energy Recovery),
              RY 1994 and RY 1995 	4-37

Figure 4-4m  Distribution of Calculation Methodologies (To On-Site Treatment), RY 1994
             and RY 1995 	4-38

-------
                         LIST OF FIGURES (Continued)

Figure 4-5     Frequency the Facility Used the Best Methodology to Estimate Releases
              and Other Waste Management Activities 	4-40

Figure 4-6a    Type of Emission Factors Used (Fugitive) 	4-43

Figure 4-6b    Type of Emission Factors Used (Stack)	4-44

Figure 4-7     Facilities Incorrectly Reporting the Quantity Sent to Treatment Rather than
              Actually Treated (Weighted)	4-46

Figure 5-1     Comparison of Facility and Site Surveyor Estimates of Total On-Site
              Releases and Transfers Off-Site for Waste Management  	5-16

Figure 5-2     Comparison of Estimates of Total On-Site Releases and Transfers
              Off-Site for Waste Management in SIC Codes 25 and 281 Surveyed for
              Reporting Year 1994  	5-17

Figure 5-3     Comparison of Estimates of Total On-Site Releases and Transfers
              Off-Site for Waste Management in SIC Codes 285 and 30 Surveyed for
              Reporting Year 1994  	5-18

Figure 5-4     Comparison of Estimates of Total On-Site Releases and Transfers
              Off-Site for Waste Management in SIC Codes 26 and 286 Surveyed for
              Reporting Year  1995  	5-19

Figure 5-5     Comparison of On-Site Releases and Transfers Off-Site for Waste
              Management from Reporting Years 1995 and 1994 	5-22

Figure 5-6     Comparison of On-Site Releases and Transfers Off-Site for Waste
              Management from Reporting Years 1988 and 1987 	5-23

Figure 5-7     Comparison of Facility Estimates and  Site Surveyor Estimates for Fugitive
              Air Emissions	5-26

Figure 5-8     Comparison of Facility Estimates and  Site Surveyor Estimates for Stack
              Air Emissions	5-28

Figure 5-9     Comparison of Facility Estimates and  Site Surveyor Estimates for Receiving
              Streams	5-30

Figure 5-10    Comparison of Facility Estimates and  Site Surveyor Estimates to Land
              On-Site	5.33

Figure 5-11    Comparison  of Facility Estimates and  Site Surveyor Estimates for Discharges
             to POTW  	5.35

-------
                        LIST OF FIGURES (Continued)

Figure 5-12   Comparison of Facility Estimates and Site Surveyor Estimates for Transfers
             Off-Site for Disposal 	5-17

Figure 5-13   Comparison of Facility Estimates and Site Surveyor Estimates for Transfers
             Off-Site for Treatment	5-39

Figure 5-14   Comparison of Facility Estimates and Site Surveyor Estimates for Off-Site
             Recycling	5-41

Figure 5-15   Comparison of Facility Estimates and Site Surveyor Estimates for Off-Site
             Recovery  	5-43

Figure 5-16   Comparison of Facility Estimates and Site Surveyor Estimates for On-Site
             Recycling	5-49

Figure 5-17   Comparison of Facility Estimates and Site Surveyor Estimates for On-Site
             Treatment	5-51

Figure 5-18   Comparison of Facility Estimates and Site Surveyor Estimates for On-Site
             Energy Recovery	5-53

Figure 5-19   Method of Estimate used by Facilities to Calculate PA/AI	5-56

Figure 6-1    Common References Used to Compile Form Rs for RY 1987 and RY 1988  .. 6-5

Figure 6-2    Common References Used to Compile Form Rs for RY 1994 and RY 1995  .. 6-6

Figure 6-3    Time Needed to Complete all Form Rs in RY 1988 for SIC Codes 28 and
             291, and 34 - 38	6-9

Figure 6-4    Time Needed to Complete all Form Rs in RY 1994 for SIC Codes 28,
             25, and 30	6-10

Figure 6-5    Time Needed to Complete Form R in RY 1995 for SIC Codes 26 and 286 .  . 6-12

Figure 6-6    Percent of Facilities Calling the Hotline by Industry for RY 1994 and
             RY 1995	6-17

Figure 6-7    Percent of Facilities using the Automated Form R for RY 1994 and
             RY 1995	6-17

-------
                                    OVERVIEW

             As part of a continuing effort to assess and to improve the quality of the data
contained in the Toxic Release Inventory (TRI) database, the U.S. Environmental Protection
Agency (EPA) conducted TRI data quality site surveys for the reporting years 1987,1988,1994,
and 1995. The goals for these site surveys were to identify areas in the TRI data collection
process that could be improved, to provide a quantitative assessment of the accuracy of the data
collected, and to disseminate further guidance on the completion of the TRI forms. The figures
in this overview present the significant findings from the site surveys conducted.

             This report focuses on surveys completed for reporting years (RY) 1994 and
1995, as previous reports have presented findings from RY 1987 and RY 1988.  Site surveys
were completed at the following facilities:

                    25 facilities in SIC Code 25, furniture manufacturing, for RY 1994;
             •      19 facilities in SIC Code 281, inorganic manufacturing, for RY 1994;
                    17 facilities in SIC Code 285, paint manufacturing, for RY 1994;
             •      23 facilities in SIC Code 30, rubber and plastics manufacturing, for RY
                    1994;
             •      10 facilities in SIC Code 26, pulp and paper manufacturing, for RY 1995;
                    and
             •      10 facilities in SIC Code 286, organic chemical manufacturing, for RY
                    1995.

-------
      Approaches Used by Facilities to Make Threshold Determinations for RY 1994
10000--,
•a
0)
| 9000-
w
a 8000-
_c
o> 7000-
•a
re
S
» 6000-
.2 o
re S
£ S 5000-
c Q.
= 5-
^ <
^ 4000-
5
1 300°
4)
£ 2000
•s
£ 1000
S
Q.
000

*,"'
!s^t
*•%*
*-^
->_'
«r*S


v '
f^C

1
^
•H
U*
• .,
|8
^j"
y


^1
.

-
I
Data for this figure can be found on Table 3-1
                           Approaches Used To Determine Thresholds
                    Facilities primarily use purchasing records to make threshold
                    determinations.


                    Facilities in chemical manufacturing (SIC Code 281 - Inorganic
                    Chemicals) use production data more frequently.

                    Facilities in chemical manufacturing are more likely to assume thresholds
                    are exceeded.
                                           11

-------
 Approach Used by Facilities to
Make Threshold Determinations
                                                                 for RY 1995
100%
                         Appr
                            ,,cheS Us^o Define
Dala for this figure can
           t,e found on Table 3-1
                     Facilities primarily use purchasing
                                                                       -
                                    data more
                      are exceeded.
                                                111

-------
          Accuracy of Threshold Determinations by Reporting Year and SIC Code
                                                                                 Facility did not Submit a Form-R for a
                                                                              Chemical that did not Exeed a Threshold (Correct)

                                                                              Facility Submitted a Form-R fora
                                                                           Chemical that Exceeded a Threshold (Correct)
                                                                             Facility Submitted a Form-R for a
                                                                         Chemical that did not Exceed a Threshold (Incorrect)
                                                                        Facility did not Submit a Form-R for a
                                                                     Chemical that Exeeded a Threshold (Incorrect)
                             Reporting Year/SIC
Note: The first tw o outcomes represent cases w here facilities incorrectly determned thresholds, w rule the last two outcomes represent cases w here
facilities correctly determned thresholds Rgure 3-5 compares the correct and incorrect threshold determinations by reporting year and SIC Code
Dara for this figure can be found on Table 3-2.
                        Facilities generally determine thresholds correctly over 90 percent of the
                        time.

                        Errors are generally evenly split between failing to report chemicals that
                        exceed thresholds, and reporting on those that do not.

                        Facilities in inorganic and organic chemical manufacturing (SIC Code 281
                        and 286) had the highest error rate, primarily for reporting for chemicals
                        that don't exceed thresholds.  This may be related to tendency in these
                        industries to assume thresholds are exceeded.
                                                    IV

-------
          Distribution of Release and Other Waste Management Activity Types,
                                 RY 1994 and RY 1995
          100%-
                               Source
Data for this figure can be found on Table 4-1.
                    Fugitive and stack releases and off-site transfers were observed at most
                    facilities in all industry sectors.

                    Some facilities in all industry sectors reported releases to POTWs.

                    Most facilities in pulp and paper manufacturing reported releases to
                    receiving streams.

-------
      Incorrectly Identified Release and Other Waste Management Activity Types
                               for RY 1994 and RY 1995
                                     Type
Data for this figure can be found in Table 4-2.
                    Facilities often correctly identified releases and other waste management
                    activities, but reported them to the wrong type (particularly between stack
                    vs fugitive and between various off-site transfers).

                    Quantities transferred to POTWs were correctly identified by most
                    facilities.

                    Releases to receiving streams and underground injection wells, and on-site
                    waste management activities were rarely observed; therefore, they were
                    rarely reported incorrectly.
                                            VI

-------
           Overlooked Releases and Other Waste Management Activity Sources
                                for RY 1994 and RY 1995
                                       Source
Data for this figure can be found on Table 4-3.
                                                                            SIC Code 286
                                                                           SIC Code 26
                                                                         SC Code 281
                                                                       SIC Code 285

                                                                      SIC Code 25

                                                                    SIC Code 30
                    Container residue was the most commonly overlooked release source.

                    Some facilities in all industry sectors overlooked releases from container
                    residue; pumps, valves, and flanges; and volatilization from process areas.

                    A significant number of facilities also overlooked releases from storage
                    tanks.
                                           vn

-------
Frequency the Facility Used the Best Available Methodology to Estimate Releases
                    and Other Waste Management Activities
    100%
    90%	
    80%
    70% ._
    60%
    50%
    40% _
    30% - _.
    20%
    10%
          SIC Code 25    SIC Code 281   SIC Code 285    SIC Code 30    SIC Code 26    SIC Code 286
                Most facilities in all industry sectors (greater than 80%) used an
                appropriate methodology to most accurately estimate releases.

                Note that this chart presents data on methodologies, and does not represent
                errors made in quantifying the releases.
                                       vin

-------
          Comparison of Facility and Site Surveyor Estimates of Total Releases
                         and Other Waste Management Activities
          SC Code 25      SIC Code 281      SIC Code 285      SIC Code 30      SC Code 26      SIC Code 286
           (W 1994)       (Rf 1994)       (FY 1994)       (RV 1994)       (Rf 1995)       (RY 1995)
                                       Industry Sector
Date for this figure can be found on Tables 5-1 through 5-12
                     Facility and site surveyor release estimates were in good agreement,
                     calculated to be within ±3% for most SIC Codes.

                     Facilities in SIC Code 286, the organic chemical manufacturing industry,
                     tended to be larger than those in the other SIC Codes surveyed, and had
                     more quantities released and other waste management activities.
                                             IX

-------
        Common References Used to Compile From Rs for RY 1994 and RY 1995
                                                                       SIC Code 286 (RY 1995)

                                                                     SICCode26(RY 1985)

                                                                  SIC Code 285 (RY 1994)

                                                                SC Code 281 (RY 1994)

                                                             SIC Code 30 (RY 1994)


                                                          SIC Code 25 (RY 1994)
                                                                      Note The Emergency Planning and
                                                                      Community Right-to-KnowAct,
                                                                      Section 313. and the Toxic Chemical
                                                                      Release Reporting Proposed Rule
                                                                      not listed in questic- nnatre to r RY
                                                                      1994
Data for this figure can be found on Table 6-3.
                     Most facilities surveyed for RY 1994 and RY 1995 use the "Toxic
                     Chemical Release Inventory Reporting Form R and Instructions" as their
                     main reference in compiling Form Rs.

                     Most facilities in SIC Code 26, the pulp  and paper manufacturing industry,
                     use trade association and NCASI guidance to complete Form Rs.  Trade
                     association guidance for this SIC Code is readily available and more
                     detailed than the guidance generally available for facilities in other SIC
                     Codes.

                     Many facilities in SIC Code 286, organic chemical manufacturing, use
                     AP-42 to estimate fugitive and stack releases.

-------
                   Time Needed to Complete all Form Rs in RY 1988 for
                             SIC Codes 28 and 291, and 34 - 38
                                                                                 SC Code 28,291
                         9-24
                                    25-40
                         Time Estimate (Hours)
                                                                           SC Code 34-38
41-160
                                                              >160
                   May not add up to 100% because not all facilities reported the time estimate
Data for this figure can be found in Table 6-4
                     The majority of the facilities in SIC Codes 34-38 take less than 24 hours to
                     complete all Form Rs.

                     Many facilities in SIC Code 28 are large and have many Form Rs to
                     complete. Thus,  the total time to complete all Form Rs at these facilities is
                     more than that of other SIC Codes.
                                             XI

-------
                  Time Needed to Complete all Form Rs in RY 1994 for
                              SIC Codes 25, 281, 285, and 30
               <20
                                          100-200
   21-50
            51-100

     Time Estimate (Hours)

ivfay not add up to 100% because not all facilities reported the timg estirrate
                                                     >200
                                                                                SIC Code 30
                                                                           SIC Code 285
                                                                      SC Code 281
                                                                SIC Code 25
Data for this figure can be found in Table 6-5
                     The majority of the facilities in SIC Codes 25, 281, 285, and 30 take less
                     than 50 hours to complete all Form Rs.

                     Using the maximum of hours in the lowest range checked, the average
                     number of hours needed to complete each Form R in RY 1994 is 11.7
                     hours.
                                             xn

-------
       Time Needed to Complete all Form Rs in RY 1995 for SIC Codes 26 and 286
                                                                                 SIC COde 266
             <8
                        9-20
                                                                           SIC Code 26
                                    21-40
                          Time Estimate (Hours)
                      Nby not add up to 100% because not all facilities reported the time estimate
                                                              >100
Data for this figure can be found in Table 6-6.
                    Facilities in SIC Code 26 tended to be smaller in size and had fewer
                    chemicals, and thus, took less time filling out Form Rs for all chemicals
                    than those facilities in SIC Code 286.

                    Using the maximum number of hours in the lowest range checked, the
                    average number of hours needed to complete each Form R in RY 1995 is
                    9.0 hours.
                                           xin

-------
      Percent of Facilities Calling the EPCRA Hotline by Industry
                       for RY 1994 and RY 1995
70%
       SIC Code 25
        (RY 1994)
SIC Code 286
 (RY1995)
                                   Industry
            Facilities in SIC Code 25, the furniture manufacturing industry, and in SIC
            Code 30, the rubber and plastics industry, called the hotline less than
            facilities in the other SIC Codes surveyed.
                                   xiv

-------
Percent of Facilities Using the Automated Form R for RY 1994 and RY 1995
       Did not
      Use APR
        46%
Used APR (found it
    helpful)
     51%
                    Used APR (found it
                      not helpful)
                         3%
             Approximately half of the facilities surveyed used the automated Form R.

             Of the facilities who used the automated Form R, most found it helped to
             reduce reporting errors.
                                   xv

-------

-------
1.0 INTRODUCTION

Section 313 of the Emergency Planning and Community Right-to-Know Act
(EPCRA) required the U.S. Environmental Protection Agency (EPA) to collect information
concerning releases and other waste management activities of toxic chemicals to the environment
from manufacturers, processors, and users of listed toxic chemicals. In order to collect such
information, EPA implements a yearly reporting requirement from such facilities. Reports
referred to as Form R chemical reports are due to EPA by July 1 each year to fulfill the reporting
requirement for the previous calendar year. The reporting requirement was first implemented for
the 1987 calendar year. The study discussed in this report reviewed data from the 1994 and 1995
reporting years (RY 1994 and RY 1995, respectively). Data for Standard Industrial
Classification (SIC) Codes 25, 281, 285, and 30 were reviewed for RY 1994 and data for SIC
Codes 286 and 26 were reviewed for RY 1995. SIC Codes 25, 281, 285, 30, 286, and 26
represent furniture manufacturing, inorganic chemical manufacturing, paint manufacturing,
rubber and plastics manufacturing, organic chemical manufacturing, and pulp and paper
manufacturing, respectively. This report also compares data for RY 1994 and RY 1995 to data
from similar studies completed for the 1987 and 1988 reporting years. The data from the Form R
chemical reports are compiled in EPA's Toxics Release Inventory database (TRIS) for future
analysis, distribution, and evaluation. The information collected under EPCRA, Section 313 is
useful for informing the general public and the communities surrounding affected facilities of
releases and other waste management activities of toxic chemicals, assisting in focusing the
Agency's research into the effects and control of toxic substances, and aiding in the development
of regulations, guidelines, and standards.

For 1994, a total of approximately 76,500 Form R reports covering all SIC Codes
required to report toxic chemicals were submitted to EPA by approximately 23,000 facilities and
entered into the TRIS database. At the time the site surveys for RY 1994 were conducted,
12,896 Form Rs had been submitted and incorporated in the TRIS database for 3,764 facilities in
SIC Codes 25, 281, 285, and 30. For RY 1995, a total of approximately 74,500 Form R reports
covering all SIC Code codes required to report toxic chemicals were submitted to EPA by
approximately 22,000 facilities. At the time the site surveys for RY 1995 were conducted, 402
1-1

-------
facilities had been identified in the TRIS database as SIC Code 286 submitters and 165 facilities
had been identified as SIC Code 26 submitters.

1.1          EPA's Overall Quality Assurance Program

             Because of the wide audience and many intended uses of the Toxics Release
Inventory database, EPA designed and implemented a program to assess the quality of the data
collected under Section 313 and to identify areas where improved guidance would be useful for
improving the accuracy of future reported data.  The site surveys described in this report are a
component of EPA's overall quality assurance program.

1.2           Site Survey Objectives

              EPA's site surveys were designed to provide a quantitative assessment of the
accuracy of the data submitted for a calendar year by identifying the frequency and the
magnitude of errors in the Form R data and the reasons these errors occurred. EPA believed that
on-site review of industrial processes, pollution control technologies, and documentation
supporting the Form R reports would reveal errors in the database not obvious from review of a
facility's Form R submissions. Expected error types included overlooked chemicals, incorrectly
included chemicals, and errors in the release and other waste management quantity estimate
calculations.  The goal of the surveys was to obtain information that could be used to improve
the Form R reporting instructions and definitions, and thus improve the quality of data in the
TRIS database in future years.

              Users of the results of the site survey program, as well as the TRI database itself,
should be aware of a basic limitation of the Emergency Planning and Community Right-to-Know
Act (EPCRA) reporting process. Under EPCRA (Title III of the Superfund Amendments and
Reauthorization Act),  facilities are not required to perform any additional monitoring or
measurement of the quantity of toxic chemicals released to the environment to calculate Form R
release estimates. Therefore, the methods selected by facilities to estimate releases and other
waste management quantities depend on the nature of the data available to facility personnel, and
the quality of these release and other waste management quantity estimates in turn depends  on
both the proper application of the estimation methods and on the quality of available data. At
                                           1-2

-------
facilities where supporting data were available, surveyors carefully examined the estimation
calculations and data sources and then recalculated the estimates. In many instances, the site
surveyors were able to identify data sources overlooked by facility personnel, and these new data
were used to recalculate release and other waste management quantity estimates during the site
visits. However, site surveyors did not conduct any monitoring or measurements during the site
visits. Site surveyors also assessed the quality of the estimation methods by recalculating
releases and other waste management quantities using alternative approaches where more
accurate estimation methods were appropriate and where available data warranted.

1.3 EPA Site Surveys

EPA has conducted four sets of quality assurance site surveys since the first
submittal of Form Rs from industry. The RY 1987 site surveys covered all SIC Codes affected
by the EPCRA Section 313 (SARA Title III) requirements. The RY 1988 site surveys covered
SIC Codes 28, 291, and 34 - 38. These SIC Codes were targeted because data for the 1987
reporting year showed that facilities in these SIC Codes accounted for a substantial portion of the
total releases from all reporting facilities in 1987.'

The approach used for the RY 1994 and RY 1995 site surveys was similar to that
used for the RY 1987 and RY 1988 programs. Training of site surveyors, the contents of the
survey instrument, and activities conducted on site for RY 1994 and RY 1995 were similar to the
previous programs. However, the SIC Codes included in the site visits differed slightly from
those studied in previous years. The RY 1994 site surveys focused on facilities in SIC Codes 25,
281, 285, and 30; and the RY 1995 site surveys focused on facilities in SIC Codes 26 and 286.
These SIC Codes were targeted because previous reporting years showed that facilities in these
SIC Codes account for a substantial portion of the total releases and other waste management
quantities. The results of the RY 1994 and RY 1995 site surveys will help EPA identify ways
additional guidance can be structured to improve the overall quality of the data generated under
EPCRA (SARA Title III, Section 313) reporting.
'The results of these surveys are provided in Radian reports entitled Assessment of Data Quality in the 1987 Toxic
Release Inventory: Site Visit Program (March 1990), and Site Visit Program to Assess 1988 Toxic Release
Inventory Data Quality (July 1991).
1-3

-------

-------
2.0           APPROACH

              A very structured approach was established for the site surveys to ensure
consistency in conducting site surveys and accuracy of the results. The approach was orginially
established for the RY 1987 and RY 1988 site surveys and was improved for the RY 1994 and
RY 1995 site surveys based on experience from the previous programs.

              The approach for performing the RY 1994 and RY 1995 site surveys, shown
schematically in Figure 2-1, consisted of the following steps:

              (1)    Revising the Survey Instrument;
              (2)    Selecting facilities to be visited (Sample Selection);
              (3)    Training site surveyors (Training);
              (4)    Arranging Site Visits;
              (5)    Performing site visits (Site Visit Methodology);
              (6)    Data Management and Data Quality Assurance; and
              (7)    Data analysis and Reporting.


Each of these steps is discussed in the following subsections.
                                          2-1

-------
Revise the Survey Instrument
                                         Define the Goals
                                          for the Sample
                                        Selection Process
                                         Select a Random
                                        Group of Facilities
                                          for the Sample
Revise the Training Manual
                                                                           Train Site Surveyors
                                     Obtain Facility's Voluntary
                                     Participation and Arrange
                                            Site Visits
Prepare for each Site Visit
                                       Perform the Site Visits
                                     and Complete the Survey
                                      Instrument for each Site
  Develop a Computerized
 Data Management System
                                        Internally Check the
                                   Consistency of the Completed
                                        Survey Instruments
                                                                         Hold an Informal Wrap-up
                                                                            Meeting to Discuss
                                                                             Qualitative Items
                                        Enter and Verify the
                                         Data in the Data
                                       Management System
  Summarize the Results
  of the Wrap-up Meeting
    in a Memorandum
                                      Generalize the Results
                                       using Facility Weights
                                        REPORT RESULTS
             Figure 2-1.  Approach used to Perform the EPCRA Section 313
                                      Site Visit Program
                                               2-2

-------
2.1           Survey Instrument

              The survey instrument, shown in Appendix A, was designed to standardize and
facilitate the review of threshold determinations, release estimate calculations, and calculations
used to assess other waste management activities at facilities. The engineers and scientists who
performed the site surveys used the survey instrument as a detailed checklist to ensure that all
pertinent items were reviewed, and also as a consistent format for recording both the data
collected during site surveys and the errors made by facility personnel on their Form R reports.
In addition to its primary focus on chemical-specific information, the survey instrument contains
questions regarding the usefulness of the reporting instructions, EPCRA Section 313 hotline, and
the other published guidance materials. A question on the usefulness of the Toxic Chemical
Release Inventory Form A (Form A) was added for the visits conducted for RY 1995. Each
section of the survey instrument focuses on identifying specific types of errors made by facility
personnel on their Form R submittals.

             The survey instrument used in the RY 1994 and RY 1995 data quality site visits
was a revised version of the survey instrument used in the RY 1987 and RY 1988 programs.
Most of the questions remained the same, but some additional questions concerning
documentation available, possible sources for threshold determinations, source reduction
activities, pollution prevention technology, and use of the Form A (for RY 1995) were added to
clarify information received during the site visits and to assess the usefulness of the new
guidance and materials available. The time increments for amount of time needed to complete
all Form Rs at the facility were adjusted slightly in the RY 1995 survey instrument to obtain a
more precise estimate of time needed. The format was also revised to make the survey
instrument easier for the site surveyors to use.

2.2          Sample Selection

             The primary objective of sample selection was to obtain a random group  of
facilities from the key  industry groups within specific SIC Codes which were being surveyed to
appropriately scale up  the results to reflect the reporting of the entire SIC Code group. This
                                          2-3

-------
sample selection approach was used for Reporting Year 1987, 1988, 1994, and 1995 survey
programs.

2.2.1 RY 1987 and RY 1988 Sample Selection

All industry groups required to report toxic chemicals under the EPCRA Section
313 program were first surveyed for Reporting Year (RY) 1987. Table 2-1 presents the
distribution of facilities sampled among the SIC Codes for each year of the data quality site
visits.

A target of 150 facilities was selected as the number of facilities needed to ensure
the statistical validity of the data collected during the site visit program for RY 1987. Appendix
B provides a detailed discussion of the procedure used to select the sample group of facilities,
and also provides a description of the weighting system (i.e., the number of facilities in the TRI
database represented by each sample point). Briefly, facilities submitting 15 or fewer Form R
reports were divided into geographic clusters on the basis of the first three digits in their zip
codes. A sample of geographic clusters was then selected according to a sampling scheme in
which probability of selection is proportional to cluster size. The cluster size measure was the
total number of facilities in the SIC Code group sampled. The clustering approach was used to
minimize costs by reducing travel costs and travel time for site survey teams. A stratified
random sample of facilities was drawn from each of the sample geographic clusters, based on the
desired number of site visits in each SIC Code. This general procedure was used for sample
selection for site surveys conducted for RY 1987 and RY 1988.

Facilities with 15 or fewer Form Rs were selected due to the limited time and
budget available. Only a few facilities have 16 or more Form Rs and site visits to those facilities
would have taken considerable time, limiting the number of facilities that could be visited. Since
the same facility personnel may complete multiple reports at a given facility, visiting more
facilities presents a better representation of the range of reporting practices.
2-4

-------
                                     Table 2-1

                Distribution of Facilities Among the SIC Codes
                  For Each Year of the Data Quality Site Visits
SIC Code
20
22
23
24
25
26
27
28
29
30
32
33
34
35
36
37
38
39
Total
Number of Facilities Visited
RY 1987
16
5
1
2
2
14
3
44
0
7
2
16
16
5
11
7
2
3
156
RY 1988







43
1



8
10
14
10
3

89
RY 1994




25


37a

23








85
RY 1995





10

10










20
"One of the facilities visited was in SIC Code 282. The results of this survey are not included in the analysis of data
for SIC Code 28 land 285.
                                         2-5

-------
A target sample size of 90 completed site visits was the goal of the site visit
program for RY 1988. Details of the sample design and weighting methodology are described in
Appendix B (and follow the same general procedure as RY 1987). For RY 1988 facilities
submitting 30 or fewer Form R reports were targeted, rather than facilities with 15 or fewer Form
Rs as in other years.

2.2.2 RY 1994 and RY 1995 Sample Selection

The key industries sampled for the RY 1994 Toxic Release Inventory (TRI) data
quality site survey were furniture manufacturing, paint manufacturing, chemical manufacturing,
and rubber and plastics manufacturing. Key industries sampled for the RY 1995 TRI data quality
site survey were the organic chemicals manufacturing and pulp and paper industries. Facilities in
these industry groups were selected because they accounted for a substantial portion
(approximately half) of the total releases reported by facilities for the 1994 and 1995 reporting
years. The sample does not include facilities outside the above listed industry groups and
therefore does not represent the entire population of facilities that reported to the TRI.

Facilities engaged in furniture manufacturing were defined as those having a two-
digit SIC Code of 25. Facilities engaged in chemical manufacturing (SIC Code 28) were
ultimately refined to include only those facilities engaged in inorganic chemical manufacturing
with an SIC Code of 281, and paint manufacturing with an SIC Code of 285. Rubber and
plastics manufacturing facilities have an SIC Code of 30.

A target sample size of 40 completed site visits, divided evenly between SIC
Codes 281 and 285, was established for the first part of the RY 1994 site visit program. A target
sample size of 50 completed visits, divided evenly between SIC Codes 25 and 30, was the goal
of the second part of the RY 1994 site visit program. As discussed in Section 2.2.1, a stratified
random sample of facilities was drawn from a set of geographic clusters. The 1995 site visit
program targeted 20 completed site visits at facilities in SIC Codes 286 and 26 (10 visits each).
The geographic clustering approach was not used for RY 1995 because the sample set in the SIC
Codes chosen was small. A total random sampling was done for RY 1995. Details of the sample
design and weighting methodology are described in Appendix B.

2-6

-------
2.3           Site Surveyor Selection and Training

              To complete the site visit program as efficiently as possible, the engineers and
scientists staffing the program were selected on the basis of their experience in performing
environmental audits of industrial processing facilities, and were required to have a thorough
understanding of chemistry, technical calculations, multimedia environmental concerns, and
pollution control technologies.  The quality assurance reviewers for the site surveys were all
from one office and  remained consistent throughout the program.

              A surveyor training program was developed to ensure consistency and high
quality work among  all site surveyors.  The training program consisted of three steps:
              1)     Compiling a comprehensive training manual, including copies of EPA
                     guidance documents and other references;
              2)     Holding training sessions to familiarize project personnel with program
                     requirements; and
              3)     Review of the completed survey instruments with the site surveyor by the
                     reviewer to maintain a consistent approach among the surveyors.

2.4           Arranging Site Visits
              The goal in arranging site visits was to provide each facility in the sample with an
equal opportunity to participate in the site visit program, thus ensuring the statistical validity of
the approach.  Participation was voluntary; the facilities were not legally required to participate.
A key factor encouraging voluntary participation was the assurance of anonymity to the facilities.
Names, location, and all other facility identification data are shielded from the Agency. Upon
facility request, a written confidentiality agreement was signed by the contractors.

              As a first step, introduction letters (copies of these letters are provided in
Appendix C) were sent to each facility's technical contact, and where  appropriate, to each
facility's senior management official. These letters contained explanations of the purpose of the
quality assessment program and the anticipated burden on and benefits to the facility, and

                                           2-7

-------
assurance to the facility that all facility-specific data would be treated as confidential.  ERG
followed these letters with telephone calls to the technical contacts at the facilities to solicit their
participation, and for those facilities agreeing to participate, to arrange a date for the site visit and
to review a preliminary agenda for the site visit.

2.5           Conducting Site Visits

              The goal of the site visit was to  collect all the information needed to complete the
survey instrument accurately, while minimizing burdens on facility staff. On-site survey
activities included tours of the facilities, which focused on material storage areas, industrial
processing operations,  and pollution control equipment; careful review of all readily available
documentation, which  could include MSDSs, production data, monitoring data, purchasing
records, and facility spreadsheets or computer software with this information; and interviews
with appropriate facility employees regarding documentation materials. Site surveyors did not
perform any monitoring or measurements during the site visits.

              The site visits were designed to  determine:

              1)     Overlooked chemicals;
              2)     Releases and other waste management activities;
              3)     Errors in the Form R reports submitted to EPA; and
              4)     Whether more accurate release estimation methods could have been used,
                     based on information available to the facilities.

Releases and other waste management estimates were either recalculated or recreated by site
surveyors from available documentation during the visit.  Site surveyors recorded these results on
the survey instrument and reviewed the results with facility personnel before leaving the site. A
wrap-up meeting at the facility with the person who filled out the Form R reports was held at the
end of the visit to discuss any issues or questions that the facility contact had and to go over the
conclusions and recommendations of the site surveyor. Follow up with the facility contact after
the on-site visit occurred when regulatory issues which needed EPA clarification or additional
research was required.
                                            2-8

-------
2.5.1          Data Collection

              Site surveyors reviewed 295 Form R chemical reports and 728 additional
chemicals with amounts used or activities which did not meet the reporting criteria at the 85
facilities visited for RY 1994, and 139 Form R chemical reports and 171 additional chemicals
with amounts used for activities which did not meet the reporting criteria at the 20 facilities
visited for RY 1995.  Threshold determinations, releases,  and other waste management estimates
were reviewed separately to identify the frequency, magnitude, and sources of errors in these
areas.  Site surveyors followed the stepwise approach described in the Form R reporting
instructions for completing threshold determinations, releases, and other waste management
estimates. In following the Form R reporting instructions, facilities must first assess which
chemicals are manufactured, processed, or otherwise used in excess of appropriate thresholds.
Facilities must then estimate and report all releases to the environment and other quantities of
listed chemicals exceeding a threshold managed as waste.

2.5.2          Threshold Determinations

              The following types of errors may be made by facilities in determining which
chemicals at their site meet a EPCRA Section 313 thresholds:

              •      Overlooking a chemical;
              •      Incorrectly calculating a threshold amount;
              •      Incorrectly applying an exemption; and
              •      Misclassifying a chemical activity.

              To identify errors in threshold determinations, site surveyors looked for problems
in a facility's documentation and, on the plant tour, site surveyors looked for evidence of
chemicals that were reported but should not have been reported, and for evidence of chemicals
that were not reported but should have been reported. Each facility's documentation was
reviewed to track the decision process used to determine whether a chemical should have been
reported. Furthermore, site surveyors used all available documentation to recalculate threshold
                                           2-9

-------
estimates for reported chemicals and for chemicals present but not reported to verify the accuracy
of facility calculations.

2.5.3          Release and Other Waste Management Estimates

              The following types of errors may be made by facilities in calculating release
estimates for EPCRA Section 313 chemicals:

              •      Overlooking a chemical;
              •      Overlooking a source of data;
              •      Incorrectly calculating a release or other waste management quantity; and
              •      Incorrectly interpreting the reporting instructions.

A two-part approach was used for identifying errors in releases and other waste management
activity estimates.  First, site surveyors always recalculated releases and other waste management
quantities using the same technical approach used by the facility. Second, whenever the site
surveyor's experience and training indicated that a calculation approach different than that used
by the facility was appropriate, the surveyor attempted to obtain the data needed to calculate
releases and other waste management quantities using the more appropriate approach. In many
such instances, data were not readily available during the site visit to recalculate these amounts
using the alternative approach. In the cases where  site surveyors were able to recalculate releases
and other waste management amounts using alternative approaches, they were able to assess the
reasonableness of the estimation techniques used by facility personnel.

              The surveyors quantified all numerical differences between the facility's estimates
and the recalculated values, even in instances where surveyors identified only small differences.
As discussed later, these numerical differences were used to assess quantitatively the accuracy of
the total aggregate releases and other waste management quantities contained in the TRI
database.
                                           2-10

-------
 2.6           Data Management/Data Quality Assurance

 Many steps were taken to ensure the data quality of the surveyor's estimates and the verification
 of the data in the database.  This section outlines the procedures taken to review the Survey
 Instruments after they have been completed by the site surveyor, the database system, and the
 data entry into the master database; the verification procedures for the data entered into the
 database; the weighting of the data to apply the results to the entire population of facilities for
 each SIC Code surveyed; and potential sources of error in the site survey program.

 2.6.1         Quality Review of Survey Instrument and Data Entry

              All survey instruments were reviewed twice by a consistent set of reviewers to
 ensure the calculations and methodologies used were correct and consistent for all site surveys.
 The data entry for all site surveys was also done twice. These database entries were compared to
 each other, and then verified with the actual survey if an inconsistency was found. Project staff
 also reviewed the database entries for internal consistency and completeness by comparing
 responses to various questions as appropriate.

 2.6.2         Data Weighting

              To allow EPA to assess the impact of the site survey program results on the TRI
 database for the SIC Codes surveyed, weighting factors were applied to the site visit data. These
 factors or "weights" represent the number of facilities in the TRI database represented by each of
 the surveyed facilities. The weights of each surveyed facility are based on the measure of size of
 the geographic cluster in which the facility is located, and the systematic probability of selecting
 that facility proportional to that measure of size.

              The weights for the sample facilities in each SIC Code group are summed up to
represent the total population of facilities included on the TRI for that SIC Code group. A total
population of 535 facilities for SIC Code 25,1872 facilities for SIC Code 30, 662 facilities for
 SIC Code 285, and 695 facilities for SIC Code 281 is represented for RY 1994. A population of
                                          2-11

-------
402 facilities for SIC Code 286 and 165 facilities for SIC Code 26 is represented for RY 1995.
The weights used for facilities in SIC Codes 25, 281, 285, and 30 are presented in Appendix B.

2.6.3 Limitations of the Analysis

The design and implementation of the survey may have introduced unavoidable
inaccuracies in the study results. The three primary sources of error are:

• sample selection bias;
• survey implementation; and
• data reduction and analysis.

The relatively small number of facilities sampled clearly introduced a sample
selection bias — the smaller the number of facilities sampled the greater the likelihood that these
facilities do not accurately represent the universe of reporting facilities. For the selected sample
size of 40 facilities in SIC Code 28, the 90 percent confidence interval is plus or minus thirteen
percent. That is to say, if 50 percent of the facilities visited reported accurate data there is a 90
percent probability that between 37 percent and 63 percent of the facilities in the national
database reported accurate data. Counting the SIC Code groups separately, there is a 90 percent
confidence level of plus or minus 18 percent for each group. For the selected sample size of 50
facilities in SIC Codes 25 and 30, the 90 percent confidence interval is plus or minus 17 percent.
Counting the SIC Code groups separately, there is a 90 percent confidence level of plus or minus
24 percent for each group. Thus, the confidence levels are based on the survey size and the total
number of facilities in the SIC Code group.

Another possible source of error concerns the fact that approximately 15 different
surveyors performed the survey. This source of inaccuracy was controlled to the extent possible
by the use of a carefully designed survey instrument and extensive quality assurance provisions.
Nevertheless, it is possible that different surveyors made different judgments in the course of the
site surveys.
2-12

-------
Finally, certain assumptions were made to simplify data analysis. The key
assumption was that the facilities and Form Rs examined in the site visits accurately represent all
facilities in their SIC Code group in terms of the accuracy of the data submitted. Aside from
possible errors introduced by the relatively small size of the sample, the sampled facilities may
not fully represent their SIC Code group because:

• The sampled facilities excluded any facility with more than 16 Form Rs for
budgetary reasons. To the extent that facilities submitting more than 16 Form Rs
report more (or less) accurate data than the sampled facilities, the latter facilities
do not fully reflect the universe of facilities in the database.

• Many facilities surveyed processed or manufactured some kind of specialty
chemical. These facilities may not accurately portray the "typical" facility within
the SIC Code group. This may overestimate a specific chemical produced and
released within the SIC Code group due to scaling and weighting factors.

2.7 Data Analysis and Reporting

Once the results of the site surveys were loaded into a database and the database
was validated through the quality assurance process described above, the data were evaluated to
discern trends in the quality of data in the TRI forms. This report presents the results of that
analysis.
2-13

-------

-------
3.0           THRESHOLD DETERMINATIONS


              This section reviews how accurately facilities determined whether Section 313
chemicals exceeded TRI reporting thresholds. Errors in threshold determinations can cause
facilities to submit Form Rs for chemicals that do not exceed applicable reporting thresholds and
also can cause facilities to fail to submit Form Rs for chemicals that exceed these thresholds.
These two scenarios may lead facilities to understate or overstate emissions reported to the TRI
database, respectively.  Using the following topics, this section discusses the extent to which
erroneous threshold determinations impact the quality of TRI data:
                     Approaches commonly used to calculate thresholds
                     Errors made when calculating thresholds
                     Reasons for making erroneous threshold determinations
              The section concludes with a review of key findings and offers several
recommendations to help facilities improve reporting practices in the future.

              It should be noted that this section does not differentiate facilities that submitted
Form As from those that submitted Form Rs. Because the magnitude of the total annual
reportable amount (the sum of Sections 8.1 - 8.7 on the Form R) ultimately determines when
facilities can use Form As, Sections 5 and 6 of this report provide specific details on the
frequency with which facilities use short reporting  forms.

3.1           Approaches Used for Determining Thresholds

              The following discussion considers  how the approaches that facilities use to
calculate thresholds affect the quality of TRI data.  Although the TRI reporting instructions
include specific criteria for determining when chemicals exceed reporting thresholds, the
instructions do not require facilities to use specific  approaches for conducting threshold
determinations. Accordingly, facilities use many different approaches to  determine whether
Section 313 chemicals exceed activity thresholds.  The most appropriate approach depends

                                           3-1

-------
largely on information available to the facility and specific uses of a chemical. For example,

purchasing and inventory data may be the best data source for evaluating thresholds for certain

raw materials, while process recipes or production data may be more appropriate data sources for

evaluating thresholds for products.  During each site audit, surveyors identified approaches used

for determining thresholds from information available in the facility's supporting documentation.

In cases where facilities did not estimate thresholds, site surveyors identified an approach that

could reasonably have been used to estimate thresholds.


              Figures 3-1, 3-2, and 3-3 summarize the approaches most commonly used by

facilities (or that could reasonably have been used by facilities) to estimate thresholds for TRI

reporting years 1987,1994, and 1995, respectively.  The quantitative data for RY 1988 were not

available.  For quick reference, Table 3-1 presents the same data for all reporting years

considered.  This table and these figures illustrate several trends:
                     For RY 1987,1994, and 1995, and for all industries considered, facilities
                     primarily used purchasing data to determine whether chemicals exceeded
                     appropriate thresholds.  By this approach, facilities typically determine
                     annual usage by calculating total purchases during a calendar year and
                     correcting these quantities for changing levels of inventory.  For reference,
                     Appendix D includes an example threshold calculation using this
                     approach.

                     Facilities in the inorganic and organic chemical manufacturing industries
                     (SIC Codes 281 and 286) use production data to calculate thresholds more
                     frequently than facilities in other industries. These industries typically use
                     production data to determine thresholds for Section 313 chemicals that are
                     produced on-site by chemical reactions.  Facilities in the furniture (25),
                     paper (26), paint (285), and plastics (30) manufacturing industries
                     generally do not have reactions that produce large quantities of Section
                     313 chemicals and therefore rarely use production data to calculate
                     thresholds.

                     Facilities in the inorganic and organic chemical manufacturing industries
                     also are more likely to assume that chemical usage exceeds reporting
                     thresholds, rather than calculate annual usage directly. Making such
                     assumptions is only advised in cases where facilities clearly produce or
                     consume extremely large quantities of Section 313 chemicals, a scenario
                     common to large chemical manufacturing plants.  As noted in Section
                                           3-2

-------
                                                           Table 3-1
                          Approaches Used by Facilities to Make Threshold Determinations
Approach Used to Make
Threshold Determination
Purchase/Inventory records
Production data
Process recipes
Emission factors a
Mass balance
Material safety data sheets
Monitoring data
Assumed threshold exceeded
(no calculations completed)
Other approach
Percent of Facilities Using Specific Approach for Threshold Determinations
RY 1987
SIC Code
20-39
85
1
3
4
3
0
1
2
0
RY 1994
SIC Code
25
96
0
16
0
4
1
0
4
12
SIC Code
281
87
49
16
0
9
13
8
63
4
SIC Code
285
96
12
4
0
0
23
0
8
1
SIC Code
30
93
15
15
6
0
4
0
0
0
RY 1995
SIC Code
26
100
0
0
50
20
0
30
10
10
SIC Code
286
90
70
50
10
20
0
10
20
20
Note: Because some facilities used multiple approaches to calculate thresholds for a given chemical, the sum of the percents for a given SIC Code may exceed
100.
a "Emission Factors" in this table refers to any factors supplied by EPA or a trade association which are technically supported and used to determine the amount
of any given chemical manufactured, processed, or otherwise used at the facility.

-------
D
91
5T
ta
I

o
 Cr
 (D
 §

 g
 5"
 CO
                                                                 Percent of Threshold Determinations
                      s
                      o
                       $F
                       (D
                       a
                       I

                       I
                        (D
                        VI


                        O.

                        a.
                        1/1
                   o   o
                                      6
                        8   8
tiec
Ol
o
o
o
i APF
o>
o
8
iroac
-j
o
8
11
CO
O
8
to
o
o
o
-n
(5'
c.
*.«
                                    Purchase/inventory records
                                              Production data
                                            Process recipes
Emission factors
                                             Mass balance
                                Material safety data sheets
                                         Monitoring data
                           Assumed threshold exceeded

                            (no calculations completed)
                                                                               •a

                                                                                3

                                                                                o
                                                                                31
                                                                                (D
                                                                                (/>

                                                                                C
                                                                                (A

                                                                                a

                                                                                rjo
1
=r.
                                                                                 &>

                                                                                 (D

                                                                                 H
                                                                                 3"

                                                                                 5
                                                                                                                             a.
                                                                                                                             a
                                                                                                                              &>
                                                                                                                              cr.
                                                                                                                              o
                                                                                                                              3
                                                                                                                              (A

-------
                                                         s-e
o
o-
-------
O
01

5T
                                                              9-e
o
01
3

O-
§

i!
                                                                Percent of Total Chemical Threshold

                                                                  Determinations in Each Category
                                                                                    fl
                                                                                   <5*
                                                                                    c
                 •a
                 •o
                 •^
                 o
                 u
                 o
                 IT
                 
-------
                     3.3.4, however, facilities that assume thresholds are exceeded tend to make
                     more errors in threshold determinations than facilities that calculate actual
                     quantities manufactured, processed or otherwise used.
              To evaluate how approaches for determining thresholds might affect the quality of
TRI data, the remainder of this section considers how these approaches may have caused
facilities to make incorrect threshold determinations.

3.2           Frequency of Errors Made When Determining Thresholds

              The following analyses indicate the frequency and type of errors made by
facilities when determining thresholds and comment on how these errors may affect the overall
quality of TRI data. During site visits, surveyors used information provided by facility contacts
to calculate thresholds for Section 313 chemicals. For reference, Attachment F includes several
examples of how site surveyors calculated thresholds.  Based on these threshold calculations,
surveyors then listed chemicals for which facilities should have submitted Form Rs. Errors in
threshold determinations were identified by comparing lists of chemicals that exceeded
thresholds, as determined by the site surveyor, to chemicals for which facilities submitted Form
R reports. These comparisons yielded four possible outcomes:

              •      The facility submitted a Form R for a chemical that exceeded a threshold.
              •      The facility did not submit a Form R for a chemical that did not exceed a
                    threshold.
              •      The facility submitted a Form R for a chemical that did not exceed a
                    threshold.
              •      The facility did not submit a Form R for a chemical that exceeded a
                    threshold.

              The first two outcomes represent cases where facilities correctly determined
thresholds, while the last two outcomes represent two general types of errors made when
calculating thresholds.  The last two errors can cause facilities to overstate or understate,
respectively, the releases and other quantities managed as waste reported to TRI. Using these
                                           3-7
-------
outcomes, Figure 3-4 and Table 3-2 summarize the frequency of errors that facilities made when

determining thresholds during reporting years 1987, 1988, 1994, and 1995. As a summary,

Figures 3-5 and 3-6 indicate the frequency with which facilities made correct and incorrect

threshold determinations. Several important observations can be made from these data:
                     Over all industries and reporting years considered, Figures 3-5 and 3-6
                     indicate that facilities consistently determined thresholds correctly for over
                     90 percent of the Section 313 chemicals used at their respective plants.
                     The errors made when determining thresholds were almost evenly split
                     between failing to submit Form-Rs for chemicals that exceeded thresholds
                     and submitting Form Rs for chemicals that did not exceed thresholds.

                     For roughly 5 percent of the Section 313 chemicals used in the industries
                     considered, facilities submitted Form Rs when thresholds were not
                     exceeded. Facilities in the inorganic and organic chemical manufacturing
                     industries (SIC Codes 281 and 286) made this error more frequently than
                     facilities in other industries. The errors made by inorganic and organic
                     chemical manufacturers may be caused by these facilities assuming that
                     thresholds were exceeded (see Figures 3-2 and 3-3) rather than actually
                     calculating annual quantities manufactured, processed or otherwise used.
                     Section 3.3 provides additional insight into the sources of these errors.

                     Also for roughly 5 percent of the Section 313 chemicals used in the
                     industries considered, facilities failed to submit Form Rs when thresholds
                     were exceeded. Again, this tendency was greatest for facilities in the
                     inorganic chemical manufacturing industry (SIC Code 281).  Section 3.3
                     examines the sources of these errors in greater detail.
              In summary, the frequency of errors suggest that industries incorrectly compute

thresholds for between 5 and 10 percent of the Section 313 chemicals used at their corresponding

facilities. These errors included a nearly even number of cases in which facilities submitted

Form Rs for chemicals that did not exceed thresholds as cases in which facilities did not submit

Form Rs for chemicals that exceeded thresholds.
                                            3-8
-------
                                                         Table 3-2
                     Accuracy of Threshold Determinations by Reporting Year and SIC Code



Outcome



Facility did not submit a
Form R for a chemical that
did not exceed a threshold
Facility submitted a Form R
for a chemical that exceeded
a threshold
Facility did not submit a
Form R for a chemical that
exceeded a threshold
Facility submitted a Form R
for a chemical that did not
exceed a threshold
Percent of Section 313 Chemicals Broken Down by Threshold Determination Outcome
RY
1987

SIC
Code
20-39

59


34


3


4

RY 1988


SIC
Code
28, 291

52


41


2


5

SIC
Code
34-36

62


34


1


3

RY 1994


SIC
Code
25

70


28


1


1

SIC
Code
281

41


43


8


8

SIC
Code
285

66


26


5


2

SIC
Code
30

76


23


2


0

RY 1995


SIC
Code
26

70


28


0


2

SIC
Code
286

16


72


1


11

Note: The first two outcomes represent cases where facilities correctly determined thresholds, while the last two outcomes represent cases where facilities
incorrectly determined thresholds. Figure 3-5 compares the correct and incorrect threshold determinations by reporting year and SIC Code.
-------
                        Figure 3-4.  Accuracy of Threshold Determinations by Reporting Year and SIC  Code
                        100%

                 -
                   S.c
                   o
                 —
U)

o

15 1
•*-«  c
                           10%
                            o% -
                                                                                                        Facility did not Submit a Form-R for a
                                                                                                    Chemical that did not Exeed a Threshold (Correct)
                                                                                                     Facility Submitted a Form-R for a
                                                                                                Chemical that Exceeded a Threshold (Correct)
                                                                                                    Facility Submitted a Form-R for a
                                                                                             Chemical that did not Exceed a Threshold (Incorrect)
                                                                                            Facility did not Submit a Form-R for a
                                                                                         Chemical that Exeeded a Threshold (Incorrect)
                                                 Reporting Year/SIC Code
         Note: The first two outcomes represent cases where facilities incorrectly determined thresholds, while the last two outcomes represent cases where facilities
         correctly determined thresholds. Figure 3-5 compares the correct and incorrect threshold determinations by reporting year and SIC Code.
         Data for this figure can be found on Table 3-2.
-------
        Figure 3-5. Accuracy of Threshold Determinations by Reporting Year and SIC Code
                                   for RY1987 and RY1988
              93%
SIC Code 20-39 (RY 1987)
                                                  93%
SIC Code 28,291 (RY 1988)
                                                                                   96%
SIC Code 34-36 (RY 1988)
                                    > Correct
           i Incorrect
-------
                        Figure 3-6. Accuracy of Threshold Determinations by Reporting Year and SIC Code
                                                    for RY1994 and RY1995
OJ
                        2%
                          98%
                   SIC Code 25 (RY 1994)
                              93%

                 SIC Code 285 (RY 1994)
                                                     16%
                                                                       84%
 SIC Code 281 (RY 1994)
         98%

SIC Code 30 (RY 1994)
        98%

 SIC Code 26 (RY 1995)
              88%

SIC Code 286 (RY 1995)
                                                    g Correct      H Incorrect
-------
3.3           Sources of Errors Made When Determining Thresholds

              This section summarizes reasons why facilities made errors when determining
thresholds. During site visits, surveyors identified these reasons from discussions with facility
contacts and from data in the facility's supporting documentation. The following subsections
examine the sources of errors from the most recent round of site surveys. To help industrial
facilities minimize errors made in threshold determinations, Section 3.4 offers several
recommendations for avoiding such errors in future reporting years.
3.3.1          Reasons Why Facilities Failed to Submit Form Rs for Chemicals That
              Exceeded Thresholds
              Table 3-3 summarizes reasons why facilities failed to submit Form Rs for
chemicals that exceeded thresholds during TRI reporting years 1994 and 1995. The data are not
classified by SIC Code, because not enough errors were observed for these reporting years to
make statistically significant conclusions for each industry. Further, the data are not compared to
those for previous reporting years, because the previous studies used slightly different sets of
categories to classify errors.  As shown in the table, the most common reasons why facilities did
not identify chemicals used at reportable levels was because facilities either overlooked the use
of Section 313 chemicals or facilities miscalculated annual thresholds. These general reasons
include a wide range of different errors, including cases where facilities assumed thresholds were
not exceeded, cases where facilities were unaware that chemicals could have exceeded reportable
quantities, and cases where facilities miscalculated total usage. The following lists indicate
specific examples of errors documented during site visits:

Overlooking a chemical activity:
                     For each TRI reporting year, a consultant for a plastic manufacturer
                     prepared Form Rs for the same set of chemicals, without first calculating
                     thresholds for all Section 313 chemicals used at the facility. The
                     consultant failed to notice that toluene should have been reported in 1995
                     due to increased use of certain solvents.
                                           3-13
-------
                                                            Table 3-3

                                      Reasons Why Facilities Failed to Submit Form Rs
                                           for Chemicals That Exceeded Thresholds
Reason for not submitting a Form R for
chemicals that exceeded thresholds
Chemical activity was overlooked
Chemical activity was misclassified
Threshold quantity was miscalculated
Chemical was incorrectly reported as a chemical category
RY 1994
Number of
Observations
9
0
7
1
Percent of
Total Errors
53
0
39
6
RY 1995
Number of
Observations
1
0
0
0
Percent of
Total Errors
100
0
0
0
u>
Note: Due to the limited number of errors identified during the site visits, the percents listed in this table may not necessarily represent the actual distribution of
reasons why facilities make errors on their Form Rs.
-------
                    A paint manufacturer did not examine the Material Safety Data Sheet
                    (MSDS) for "commercial grade" xylene, which indicated that the mixture
                    contained 15 percent ethylbenzene. Although the facility correctly
                    reported for xylene, the facility failed to report for ethylbenzene.

                    A furniture manufacturer failed to notice that a pigment used to coat metal
                    products contained several metal compounds on the TRI reporting list.
                    The MSDS for this pigment and the annual usage of the pigments suggest
                    that the facility should have reported for the metal compounds.

                    A paint manufacturer used a solvent containing glycol ethers to thin
                    several paint products but assumed that the limited usage could not
                    possibly have exceeded threshold quantities. Review of purchasing data
                    indicated that total annual usage was significantly greater than threshold
                    amounts.

                    A chemical manufacturer correctly reported for all Section 313 chemicals
                    that were in reactants and products but did not consider Section 313
                    chemicals that were components of wastewater treatment mixtures and
                    catalysts. The site surveyor noted that some of these chemicals that were
                    not used directly for production exceeded reporting thresholds.
Miscalculating a threshold:
                    A furniture manufacturer did not report for xylene (mixed isomers), but
                    made several calculation errors when determining annual quantities
                    otherwise used from individual purchasing invoices. The site surveyor
                    loaded the purchasing data into a spreadsheet and calculated an annual
                    quantity otherwise used exceeding corresponding threshold quantities.

                    A chemical manufacturer used the lower bound of a concentration range to
                    make a threshold determination for a Section 313 chemical. Using the
                    midpoint of the concentration range (as required by the TRI reporting
                    instructions), the site surveyor found the chemical exceeded threshold
                    amounts.
3.3.2          Reasons Why Facilities Submitted Form Rs for Chemicals That Did Not
              Exceed Thresholds
              Table 3-4 summarizes why facilities submitted Form Rs for chemicals that did not

exceed thresholds during TRI reporting years 1994 and 1995. For the same reasons as given in

the previous section, the data are not classified by SIC Code and are not compared to previous
                                          3-15
-------
                                                            Table 3-4

                                         Reasons Why Facilities Submitted Form Rs
                                       for Chemicals That Did Not Exceed Thresholds
Reason for submitting a Form R for
chemicals that did not exceed thresholds
Facility reported despite noting that threshold was not
exceeded
Facility assumed threshold was exceeded
Chemical activity was misclassified
Threshold quantity was miscalculated
Chemical was exempt
Facility misinterpreted revised reporting guidelines
Other
RY 1994
Number of
Observations
4
2
1
2
0
11
2
Percent of
Total Errors
18
9
5
9
0
50
9
RY 1995
Number of
Observations
2
3
2
2
1
3
1
Percent of
Total Errors
14
21
14
14
7
21
7
o\
     Note:  Due to the limited number of errors identified during the site visits, the percents listed in this table may not necessarily represent the actual distribution of
     reasons why facilities make errors on their Form Rs.
-------
reporting years. As shown in the table, approximately two thirds of the incorrectly submitted

Form Rs resulted from facilities misinterpreting revised reporting threshold guidance or from
facilities submitting Form Rs despite calculating annual usages below threshold quantities. The

following list describes specific instances when site surveyors classified threshold determination

errors in these two categories.


Facility misinterpreted revised reporting guidelines:
              •       A paint manufacturing facility submitted a Form R for acetone in reporting
                     year 1994 even though EPA removed acetone from the list of reportable
                     chemicals. The site surveyor noted that acetone was delisted and that the
                     form should be withdrawn.

              •       A paper manufacturing facility used aqueous ammonia in several process
                     areas but did not account for the revised reporting guidance indicating that
                     only 10 percent of .aqueous ammonia should be counted towards threshold
                     determinations. Using this guidance, the site surveyor determined that the
                     quantities of ammonia used did not exceed reporting thresholds. For
                     reference, Appendix F includes a specific example of computing
                     thresholds for aqueous ammonia solutions.

              •       A chemical manufacturing facility submitted a Form R for aqueous
                     sulfuric acid  used to neutralize wastewater, without considering the
                     sulfuric acid  activity qualifier for "acid aerosols." The site surveyor noted
                     that insignificant quantities of the sulfuric acid existed as aerosols and
                     concluded that the chemical should not have been reported.

Facility reported for the chemical, despite calculating a manufacture, process,  or otherwise use
quantity below threshold quantities:

              •       An organic chemical manufacturing facility submitted a Form R for
                     xylene in every reporting year since 1987. In reporting year 1995,
                     however, the facility noted that usage of xylene was below the
                     corresponding thresholds. Fearing that not submitting a Form R for a
                     chemical that was previously reported might somehow trigger an audit or
                     enforcement  response, the facility reported for xylene anyway.

              •       A paper manufacturing facility correctly determined that 9,700 pounds of
                     chlorine were "otherwise used" during reporting year 1995. The facility
                     submitted a Form R for chlorine anyway, noting that it would be better to
                     report a chemical that may not have exceeded a threshold than to not
                     report a chemical that exceeded a threshold.
                                          3-17
-------
3.3.3          Chemical Activity Classification


              Because appropriate TRI reporting thresholds (i.e., 10,000 or 25,000 pounds)

depend on how facilities use Section 313 chemicals, it is important that facilities correctly

classify chemical activities as either "manufacture," "process," or "otherwise use." For instance,

Table 3-4 indicates that several facilities submitted Form Rs for chemicals that did not exceed

thresholds due to incorrect chemical activity classifications. To evaluate how accurately

facilities classify chemicals, site surveyors documented during each site visit activities for all

Section 313 chemicals based on information provided by facility contacts and on observations

made during facility tours. Table 3-5 compares chemical activity classifications made by

facilities to those made by site surveyors for reporting years 1994 and 1995, respectively; Figure

3-7 also displays this data but in bar chart format.  Similar data are not available for previous
reporting years.  Based on these data, site surveyors note that:
                     Facilities in all industries made errors when classifying chemical
                     activities, with rubber and plastic manufacturers (SIC Code 30) making
                     the fewest and paint manufacturers (SIC Code 285) making the most.

                     Of the three chemical activities, facilities made most errors determining
                     whether chemicals were "processed" or "otherwise used." These
                     particular errors resulted to a great extent from facilities misclassifying
                     chemical activities for solvents. Part of this confusion may originate from
                     text in the TRI Reporting Instructions handbook which correctly lists
                     solvents as an example of a formulation component (under "processed") as
                     well as an example of a chemical processing aid (under "otherwise used").
                     Some facilities did not understand the distinction between these two
                     activities.

                     Some classification errors resulted from facilities being unaware that a
                     chemical with multiple uses could be classified under more than one
                     chemical activity.
                                           3-18
-------
                                                      Table 3-5
      Comparison of Chemical Activity Classifications Made by Reviewers to Those Made by Facilities,
                                         by Reporting Year and SIC Code
Chemical
Activity
Manufacture
Process
Otherwise
Use
Number of Chemicals Used at Selected Facilities, Classified by Activity
Reporting Year 1994
SIC Code 25
Facility
0
12
62
Reviewer
0
10
67
SIC Code 281
FacUity
26
65
34
Reviewer
17
73
35
SIC Code 285
Facility
1
67
15
Reviewer
0
83
25
SIC Code 30
Facility
0
30
26
Reviewer
0
31
26
Reporting Year 1995
SIC Code 26
Facility
56
4
37
Reviewer
64
3
29
SIC Code 286
Facility
77
46
26
Reviewer
56
58
29
Note: Reviewers and facilities may have classified selected Section 313 chemicals under multiple chemical activity categories. Therefore, the total number of
chemicals classified under "facility" for a given SIC Code does not necessarily equal the total classified under "reviewer."
-------
           Figure 3-7. Comparison of Chemical Activity Classifications made
                         by Facilities to those made by Reviewers
           Manufacture
                        Process
          Otherwise Use
                                                            Manufacture
                                                  Process
                                                                                    Otherwise Use
                  SIC Code 25 (RY 1994)
                                           SIC Code 281 (RY1994)
          Manufacture
Process
Otherwise Use
               SIC Code 285 (RY 1994)
                                                 a
                                                 O o
                                                 «> <
                                                 1!
                                                 15
                                                   I
                              100
                               90
                               80
                               70
                               60
                               50
                               40
                               30
                               20
                               10
                                0
                                                           Manufacture
                                                  Process
                                                                                    Otherwise Use
                                           SIC Code 30 (RY 1994)
          Manufacture
Process
Otherwise Use
                SIC Code 26 (RY 1995)
                                                           Manufacture
                                                  Process
                                                                                    Otherwise Use
                                           SIC Code 286 (RY 1995)
                                    D Reviewer
                               I Facility
Date for this figure can be found on Table 3-5.
                                            3-20
-------
              Although misclassified chemical activities account for only a small fraction of
errors in threshold determinations (see Table 3-3 and 3-4), EPA can help minimize these errors in
future reporting years by informing facilities of common mistakes and of correct classifications
for chemicals that are frequently misclassified, such as solvents.

3.3.4         Impact of Not Calculating Thresholds

              An important factor to consider in the accuracy of threshold determinations is
whether facilities actually calculated threshold levels for Section 313 chemicals or assumed that
thresholds were, or were not, exceeded. At each facility visited, site surveyors used feedback
from facility contacts and data in supporting documentation to determine which method was
adopted to make threshold determinations. For Section 313 chemicals found to exceed reporting
thresholds, Figure 3-8 summarizes the frequency with which facilities in the selected industries
actually calculated annual usages. The data in Figure 3-8 may, at first, seem to contradict the
data in Table 3-1  (approaches used for determining thresholds). The reader should note,
however, that Table 3-1 indicated approaches that facilities actually used to determine thresholds
for each chemical reported. Figure 3-8 counts thresholds calculated at least once (for any
chemical) at a given facility.

              Not surprisingly, facilities in the industries that calculated thresholds most often
(SIC Code  25 and SIC Code 30) made fewer errors when determining thresholds than facilities in
the industries that calculated thresholds less frequently (SIC Code 281 and SIC Code 285).  This
observation suggests that errors in threshold determinations may be significantly reduced if
facilities actually calculate annual usages for Section 313 chemicals, as opposed to assuming that
chemicals are below or above reporting thresholds.

3.4           Lessons Learned

              In summary, site surveyors found that facilities in the furniture, rubber, paper, and
plastic manufacturing industries (SIC Codes 25,26, and 30) determined thresholds more
accurately than facilities in the inorganic, paint, and organic chemical manufacturing industries
                                           3-21
-------
                                Figure 3-8.  Frequency with which Facilities Calculated Thresholds for
                                                   EPCRA Section 313 Chemicals
                                                                              44%
                               93%

                   SIC 25 Code (RY 1994)
                                                56%
                                                                                            18%
                                                                                                                  82%
SIC Code 281 (RY 1994)
SIC Code 26 (RY 1995)
to
            26%
                                         74%
                                                                                               12%
                   SIC Code 285 {RY 1994)
          95%

 SIC Code 30 (RY 1994)
               88%

SIC Code 286 (RY 1995)
                                    I Facility Actually Calculated Threshold
                I Facilty did not Calculate Threshold
-------
(SIC Codes 281, 285, and 286).  Further, facilities as a whole correctly calculated thresholds for
over 90 percent of the Section 313 chemicals used at the selected industries.  For nearly 5 percent
of the Section 313 chemicals, however, facilities failed to submit Form Rs in cases where
thresholds were exceeded; and, for the remaining 5 percent of chemicals, facilities submitted
Form Rs when quantities manufactured, processed or otherwise used did not exceed thresholds.
Therefore, according to the most recent site surveys, a small fraction of the Form Rs currently
logged in the TRI database need not have been filed, but facilities failed to submit a nearly equal
amount of Form Rs for chemicals that exceeded threshold levels. Although this observation may
suggest that the total number of Form Rs in the TRI database is highly representative of the
actual amount of Section 313 chemicals that exceed thresholds, it must be noted that the site
survey data may be influenced by limitations posed by sample selection (see Section 2.2).
                                          3-23
-------
-------
4.0           SOURCES AND TYPES OF RELEASES, OFF-SITE TRANSFERS, AND ON-SITE
              WASTE MANAGEMENT ACTIVITIES
              This section provides an overview of the sources of releases, off-site transfers for
further waste management, and on-site waste management activities as well as the release types
that were both claimed and observed at each facility visited during the site surveys. Statistically
weighted percentages of data are presented to show the distribution of release sources and release
types within each SIC Code (see Section 2 for a discussion of statistical weighting).
Additionally, percentages of incorrectly reported data and overlooked data are presented. Trends
and corresponding discussions regarding observations made during the site visits are presented,
as applicable.

              For the purposes of this report, "sources" are defined as the streams or units that
generate the release, off-site transfer, or on-site waste management activity (such as process
vents, container residue, or spills) and "release types" are defined as the environmental media
corresponding to elements in Sections 5 through 7 of the Form R (such as releases to fugitive air,
releases to stack air, releases to water, releases to land, and transfers to off site disposal). In most
cases, data has been presented both in a tabular form for quantitative analysis and in a graphical
format for qualitative trend analyses.

              Data is presented for RY 1994 and RY 1995.  A trend analysis has been
conducted whenever applicable between the six SIC Codes visited for RY 1994 and RY 1995.
4.1           Observed On-Site Releases. Off-Site Transfers, and On-Site Waste
              Management Activities
              Table 4-1 presents the distribution (weighted) of sources and the corresponding
release type, off-site transfer, or on-site waste management activity that was observed during the
site visits for each SIC Code. It also lists a "total" row for each of these activities. This
represents the number of facilities that reported at least one release or other waste management
activity from any source to that activity or release type.  Figure 4-la presents the "total" by
release type or other waste management activity and Figure 4-lb through 4-lh present the data
                                           4-1
-------
                             Table 4-1
Distribution of Release Sources and Off-Site Waste Management Activities
                       RY 1994 and RY 1995
Release or
Waste Management
Activity Type
Fugitive
Stack
Source
Volatilization from Process Areas
Pumps/Valves/Flanges
Storage Tank/Stock Pile Losses
Housekeeping Practices/Clean-up Wastes
Accidental Spills/Releases
Process Vents
Volatilization from Treatment Areas
Container Residue
Other2
TOTAL: Reporting from at Least One Source1:
Volatilization from Process Areas
Pumps/Valves/Flanges
Storage Tank/Stock Pile Losses
Housekeeping Practices/Clean-up Wastes
Accidental Spills/Releases
Process Vents
Volatilization from Treatment Areas
Percent of Facilities Documenting Releases or
Waste Management Activity (weighted)
SIC
Code
25
76.0%
43.0%
14.5%
29.0%
0.0%
0.0%
0.0%
0.0%
0.0%
92.8%
100.0%
1.0%
27.9%
20.5%
4.0%
0.0%
4.0%
SIC
Code
281
81.8%
70.7%
30.5%
20.0%
44.3%
23.4%
39.2%
5.5%
7.6%
100.0%
65.7%
21.7%
60.0%
0.0%
8.3%
71.2%
5.5%
SIC
Code
285
100.0%
50.1%
53.1%
22.5%
22.5%
12.9%
8.2%
0.0%
0.0%
100.0%
29.9%
0.0%
57.9%
14.6%
0.0%
38.1%
0.0%
SIC
Code
30
63.4%
23.3%
14.4%
8.8%
0.0%
0.0%
0.0%
0.0%
0.0%
67.4%
50.9%
6.0%
12.8%
0.0%
0.0%
0.0%
5.4%
SIC
Code
26
70.0%
30.0%
30.0%
0.0%
10.0%
0.0%
80.0%
0.0%
0.0%
90.0%
80.0%
0.0%
30.0%
0.0%
0.0%
0.0%
30.0%
SIC
Code
286
80.0%
60.0%
10.0%
30.0%
10.0%
0.0%
30.0%
0.0%
0.0%
80.0%
70.0%
10.0%
60.0%
10.0%
0.0%
0.0%
30.0%
-------
Table 4-1 (Continued)
Release or
Waste Management
Activity Type
Stack (Cont.)
Receiving Stream
Underground Injection
Land On-Site
Source
Process Discharge Streams
Combustion By-Products
Other2
TOTAL: Reporting from at Least One Source1:
Accidental Spills/Releases
Waste Treatment Discharge Streams
Stormwater Runoff
Process Discharge Streams
Other2
TOTAL: Reporting from at Least One Source1
Process Discharge Streams
Other2
TOTAL: Reporting from at Least One Source1
Accidental Spills/Releases
Container Residue
Treatment Sludges, Recycling or Energy Recovery By-
Product
TOTAL: Reporting from at Least One Source1
Percent of Facilities Documenting Releases or
Waste Management Activity (weighted)
SIC
Code
25
0.0%
0.0%
0.0%
100.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
SIC
Code
281
0.0%
0.0%
0.0%
79.6%
7.9%
7.9%
21.1%
11.5%
0.0%
28.8%
10.1%
1.7%
10.1%
5.3%
0.0%
0.0%
5.3%
SIC
Code
285
0.0%
0.0%
0.0%
60.7%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
SIC
Code
30
0.0%
0.0%
6.8%
64.1%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
SIC
Code
26
10.0%
50.0%
0.0%
90.0%
0.0%
80.0%
20.0%
0.0%
0.0%
80.0%
0.0%
0.0%
0.0%
0.0%
0.0%
30.0%
30.0%
SIC
Code
286
0.0%
10.0%
0.0%
70.0%
10.0%
10.0%
20.0%
10.0%
10.0%
30.0%
0.0%
0.0%
0.0%
0.0%
10.0%
0.0%
10.0%
-------
                                                       Table 4-1 (Continued)
Release or
Waste Management
Activity Type
POTW
Off-Site Transfer
Source
Housekeeping Practices/Clean-up Wastes
Accidental Spills/Releases
Waste Treatment Discharge Streams
Stormwater Runoff
Process Discharge Streams
TOTAL: Reporting from at Least One Source1:
Housekeeping Practices/Clean-up Wastes
Accidental Spills/Releases
Waste Treatment Discharge Streams
Process Discharge Streams
Container Residue
Treatment Sludges, Recycling or Energy Recovery By-
Product
Combustion By-Products
Other2
TOTAL: Reporting from at Least One Source1:
Percent of Facilities Documenting Releases or
Waste Management Activity (weighted)
SIC
Code
25
0.0%
0.0%
6.1%
0.0%
9.8%
9.8%
75.6%
0.0%
0.0%
33.9%
30.9%
16.1%
0.0%
1.4%
88.7%
SIC
Code
281
14.4%
11.4%
24.4%
10.2%
14.4%
24.4%
31.9%
12.0%
20.0%
28.1%
15.3%
9.2%
0.0%
7.5%
55.5%
SIC
Code
285
12.4%
0.0%
20.6%
0.0%
12.4%
20.6%
38.0%
0.0%
0.0%
0.0%
43.3%
31.1%
0.0%
6.6%
63.1%
SIC
Code
30
9.1%
0.0%
0.0%
0.0%
9.1%
12.0%
38.6%
18.5%
0.0%
28.3%
27.4%
6.7%
0.0%
29.0%
57.5%
SIC
Code
26
0.0%
0.0%
10.0%
0.0%
0.0%
10.0%
0.0%
0.0%
30.0%
0.0%
0.0%
40.0%
10.0%
10.0%
40.0%
SIC
Code
286
0.0%
0.0%
50.0%
0.0%
30.0%
70.0%
20.0%
0.0%
30.0%
60.0%
10.0%
30.0%
0.0%
10.0%
70.0%
'Total is not additive as facilities may report a release type from multiple sources.
2Source listed as "other" include: off-spec product, uniform laundering, baghouse dust, cooling system wastewater, tank heel, sampling residue, and injection
well treatment.
-------
a
01
a?
3-

Si'
I

o
0)
3

a-
(D




I
CL

O
Percent of Facilities Documenting Releases to

                  this Type
                             Fugitive
                              Stack
                   Receiving Stream
         o
         o    Underground Injection
                     Land On-Site
                          POTW
               Off-Site Transfer
                                                                         (Q

                                                                          I
                                                                          0)

                                                                          q
                                                                          55'
                                                                          =f
                                                                          CT
                                                                          C
                                                                          o

                                                                          73


                                                                      <  |

                                                                      <0  (A
                                                                      
-------
                                                                 9-f
5T
s-
t
1
s-

I
o
3
Cr


                                                                                                                                 |T

                                                                                                                                 0
                                                                                                                                 55'
                                                                                                                             to
                                                                                                                             (O
                                                                                                                                 
-------
Co'
eg1
o
Q)
3

O-
CD

s-


Q.

§
c?
                                                        Percent of Facilities Documenting Releases

                                                               or Waste Management Activity
               Volatilization from Process Areas



                       Pumps/Valves/Flanges



               Storage Tank/Stock Pile Losses



     Housekeeping Practices/Clean-up Wastes



                            Process Vents



                 Accidental Spills/Releases



         Volatilization from Treatment Areas



      Waste Treatment Discharge Streams



                     Stormwater Runoff



            Process Discharge Streams



                   Container Residue
Treatment Sludges, Recycling or Energy

        Recovery By-Products



            Combustion By-Products




                           Other
                                                                                                                                  (Q
                                                                                                                                  O

                                                                                                                                  to'
                                                                                                                              l
                                                                                                                              0)
                                                                                                                              -»•  
-------
                                                                8-17
Q)

?
3
(n


I

i
                                                       Percent of Facilities Documenting Releases or
                                                                 Waste Management Activity
Q.
O
3
s1
I
                 Volatilization from Process Areas



                        Pumps/Valves/Flanges



f-              Storage Tank/Stock Pile Losses



       Housekeeping Practices/Clean-up Wastes



                             Process Vents



                  Accidental Spills/Releases



   O     Volatilization from Treatment Areas


   I
        Waste Treatment Discharge Streams



                      Stormwater Runoff



              Process Discharge Streams



                     Container Residue


 Treatment Sludges, Recycling or Energy

         Recovery By-Products


              Combustion By-Products



                             Other
(Q
 C




 Q.

 g

 i-r
 -^
 Er
 c
 F*
 5'


 o"

                                                                                                                             tO  rJ


                                                                                                                             %  I

                                                                                                                                 <\

                                                                                                                                 5'
                                                                                                                                 (Q
                                                                                                                                 Q)


                                                                                                                                 71
                                                                                                                                 2.
                                                                                                                                 CD
                                                                                                                                 £0
                                                                                                                                 (A
                                                                                                                                 (D
                                                                                                                                 (A
-------
Si'



I


§

I
i
-------
ID
3-
f
§
§
a
§
Cr
(D
                                                        Percent of Facilities Documenting Releases or
                                                                 Waste Management Activity
          Volatilization from Process Areas


                  PumpsA/alves/Flanges


         Storage Tank/Stock Pile Losses


Housekeeping Practices/Clean-up Wastes


                      Process Vents


            Accidental Spills/Releases


   Volatilization from Treatment Areas
  O
  JJ   Waste Treatment Discharge Streams
                      Stormwater Runoff
             Process Discharge Streams
                    Container Residue


 Treatment Sludges, Recycling or Energy
         Recovery By-Products


             Combustion By-Products
                             Other
                                                                                                                                 (Q
                                                                                                                                  C
    g
    35'

32
K  o-
•J.  C
IP  1-^
co  5'
*•  3

3  O
                                                                                                                       -A  O
                                                                                                                       CO  
-------
D
01

Q?

3-


3
I
i
I
Q.

O
3




g
                                                  Percent of Facilities Documenting Releases or Waste

                                                                   Management Activity
  o
                Volatilization from Process Areas




                       Pumps/Valves/Flanges




               Storage Tank/Stock Pile Losses




      Housekeeping Practices/Clean-up Wastes




                            Process Vents




                  Accidental Spills/Releases




          Volatilization from Treatment Areas




       Waste Treatment Discharge Streams




                      Stormwater Runoff




             Process Discharge Streams




                    Container Residue
 Treatment Sludges, Recycling or Energy

         Recovery By-Products



             Combustion By-Products




                            Other
                                                                                                                             C

                                                                                                                             3
    q

    55'
    l-K


3t
 ^  EH
co  5*
(O  3
                                                                                                                          Q.  o


                                                                                                                          5  3
CO  rf
01  O


    •o

    O
                                                                                                                              (A

                                                                                                                              31
-------
                                                               u-v
3*

3
Co'

S1
§
I
S1
§
Q.
§

s
                                                     Percent of Facilities Documenting Releases or

                                                              Waste Management Activity
             Volatilization from Process Areas



                     Pumps/Valves/Flanges



             Storage Tank/Stock Pile Losses



   Housekeeping Practices/Clean-up Wastes



                          Process Vents



               Accidental Spills/Releases



       Volatilization from Treatment Areas

to
o

£;   Waste Treatment Discharge Streams
o
CD


                   Stormwater Runoff




          Process Discharge Streams
                     Container Residue



 Treatment Sludges, Recycling or Energy

         Recovery By-Products



              Combustion By-Products
                             Other
                                                                                                                                (D
                                                                                                                                 q
                                                                                                                                 55'
                                                                                                                             to  o
                                                                                                                             tO  3

                                                                                                                             ?ff
•<  o
_i.  
-------
graphically by release type or other waste management activity for each source. Table 4-2
presents the percentage of occurrences (weighted) in which facilities incorrectly identified the
release type or other waste management activity. Figure 4-2 presents the data graphically. Table
4-3 presents the percentage of occurrences (weighted) in which various release and other waste
management sources were overlooked by facilities. Figure 4-3 presents the data graphically.

              In general, most facilities in each SIC Code reported fugitive and stack releases
and some type of transfer off-site for further waste management. Many facilities also reported
transfers to POTWs. Other release types and waste management activities including those to on-
site land were rarely observed.
4.2           Incorrectly Reported On-Site Release. Off-Site Transfer, and On-Site Waste
              Management Activity Types
              A comparison of the on-site releases, off-site transfers, and on-site waste
management activities reported by facilities and those identified by site surveyors showed that a
large number of on-site releases, off-site transfers, and on-site waste management activities were
reported to the wrong release or waste management activity type. This section discusses those
types that were incorrectly reported and presents a qualitative discussion regarding the
corresponding error in release and other waste management activity estimates. A detailed
discussion of these quantities is presented in Section 5.  Table 4-2 presents the weighted percent
of reports that had release or other waste management activity types that were incorrectly
identified.

              In many circumstances, the overall estimates that were reported were correct, but
they were assigned to the wrong type. For example, it was observed that many paint
manufacturing facilities (SIC Code 285) correctly identified, and accurately estimated, air
releases. However, the releases were incorrectly reported as stack releases (Section 5.2 of Form
R) rather than as fugitive releases (Section 5.1). The main source of this error is that state
reporting requirements and other federal reporting requirements often differ on the definition of
stack vs. fugitive emissions, causing confusion for facilities. In some instances general room  air
                                           4-13
-------
that is channeled to one vent on the building roof is considered a stack release, regardless of
whether there is an associated air pollution control device (APCD). This is considered a fugitive
release in other circumstances.  This caused confusion when facilities completed various
reporting requirements because they did not want to claim a release as a fugitive for one report
and as a stack for another. This source of error was more common at facilities that had fugitive
releases from indoor process areas without sophisticated air pollution control systems. Per TRI
guidance, the use of an APCD makes this type of release a stack emission, which coincides with
the definition of stack releases from most state requirements; thereby eliminating this source of
error. While some facilities in each SIC Code incorrectly reported stack emissions, paint
manufacturers (SIC Code 285) were the most likely to have this type of process (and
corresponding error), and this type of error was rarely observed for furniture manufacturing (SIC
Code 25) because typical facilities employed APCDs on building vents.

              Table 4-2  and Figure 4-2 also show that many facilities (in most SIC Codes)
incorrectly identified transfers off-site for further waste management (off-site recycling, off-site
disposal, off-site energy recovery, and off-site treatment). Again, it was observed that the
transfers were often correctly identified and estimated, but the reported disposition was incorrect.
One source of this error is that many facilities expressed confusion as to how toxic chemical
waste sent  off-site should be classified. Many facilities did not investigate the ultimate
disposition of the toxic chemical waste (nor felt it was their responsibility to do so). They simply
guessed as to whether the waste would be treated, recycled, or disposed. This error was not
typically observed at facilities that sent waste solvents off-site for energy recovery.  This may be
due to the fact that waste solvents are often sent off-site in large quantities (requiring large fees)
and the receiving companies rigorously test, track, and charge by the quantity received.

              There were significant quantities of transfers to POTWs. However, these transfers
were typically identified  correctly by facilities (although the estimated quantity transferred may
have been in error). This was expected because there are typically federal, state, and local limits
on the water discharged to POTWs, and most POTWs require discharge monitoring. Therefore,
facilities were aware of these discharges and had already invested time and effort to determine
their quantity and source.
                                           4-14
-------
                                           Table 4-2
Incorrectly Identified On-Site Releases, Off-Site Transfers, or On-Site Waste Management Activity Types,
                                     RY1994 and RY1995
Release or Waste Management Activity Type
Fugitive
Stack
Receiving Stream
Underground Injection
Land On-Site
POTW
Recycling (On-Site)
Treatment (On-Site)
To Off-Site Disposal
To Off-Site Energy Recovery
To Off-Site Recycle
To Off-Site Treatment
Percent of Reports Identified (weighted)
SIC
Code
25
25.7%
0:3%
0.0%
0.0%
0.0%
0.9%
3.3%
0.0%
34.4%
15.4%
25.4%
9.8%
SIC
Code
281
9.0%
18.6%
5.5%
5.1%
0.0%
4.7%
7.3%
0.5%
6.9%
1.1%
3.5%
0.8%
SIC
Code
285
11.4%
32.8%
0.0%
0.0%
0.0%
3.1%
11.8%
6.6%
36.9%
6.5%
20.7%
14.6%
SIC
Code
38
5.2%
12.8%
0.0%
0.0%
0.0%
4.2%
2.3%
6.5%
5.9%
0.0%
19.7%
0.0%
SIC
Code
2fr
9.0%
2.3%
6.8%
0.0%
4.9%
0.0%
0.0%
0.0%
4.5%
0.0%
0.0%
2.3%
SIC
Code
286
9.4%
4.7%
4.7%
0.0%
0.0%
6.3%
0.0%
0.0%
6.5%
1.6%
0.0%
3.2%
-------
s
In'


I
Cr



                                                                    5"
                                                                    o
                                                                    o
                                                                    T

                                                                    o
                                                                    1-f
                                                                    ^

                                                                    S
                                                                    (D
                                                                    3
                                                                    
                                                                §
                                                                    (D
                                                                    tf)
                                                                    0)
Q.

7)
                                                                (O
                                                                
-------
              Releases to receiving streams, underground injection wells, on-site land and on-
site energy recovery were rarely observed during the site visits. Therefore, the potential for
incorrectly identifying releases or on-site energy recovery was low or non-existent.

4.3           Overlooked Releases and Other Waste Management Activities

              As shown on Table 4-3 and Figure 4-3, several facilities overlooked some releases
and other waste management activities entirely. In some cases this resulted in an
underestimation of the overall quantity of the toxic chemical managed as waste by the facility.
However, in cases where a mass balance was used as the method to determine the quantity of the
toxic chemical managed as waste, the facility may have included the quantity that was
overlooked in another release type.  For example,  a facility may have overlooked a release or
transfer off-site from container residual. However, after conducting a material balance and
analyzing the total throughput and quantifiable releases and other waste management activities,
this quantity may have been unaccounted for.  The facility may have assumed this quantity was
released from process areas as fugitive emissions. In this case, the release or transfer off-site to
one type would have been under reported, while the fugitive air emissions would have been over
reported.

              It was observed that the primary overlooked sources were from chemicals sent
off-site as container residue (typically as liquid residue in "empty" drums), stack emissions of
volatile chemicals from on-site storage tanks, liquid discharges to POTWs or receiving streams
(from aqueous washwater and spent solvents from waste cleaning materials), and fugitive
releases from process areas and process lines (pumps, valves, and flanges).

              The largest source of overlooked releases and other waste management activities
(both frequency and overall quantity) was from container residue. Although the  EPCRA Section
313 instructions specify that container residue should be considered as  a release,  most facilities
assumed that all used drums, totes,  or small containers were completely empty and the
subsequent transfer of the empty containers off site (or the disposal on site) did not result in any
release or transfer of EPCRA Section 313 chemicals. Many facilities did not consider the
potential for reportable quantities of residual chemicals in these containers. Other facilities
                                          4-17
-------
                                                     Table 4-3

                        Overlooked Release and Other Waste Management Activity Sources
                                               RY1994 and RY1995
oo
Source
Container Residue
Storage Tank/Stock Pile Losses
Housekeeping Practices/Clean-up Wastes
Pumps/Valves/Flanges
Volatilization from Process Areas
Process Vents
Volatilization from Treatment Areas
Accidental Spills/Releases
Waste Treatment Discharge Streams
Process Discharge Streams
Treatment Sludges, Recycling or Energy Recovery By-
Products
Combustion By-Products
Other1
Percent of Reports Identified (weighted)
SIC
Code
25
61.4%
0.0%
16.6%
12.2%
12.9%
0.0%
0.0%
0.0%
0.0%
21.5%
0.0%
0.0%
0.0%
SIC
Code
281
17.7%
22.9%
5.3%
8.6%
13.0%
12.3%
7.5%
3.4%
0.0%
10.1%
0.0%
0.0%
3.4%
SIC
Code
285
49.0%
32.1%
16.3%
14.3%
24.5%
0.0%
0.0%
7.2%
8.2%
0.0%
12.3%
0.0%
0.0%
SIC
Code
30
50.3%
2.1%
27.9%
2.1%
14.0%
0.0%
0.0%
0.0%
0.0%
9.1%
0.0%
0.0%
15.9%
SIC
Code
26
10.0%
10.0%
0.0%
10.0%
10.0%
0.0%
20.0%
10.0%
0.0%
0.0%
0.0%
10.0%
0.0%
SIC
Code
286
30.0%
10.0%
0.0%
10.0%
30.0%
0.0%
10.0%
0.0%
10.0%
0.0%
0.0%
0.0%
0.0%
                 'Source of "other" include: baghouse dust and repackaging losses.
-------
                                                                  61-17
sr

$
5
-------
considered this potential release or waste management quantity but felt it was negligible (and did
not report it) if drums were shipped as "empty", as defined by federal and/or state shipping
regulations.

             In practice, liquids are often removed from drums by gravity draining or by
pumping. Neither of these methods removes all material from the drum and an appreciable
quantity may remain. Additionally, some Department of Transportation (DOT) and RCRA
Regulations require special handling precautions when transporting drums containing hazardous
materials. Drums that once contained these materials that have been emptied may be exempt
from these regulations.  The definition can vary, but drums are often defined as empty for
shipping purposes if they contain less than two inches of a liquid substance.  Therefore, facilities
often empty drums to comply with these regulations, but they do not completely empty them,
due to economical considerations. It should be noted that many facilities sent hundreds of
"empty" drums off-site and that if each drum contained some residual chemical, a significant
quantity of release and otherwise managed was overlooked. Additionally, many facilities
overlooked releases and other waste management quantities due to residual powdered EPCRA
Section 313 chemicals in empty bags.

              Most of the liquid releases and other waste management quantities from
overlooked container residue should have been reported as off-site transfers to a disposal facility.
However, some should have been reported to off-site recycling, off-site treatment,  or off-site
energy recovery. Other overlooked liquid discharges should have been reported to  either
POTWs or to receiving streams because the drums were rinsed on site and the rinsewater was
collected and disposed to the local POTW or receiving stream. Most overlooked solid releases
from bag residue should have been reported as being disposed to on-site landfills or to off-site
disposal.

              Volatilization from treatment areas was rarely overlooked.  This is presumably
because most treatment chemicals are either non-volatile or are completely destroyed during the
treatment process.
                                          4-20
-------
              Accidental spills and releases were also rarely overlooked.  However, a number of
facilities questioned the definition of a spill and requested additional guidance.  The primary
points of confusion pertained to the quantity and frequency of occurrence of spills. For example,
it is common that paint manufacturing facilities have "spills" of paints containing EPCRA
Section 313 chemicals or solvents used to make the paints on a daily basis. These "spills" are
typically collected  and sent to disposal, or the EPCRA Section 313 chemical was assumed to be
volatilized and lost as a fugitive emission. Most facilities did not claim this as a catastrophic
release in  Section 8.8. However, they were unsure how to estimate the quantity and how to
report it. Additionally, these "spills" are typically small (drippings that are less than one liter per
occurrence).  However, occasionally a pail, barrel, or drum may be knocked over. In these cases,
facilities have asked for guidance as to when this should be reported as a catastrophic release
rather than a "typical" release from the process.

              Site surveyors did not identify any overlooked releases or other waste
management activities from combustion by-products at facilities in SIC Codes 25, 281, 285, or
30. This can be attributed to the fact that very few of these facilities utilized on-site boilers,
industrial  furnaces, or incinerators. Those that did (typically chemical facilities or furniture
manufacturers) used clean fuels (such as natural gas) that generated quantities of EPCRA Section
313 chemicals that were below reporting thresholds.  However, some facilities in SIC Codes 26
and 286 that used coal and/or fuel oil for on-site boilers overlooked the incidental manufacture of
EPCRA Section 313 chemicals from these units.

              The pulp and paper industry often uses on-site boilers or recovery furnaces for the
destruction of unwanted byproducts and the concurrent generation of steam for use in the
manufacturing process.  Coal and/or fuel oil are typical fuels for these units. Combustion of
these fuels can result in the coincidental manufacture and subsequent release of EPCRA Section
313 chemicals above the reporting threshold. Some facilities overlooked this potential
manufacture and release. EPCRA Section 313 chemicals that were manufactured above the
threshold, but overlooked included sulfuric acid (acid aerosols) and hydrochloric acid (acid
aerosols).  Formaldehyde was also manufactured in appreciable quantities, and overlooked, by
                                          4-21
-------
some facilities. However, it was not manufactured above the reporting threshold at any sites that
were visited. It is expected that facilities in other SIC Codes that utilize large coal or fuel oil
burning recovery boilers may have also overlooked this release source.

4.4           Calculation Methodologies

              EPA requires facilities to designate one of four categories of calculation
methodology that were used for each release or other waste management activity estimate
(monitoring data, mass balances, emission factors, and engineering judgment or calculations).
Table 4-4 presents the distribution of calculation methodologies that were used (weighted) to
determine estimates for each release or other waste management activity type. It was observed
during the review of facility notes that facilities  often used multiple methods or reported a
method that was inconsistent with the method actually used.  Therefore, the data reported in
Table 4-4 represents the site surveyor's opinion  as to the primary method actually used by the
facility, not necessarily the method reported on the facility's Form R.  This allows for analysis of
data accuracy when compared to the actual methods used. Additionally, a significant number of
facilities used hazardous waste manifests to calculate estimates of off site  transfers. Site
surveyors noted these occurrences and their frequency of use is presented along with the four
EPA-accepted methods when applicable.  Figures 4-4a through 4-4m present the calculation
methodology data graphically. There was considerable difference in the methodologies used
between each SIC Code and in those used within SIC Codes for each release or other waste
management activity type. It should be noted that there were few or no releases reported to
several release types.  In these circumstances the table and corresponding figures currently show
0%.

              Nearly all facilities reported at least one fugitive release.  Site surveyors observed
that fugitive releases were typically the most difficult for facilities to estimate. Engineering
calculations, as presented in Table 4-4 and the corresponding figures,  are the predominant
method used by most facilities. Site surveyors observed that many  facilities actually used one or
more of the methods to estimate fugitive emissions, and applied engineering judgement to total
the emissions from all sources. This included engineering judgement for partitioning releases
between stack and fugitive if monitoring data was not  available. Mass balances, monitoring data,
                                           4-22
-------
                                                Table 4-4
                       Distribution of Calculation Methodologies, RY 1994 and RY 1995
Release or
Other Waste Management
Activity Type
Fugitive
Stack
Receiving Stream
Underground Injection
Land On-Site
Calculation Methodology
Engineering Calculations
Mass Balance
Monitoring Data
Emission Factors
Other1
Engineering Calculations
Mass Balance
Monitoring Data
Emission Factors
Other1
Engineering Calculations
Mass Balance
Monitoring Data
Emission Factors
Other1
Engineering Calculations
Engineering Calculations
Monitoring Data
Emission Factors
Percent of Facilities Documenting Releases (weighted)
SIC
Code
25
26.8%
62.1%
0.0%
1.8%
9.3%
19.4%
72.1%
0.0%
1.7%
6.8%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
SIC
Code
281
59.5%
1.6%
6.8%
30.9%
1.2%
62.0%
3.3%
17.8%
11.1%
5.7%
18.4%
0.0%
81.6%
0.0%
0.0%
100.0%
100.0%
0.0%
0.0%
SIC
Code
285
62.4%
8.7%
0.0%
21.5%
7.5%
59.7%
11.0%
9.7%
19.7%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
SIC
Code
30
38.5%
23.6%
0.0%
34.5%
3.4%
52.5%
24.2%
5.8%
14.1%
3.4%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
SIC
Code
26
40.7%
0.0%
0.0%
51.9%
7,4%
15.1%
3.0%
12.1%
69.7%
0.0%
36.7%
3.3%
33.3%
20.0%
6.7%
100.0%
60.0%
20.0%
10.0%
sic
Code
286
22.6%
9.4%
15.1%
52.8%
0.0%
46.3%
0.0%
11.1%
40.7%
1.9%
13.0%
8.7%
78.3%
0.0%
0.0%
0.0%
100.0%
0.0%
0.0%
*>.
OJ
-------
                                          Table 4-4 (Continued)
                       Distribution of Calculation Methodologies, RY 1994 and RY 1995
Release or
Other Waste Management
Activity Type
Land On-Site (Cont.)
POTW
To Off-Site Disposal
To Off-Site Treatment
To Off-Site Recycle
Calculation Methodology
Other1
Engineering Calculations
Mass Balance
Monitoring Data
Engineering Calculations
Mass Balance
Monitoring Data
Emission Factors
Other1
Engineering Calculations
Mass Balance
Monitoring Data
Hazardous Waste Manifests
Emission Factors
Other1
Engineering Calculations
Monitoring Data
Hazardous Waste Manifests
Percent of Facilities Documenting Releases (weighted)
SIC
Code
25
0.0%
15.7%
84.3%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
87.5%
0.0%
0.0%
12.5%
4.9%
0.0%
92.3%
SIC
Code
281
0.0%
40.8%
37.7%
21.6%
22.2%
12.1%
65.7%
0.0%
0.0%
21.1%
10.5%
34.2%
34.2%
0.0%
0.0%
0.0%
100.0%
0.0%
SIC
Code
285
0.0%
100.0%
0.0%
0.0%
92.3%
6.8%
0.9%
0.0%
0.0%
71.1%
0.0%
28.9%
0.0%
0.0%
0.0%
63.3%
29.6%
7.1%
SIC
Code
30
0.0%
0.0%
0.0%
100.0%
47.3%
53.7%
0.0%
0.0%
0.0%
27.2%
33.2%
39.5%
0.0%
0.0%
0.0%
73.3%
26.7%
0.0%
SIC
Code
26
10.0%
0.0%
0.0%
100.0%
42.9%
28.6%
0.0%
14.3%
14.3%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0.%
SIC
Code
286
0.0%
26.8%
9.8%
63.4%
75.0%
0.0%
25.0%
0.0%
0.0%
4.8%
0.0%
95.2%
0.0%
0.0%
0.0%
0.0%
100.0%
0.0%
to
-------
                                                      Table 4-4 (Continued)
                             Distribution of Calculation Methodologies, RY 1994 and RY 1995
Release or
Other Waste Management
Activity Type
To Off-Site Recycle (Cont.)
To Off-Site Energy Recovery
On-Site Treatment
On-Site Energy Recovery
On-Site Recycling
Calculation Methodology
Other1
Engineering Calculations
Mass Balance
Monitoring Data
Hazardous Waste Manifests
Other1
Engineering Calculations
Mass Balance
Monitoring Data
Emission Factors
Engineering Calculations
Mass Balance
Monitoring Data
Emission Factors
Engineering Calculations
Mass Balance
Monitoring Data
Other1
Percent of Facilities Documenting Releases (weighted)
SIC
Code
25
2.7%
19.9%
13.9%
31.6%
23.2%
11.4%
0.0%
100.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
18.2%
3.2%
33.2%
45.3%
SIC
Code
281
0.0%
71.6%
0.0%
0.0%
28.4%
0.0%
33.5%
32.8%
29.0%
4.6%
0.0%
0.0%
0.0%
0.0%
47.5%
0.0%
52.5%
20.5%
SIC
Code
285
0.0%
8.1%
0.0%
89.8%
2.1%
0.0%
100.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
48.7%
21.1%
26.5%
3.6%
SIC
Code
30
0.0%
16.7%
0.0%
0.0%
83.3%
0.0%
20.9%
0.0%
79.1%
0.0%
0.0%
0.0%
0.0%
0.0%
92.0%
0.0%
8.0%
0.0%
SIC
Code
26
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
30.3%
30.3%
18.2%
21.2%
42.9%
28.6%
0.0%
28.6%
100.0%
0.0%
0.0%
0.0%
SIC
Code
286
0.0%
4.5%
13.6%
81.8%
0.0%
0.0%
29.4%
5.9%
64.7%
0.0%
0.0%
0.0%
100.0%
0.0%
71.4%
0.0%
28.6%
0.0%
to
     1 "Other" methodologies according to facility notes include: hazardous waste manifests, off-site facility test reports, facility or trade association computer
     modeling, air permit limits, and "undocumented".
-------
                                                     9Z-P
£


{
o
3-

I
                                               Percent of Release Estimates Made Using

                                                        Specified Methodology
          o
          C
          5T
                     Engineering Calculations
                            Mass Balance
                         Monitoring Data
          o   Hazardous Waste Manifests
          o
                      Emission Factors
                             Other
                                                                                                            

    q
    55'
    q
                                                                                                         31
a E
3 a
-< o
-t 3
(O 9
                                                                                                        01
                                                                                                           o
                                                                                                           a
                                                                                                           p_
                                                                                                           o
                                                                                                           (Q
                                                                                                           (D*
                                                                                                           (A
                                                                                                           C
                                                                                                           -------
o
u
sr

S
3

n
0)
3

Cr
-------
O
0)

5J'
3-
fa'
ai


I
o
ffi
3
O-
  -

Q.

i
                                                    Percent of Release Estimates Made Using

                                                             Specified Methodology
                                                                                                     Tl
                                                                                                     (5*
                                                                                                     c
            o
            o>
            cT
            o
            (D



            I
                       Engineering Calculations
              Mass Balance
                            Monitoring Data
Hazardous Waste Manifests
                         Emission Factors
                                 Other
                                                                                                                      o
    5.
    D1
    C
    E*
    O
    SL
(O  O

                                                                                                                      3
-------
o
ia
iu
I
O-
CD
Q.

O
 T
.u
-k
                                              Percent of Waste Management Estimates

                                                 Made Using Specified Methodology
          2?
          o
                    Engineering Calculations
                           Mass Balance
                         Monitoring Data
               Hazardous Waste Manifests
                      Emission Factors
                             Other
  (Q
   C
   k
   a
   O"
   C
   tf.
   o
   •3
   O
   ^t

7J 0)_
•< o

it
»§
i
(o S.
01 o
   (Q
   5'
   (A
                                                                                                        Q.
                                                                                                        (D

                                                                                                       (3

                                                                                                        o

                                                                                                        •3
                                                                                                        Q.
-------
                                                    0£~t7
D
SI
5?
I
O
01
3
cr
 3
 Q.

 §
 o-
 5"
Percent of Release Estimates Made Using
         Specified Methodology
2
o
5T
i-i*
5'
                      Engineering Calculations
                              Mass Balance
                            Monitoring Data
                 Hazardous Waste Manifests
                         Emission Factors
                                 Other
                                                                                                  (O
                                                                                                   c
                                                              CD
                                                              q

                                                              I.
                                                              cr
                                                              (-*•
                                                              o'
                                                           SI
                                                               1
                                                            to   -
                                                            -------
D

Si
ST
5'
I
D-
(D


3-
§
-------
SJ.
ST
o

s
o-


31


Q.

i
                                                   Percent of Transfer Estimates Made Using

                                                             Specified Methodology
                                                                                       (Q

                                                                                        C
           S?
           o
           o
           o_

           o
                       Engineering Calculations
Mass Balance
                            Monitoring Data
                Hazardous Waste Manifests
                        Emission Factors
                                 Other
                                                                                       (Q


                                                                                        a
                                                                                        w1

                                                                                        a
                                                                                        cr
                                                                                                                  3 °
                                                                                                                  -< a,

                                                                                                                  -*• o
                                                                                                                  
                                                                                                                      0)
-------
Cr
CD
Q.
i
s1
la-
                       Engineering Calculations
                               Mass Balance
            g-               Monitoring Data
            s

            (0
            »*
            o   Hazardous Waste Manifests




                          Emission Factors




                                  Other
Percent of Transfer Estimates Made Using
          Specified Methodology
                                                                  (Q
                                                                  C
                                                                   0

                                                                   $
                                                                   CT
                                                                   C
                                                                   A
                                                                   O
                                                                   3
                                                               -* c
                                                               (O —
                                                               to o>
                                                                j. O
                                                                to Q.
                                                                (O O
                                                                w 5-
                                                                   (Q
                                                                   5'
                                                                   (A
                                                                                                                        (D
                                                                                                                        SJ-

                                                                                                                        (D
-------
0>

?

3
=!]

   a
   cr
7)  O

•<  9L
-».  o
(O  C
s
    o
    (Q
    5'
    (A
                                                                                                                        70
                                                                                                                        (0
                                                                                                                        o
                                                                                                                        (D
-------

                                                                                                  o"
                                                                                                  -
                                                                                               tO °
                                                                                               5 3

                                                                                               II
                                                                                                    .
                                                                                               •< O
                                                                                               -»• O
                                                                                               {g«e.
                                                                                               01
                                                                                                                ff
                                                                                                                m
                                                                                                                3
                                                                                                                0
                                                                                                                (2
                                                                                                                (D
                                                                                                                O
                                                                                                                 (D
-------
f
o-
CD
a
o

i
                     Percent of Waste Management Estimates

                        Made Using Specified Methodology
           o
           £L
           o

           5"
           §•
           o
                      Engineering Calculations
Mass Balance
                           Monitoring Data
                Hazardous Waste Manifests
                        Emission Factors
                                Other
                                                                                    q
                                                                                    tn
                                                                                    a
                                                                                    cr
                                                                                 -• c
                                                                                 to £.
                                                                                 
-------
D-

CD
a.
o
£
o"

f-
                                                 Percent of Waste Management Estimates

                                                     Made Using Specified Methodology
                      Engineering Calculations
                              Mass Balance
                            Monitoring Data
                Hazardous Waste Manifests
                         Emission Factors
                                 Other
  -n
  <5i


  3
   q
   v>

   3.
   &
   c.
   i-^
   6'
   o
   SL
   o
<0 3

* S


=
                                                                                                              -*
                                                                                                              S«s.
                                                                                                              01
                                                                                                                 o

                                                                                                                 O
                                                                                                                  m
                                                                                                                  3
                                                                                                                  to

                                                                                                                 a
                                                                                                                 <

                                                                                                                  7)
                                                                                                                  (D
                                                                                                                  O
                                                                                                                  O
-------
a-
5'



f
o
§
o-
CD
O.
o
                                                   Percent ol Waste Management Estimates

                                                      Made Using Specified Methodology
                                                                                                                   (Q
           r>
           c_
           fit
           *4>
           o
           a.
           o_
           o
          5
                      Engineering Calculations
                              Mass Balance
Monitoring Data
                Hazardous Waste Manifests
                        Emission Factors
                                Other
                                                                                        q
                                                                                        55'
                                                                                        a
                                                                                        JT
                                                                                        i-i-
                                                                                        5'
                                                                                                                7J O
                                                                                                                •< SL
                                                                                                                
-------
and emission factors have been presented in Table 4-4 only when they were the predominant
method used.

              It was uncommon for facilities to have access to monitoring data for fugitive
releases.  However, it was used when available  (typically in the form of periodic leak tests).
Emission factors were used by several facilities (except in SIC Code 25). The type of emission
factors used and a subsequent discussion is presented below. Mass balances were also used by
many facilities to determine fugitive releases from at least one process line or unit operation
when a material balance around the entire facility resulted in a quantity of chemical that was
unaccounted for.

              Most facilities reported a stack release. Although facilities had difficulty in
estimating these releases, they typically indicated less difficulty in identifying and quantifying
these releases than observed with fugitives. Engineering calculations and mass balances were the
most often used methods. However, the use of emission factors (including emission factors for
releases from storage tanks and facility-derived factors for releases from stacks) and monitoring
data (actual releases from stack tests) were often observed. Facilities with sophisticated
monitoring equipment associated with large stack emissions, such as chemical manufacturers and
pulp and paper mills, were more likely to use monitoring data and/or associated emission factors
than smaller facilities, such as paint manufacturers which typically used mass balances or
engineering judgement.

              Table 4-4 and the  corresponding figures also show that most facilities used
monitoring data and/or hazardous waste manifests in conjunction with engineering calculations
to estimate transfers off-site for further waste management. There were two main sources of
these data.  One was from periodic facility sampling of the waste that was collected prior to
shipment. The second was from  sampling conducted by the receiving facility. Documentation
for this data was typically more prevalent and more complete (and therefore, presumably more
accurate) than methods used to estimate releases to most other sources.
                                          4-39
-------
      Figure 4-5. Frequency the Facility Used the Best Methodology to Estimate Releases
                           and Other Waste Management Activities
-t^
o
100%
 90%
 80% - -
 70%
 60% -
 50% -
 40%
 30%
 20%
 10%
        SIC Code 25    SIC Code 281    SIC Code 285     SIC Code 30    SIC Code 26     SIC Code 286
-------
              Some facilities reported discharges to POTWs and/or receiving streams
(predominantly organic chemical manufacturer and pulp and paper mills, which have large water
releases).  A mass balance and/or engineering judgement were the primary methods used for
POTW discharges. Facilities typically used a mass balance around the entire facility to
determine the quantity of EPCRA Section 313 chemical that could not be accounted for. Then,
engineering judgement (usually based on knowledge of chemical volatility and solubility) was
used to estimate a partition factor between releases of the unaccounted quantity that would be
lost to fugitive air vs. that sent to a POTW. This method was also used to determine discharges
to receiving streams, when applicable. However, discharges to receiving streams were often
monitored for compliance with various local, state, or other federal regulations, resulting in a
more accurate estimate.

              Figure 4-5 presents the frequency that site surveyors felt the method used by the
facility would result in the most accurate estimate of release or other waste management quantity
based on information and data available to the surveyor at the time of the site visit. It does not
present the frequency that the facility correctly calculated the quantity of release or other waste
management activity. This figure shows concurrence with the selected method in most cases.  It
should be noted that during many visits the surveyor identified another, more accurate method
that could have been used to estimate releases and other waste management quantities, if a
particular variable had been tracked for the reporting year.  In many cases, the facility contact
indicated that it would have been fairly easy for the facility to implement the suggestion and that
they planned to take the surveyors advice for subsequent years. However, there was no way to
recreate the required variable for the reporting year(s) surveyed. Another limitation to this
analysis is the fact that surveyors often identified a more accurate method that could be used
based on data the facility claimed to have. However, the facility stated that they could not gather
the information in a reasonable time period for use by the site surveyor.

              Emission factors were frequently used to estimate fugitive and stack releases.
EPA instructed site surveyors to determine the type of emission factors used, when applicable.
The potential types were designated as facility-derived, EPA-approved or published, trade
                                          4-41
-------
                       Table 4-5
Types of Emission Factors Used for Fugitive and Stack Releases
                  RY1994 and RY1995
Release and Other Waste
Management Activity Type
Fugitive
Stack
Release and Other Waste
Management Activity Source
Facility Derived
EPA Derived
Trade Association Derived
Other
TOTAL:
Facility Derived
EPA Derived
Trade Association Derived
Other
TOTAL:
Percent (weighted by chemical)
sic
Code 25
93.9%
0.0%
0.0%
6.1%
100.0%
0.0%
61.6%
0.0%
38.4%
100.0%
SIC
Code 281
26.9%
8.4%
51.8%
12.9%
100.0%
53.2%
38.4%
0.0%
8.4%
100.0%
SIC
Code 285
38.7%
61.3%
0.0%
0.0%
100.0%
82.7%
17.3%
0.0%
0.0%
100.0%
SIC
Code 30
35.3%
27.6%
37.1%
0.0%
100.0%
81.9%
12.9%
5.2%
0.0%
100.0%
SIC
Code 26
5.0%
0.0%
90.0%
5.0%
100.0%
10.7%
3.6%
85.7%
0.0%
100.0%
SIC
Code 286
13.9%
41.7%
36.1%
8.3%
100.0%
33.3%
56.7%
10.0%
0.0%
100.0%
-------
                         Figure 4-6a. Type of Emission Factors Used (Fugitive)
                                          RY1994 and RY1995
  12.9%
                       26.9%
                                                   8.3%
                                                                 38.7%
                                                                   13.9%
51.8%
    SIC Code 281 (RY
      6.1%
                            '8.4%
                93.9%
     SIC Code 25 (RY 1994)
                                    61.3%
           SIC Code 285 (RY 1994)
                                         37.1%
                                                              35.3%
         27.6%

SIC Code 30 (RY 1994)
                                                                        41.7%
                                                                           36.1%
                                        SIC Code 286 (RY 1995)
                                                                                       5.0%     5.0%
                                                         90.0%

                                                   SI C Code 26 (RY 1995)
     rj Facility Derived
I EPA Derived
               I Trade Association Derived
I Other
-------
                             Figure 4-6b.  Type of Emission Factors Used (Stack)
                                            RY1994 and RY1995
          8.4%
38.4%
       SIC Code 281 (RY
                           53.2%
                                                                                      10.6%
         SIC Code 285 (RY 1994)
                                                                                                         35.4%
                                                                              53.9%
                         SIC Code 286 (RY 1995)
  38.4%
                                                 5.2%
                            61.6%
       SIC Code 25 (RY 1994)
                                      12.9%
                                                               81.9%
          SIC Code 30 (RY 1994)
                                                                                                    10.7%
                                                                                                           3.6%
                                                                                   85.7%
                         SIC Code 26 (RY 1995)
       rj Facility Derived
EPA Derived
i Trade Association Derived
I Other
-------
association-derived, and other. Table 4-5 and Figures 4-6a and 4-6b present the percentage of
use for each type of emission factor (weighted).

              These factors were typically employed to estimate fugitive releases of volatile
chemicals from process areas (open mix tanks or vats) or piping (leaks from pumps, valves,
flanges, etc.) or to estimate stack releases from storage tanks and stack releases from gasses
generated by unit operations that were channeled through stacks (typically stacks from various
air pollution control devices).
4.5           On-Site Waste Management Activities frecvcling. treatment and energy
              recovery)
              Quantities of the toxic chemicals in waste managed by on-site waste management
activities (recycling, treatment, and energy recovery) were rarely observed during the site visits.
Table 4-2 and Figure 4-2 show that a considerable number of facilities in SIC Codes 26 and 286
incorrectly identified these releases. It should be noted that there may be considerable
uncertainty in the quantitative values presented because most facilities were confused by the
definition of "recycling".  EPA recognized, before the RY 1994 and RY 1995 site surveys were
initiated, that this potential might exist and instructed site surveyors only to analyze releases to
recycling activities if the facility reported them. Therefore, site surveyors only recorded on-site
waste management  activities as incorrect if such activities were claimed but did not exist.
Facilities typically correctly identified on-site treatment activities when they existed. However,
there was considerable confusion and error when the releases and other waste management
activities were quantified for Section 8 of the Form R.

              Additionally, for the site visits completed through May 1997 (those pertaining to
RY94) EPA only asked site surveyors to compile the data for source reduction and on-site
recycling.  Site surveyors were not requested to investigate these issues further or discuss them
with the facility contacts.  Therefore, specific, quantitative input from these visits cannot be
provided, other than raw data based on what was reported by each facility.
                                           4-45
-------
              Additionally, EPA was concerned that facilities may incorrectly report the
quantity of EPCRA Section 313 Chemicals to on-site treatment in Section 8 of the Form R due to
potential confusion between requirements for Section 7 and Section 8. Site surveyors
specifically determined whether the quantities reported were quantities sent to treatment or
quantities actually treated.  Figure 4-7 presents the weighted percent of facilities that incorrectly
reported the quantity sent to treatment rather than that actually treated. A significant number of
facilities in SIC Codes 26 and 286 incorrectly reported this quantity (25.8%  and 48.6%,
respectively) while only one facility incorrectly reported from SIC Code 281 (representing 1.2%)
and no facilities from SIC Codes 25, 30, or 281 incorrectly reported.
               Figure 4-7.  Facilities Incorrectly Reporting the Quantity sent to Treatment
                           Rather than that Actually Treated (Weighted)
           60%
           50%
           40%
           30%
      o
      o
           20%
      §
      ffi
      Q.
10%
           0%
                                                                        48.6%
                  0.0%
                  0.0%
                                        1.2%
0.0%
                SIC Code 25   SIC Code 281  SIC Code 285   SIC Code 30   SIC Code 26   SIC Code 286
              It was observed that many facilities in the organic chemical and the paper
manufacturing SIC Codes, 286 and 26, respectively, had large on-site wastewater treatment
plants. Several EPCRA Section 313 chemicals (such as ammonia, sulfuric acid, and chlorine)
were used in the treatment process.  Several facilities incorrectly reported that these chemicals
were treated themselves because they were destroyed during the treatment process. EPCRA
Section 313 chemicals added to waste treatment units are not considered to be treated
themselves. This situation was never observed at facilities in SIC Codes 25, 281, 285, or 30
because these facilities did not typically have on-site wastewater treatment plants or the quantity
of treatment chemical used was below the threshold.
                                           4-46
-------
              EPA requested site surveyors to discuss source reduction and on-site recycling

issues and acquire feedback from facility contacts during visits to SIC Codes 26 and 286 (those

pertaining to RY 1995).


              The following points were raised by facility contacts during the 20 visits to Paper

and/or Organic Chemical facilities for RY 1995.
              •      Facilities tend to only claim source reduction or recycling if they
                     implement a procedure for the specific purpose of reducing releases.
                     Often a facility implements one or more of the items that EPA considers
                     source reduction, but they do not bother to go through the entire list to see
                     if they can claim it. For example, a facility may change their raw material
                     transfer operations due to a management decision. This change may result
                     in source reduction as a side effect, but the facility does not claim it
                     because the purpose was not source reduction.

              •      Facilities often do not claim source reduction or recycling activities due to
                     what they consider to be  a lack of detailed definitions. For example,
                     facilities believe that many of the codes and corresponding descriptions
                     are vague and they do not feel comfortable claiming an activity without
                     better guidance.

              •      Facilities have stated that they would rather be conservative and only
                     claim a source reduction  or recycling activity if they can verify and
                     document it. For example, some categories such as "better management
                     practices" are vague and  facilities do not claim it because they do not
                     know how to verify it.

              •      Some trade associations instruct their members not to claim a source
                     reduction activity unless  they can document a corresponding reduction in
                     releases - even if the facility specifically installs a unit or practice that is
                     intended to serve as recovery or recycling.

              •      Facilities have stated that there are so many codes that they do not bother
                     to analyze each code every year to see if any changes in their processes
                     apply.

              Tables 4-6 through 4-11 summarize data that was collected for on-site recycling

that was observed during site visits pertaining to RY 1994 and RY 1995.  Table 4-11  presents the

frequency that each chemical was recycled, as reported by these facilities.
                                          4-47
-------
             Table 4-6
Observed On-Site Recycling Activities
          (SIC Code 281)
# Of Facilities
Reporting
1
1
1
2
3
1
1
Type of Recycling
Claimed
Cleaning Waste
Spent Process Solvent
Other
Other
Other
Other
Other
Description of Recycling Stream
Not Specified
Not Specified
Scrubber Water
Dust Collector Waste
Off-Spec. Product
Ion Exchange Waste
Vapor Recovery Unit
             Table 4-7

Observed On-Site Recycling Activities
          (SIC Code 285)
# Of Faculties
Reporting
5
5
Type of Recycling
Claimed
Cleaning Waste
Spent Process Solvent
Description of Recycling Stream
Not Specified
Not Specified
                4-48
-------
             Table 4-8
Observed On-Site Recycling Activities
           (SIC Code 25)
# Of Facilities
Reporting
1
1
1
1
1
1
Type of Recycling
Claimed
Spent Process Solvent
Spent Process Solvent
Spent Process Solvent
Other (obsolete material)
Cleaning Waste
Cleaning Waste
Description of Recycling Stream
Distillation Unit
Not Specified
Batch Still and Thin-Film Evaporation
Batch Still
Not Specified
Batch Still and Thin-Film Evaporation
             Table 4-9

Observed On-Site Recycling Activities
           (SIC Code 30)
# Of Facilities
Reporting
1
1
1
Type of Recycling
Claimed
Spent Process Solvent
Other
Other (off-spec product)
Description of Recycling Stream
Not Specified
Resin from waste plastic
Reuse in subsequent batch
                4-49
-------
             Table 4-10
Observed On-Site Recycling Activities
          (SIC Code 286)
# Of Faculties
Reporting
1
4
1
3
Type of Recycling
Claimed
Not specified
Spent Process Solvent
Other
Other (unreacted raw
materials)
Description of Recycling Stream
Removed Stack Emissions
Not Specified
Scrubber Water
Process Waste (off-spec product)
                4-50
-------
                Table 4-11
Chemicals For Which Recycling Was Claimed
(SIC Codes 281,285,25,30, and 286 Combined)
CHEMICAL
Xylene
Toluene
Methanol
MffiK
1,3 -Butadiene
Ammonia
Ethylbenzene
Ethylene glycol
Glycol enters
2-ethoxyethanol
4,4'-isopropylidenediphenol
Ammonium Nitrate
Aniline
Chlorine
Copper Compounds
Cyanide Compounds
Di-(2-ethylhexyl)Phthalate
Dichloromethane
Freon
HC1
Mercury
N-butyl alcohol
N-butyl alcohol
Nitric Acid
N-N-Dimethylaniline
Phosphoric Acid
# CW FACILITIES 8EFOHHNG
7
5
3
3
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
                   4-51
-------
-------
5.0           RELEASES AND OTHER WASTE MANAGEMENT ACTIVITIES

5.1           On-Site Release and Transfers Off-Site for Waste Management Estimates

              Release and other waste management estimates are the most highly scrutinized
and publicized data in the TRI program.  This section and section 5.2 discuss release estimates
and other waste management activities made by facilities and by site surveyors. Major
differences in these estimates between the facilities and the site surveyors are noted and, if
possible, the reasons for the differences are explained. A discussion of the methodology used by
the site surveyors to gather the data necessary to estimate these quantities is contained in Section
2. A discussion of the specific techniques used by the facilities and by the site surveyors when
estimating releases and other waste management quantities is contained in Section 4.

              Releases and transfers off-site for waste management estimates are reported by
chemical and by the medium to which the chemical was released or transferred. When
completing the Form R, facilities must assign on-site releases to one of the following five
categories:

                    —    Fugitive or non-point air emissions;
                    —    Stack or point air emissions;
                    —    Discharges to receiving streams or water bodies;
                    —    Underground injections on site; or
                    —    Releases to land on site.
Transfers to other off-site locations for other waste management practices are further subdivided
into:
                    —    Discharges to Publicly Owned Treatment Works (POTWs);
                    —    Off-site transfer for disposal;
                    —    Off-site transfer for treatment;
                    —    Off-site transfer for recycling; and
                    —    Off-site transfer for energy recovery.
This section also contains a discussion of releases and other waste management practices to each
medium, how facility estimates compared to site surveyors estimates for that medium, and how
estimates for reporting years  1994  and 1995 compared to estimates from reporting years 1987
and 1988.
                                          5-1
-------
When comparing the release and other waste management activity estimates of a facility to the
estimate of a site surveyor, the percent difference between the two estimates is used. The percent
difference between the facility estimate and the site surveyor estimate is calculated as follows:

                            Percent Difference = (Fa - SS)/SS x 100
                            where: Fa = Facility Estimate
                                         SS = Site Surveyor Estimate
5.1.1          Overview of Oil-Site Releases and Transfers Off-Site for Waste Management
              as Reported by Facilities and by Site Surveyors
              On-site releases and transfers off-site for waste management quantities as reported
by the facilities and the site surveyors were summed for all chemicals to get total facility
estimates. Total facility estimates were scaled and summed for all facilities to get total releases
and other waste management quantities for each SIC Code. The total quantity for each SIC Code
are presented by medium in Tables 5-1, 5-3, 5-5, 5-7, 5-9, and 5-11. Facility estimates were
lower than site surveyor estimates in all SIC Codes for fugitive air releases, transfers to off-site
recycling, and transfers to off-site energy recovery. In general, facility estimates were higher
than site surveyor estimates in each SIC Code for stack air releases.  For all SIC Codes surveyed
for reporting year (RY) 1994, total quantities for the SIC  Code as estimated by the facility are
lower than total quantities for the SIC Code as estimated by the site surveyor. Total on-site
releases and transfers off-site for waste management for SIC Codes 26 and 286,  surveyed for RY
1995, as estimated by the facility are within 2% of the total quantities estimated  by the site
surveyor.
                                           5-2
-------
              Table 5-1 is a summary of SIC Code 25 TRI on-site releases and transfers off-site
for waste management quantities for the RY 1994.  The greatest percent difference in estimates
by facility and by site  surveyor are for off-site transfers to treatment, where facility estimates
were 250% greater than site surveyor estimates. The overall impact of this difference in off-site
treatment is not significant, as transfer to off-site treatment makes up only a small portion of the
total quantity. None of the facilities surveyed in SIC Code 25 had releases to receiving streams,
performed underground injection, or had releases to land on site. Total on-site releases and
transfers off-site for waste management estimated by facilities and site surveyors were in close
agreement.
                                       Table 5-1
  Summary of SIC Code 25 TRI On-Site Releases and Transfers Off-Site for
         Waste Management for Reporting Year 1994 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Disposal
Off-Site Treatment
Off-Site Recycling
Off-Site Energy Recovery
Total
Quantity of Chemicals as
Reported by the Facilities
3.72
37.9
0.00
0.00
0.00
0.647
0.00
1.13
0.733
9.12
53.3
Quantity of Chemicals as
Reported by the Site
Surveyors
4.54
36.9
0.00
0.00
0.00
0.783
0.437
0.319
1.20
10.6
54.8
Percent Difference*
-18%
2.8%
NA
NA
NA
-17%
-100%
250%
-39%
-14%
-2.8%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no releases to this medium at the facilities surveyed in this SIC Code.
                                          5-3
-------
              Table 5-2 presents two forms of the 1994 on-site releases and transfers off-site for
waste management for SIC Code 25. In the second column, on-site releases and transfers off-site
for waste management as reported by the facilities surveyed were scaled-up to represent the total
releases and transfers off-site for waste management for SIC Code 25 for those facilities with less
than 16 Form Rs. Thus, each facility surveyed represents a group of facilities in the TRIS
database to determine the scaled-up total. The third column is the total on-site releases and
transfers off-site for waste management amount for SIC Code 25 as reported by SIC Code 25
facilities with less than 16 Form Rs taken from the TRIS database.  This comparison examines
how closely the surveyed facilities match the overall SIC Code 25 release profile. (Site
surveyors estimates are not presented on this table).  As discussed in Section 2, facility site
selection excluded facilities that reported more than 15 chemicals.  Most facilities that
manufacture, process, or otherwise use more than 15 chemicals would have larger quantities than
the average facility. The percent difference in total on-site releases and transfers off-site for
waste management quantities between the scaled-up estimate and the TRI database totals is -
17%.
                                       Table 5-2
     1994 Reported TRI On-Site Releases and Transfers Off-Site for Waste
                  Management for SIC Code 25 (millions of Ibs.)
Medium
Fugitive and Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Scaled Quantity of
Chemicals as Reported
by the Facilities Surveyed
41.6
0.00
0.00
0.00
0.647
11
53.3
Quantity of Chemicals as
Reported by All Facilities
in the TRIS Database
51.5
0.000266
0.00
0.0637
0.145
12.6
64.3
Percent Difference*
-19%
-100%
0.0%
-100%
346%
-13%
-17%
 *Percent Difference = (Sea - TRI)/TRI x 100, where Sea = Scaled Facility Estimate Total and TRI = Facility
 Estimate Total as Reported to TRI.
                                           5-4
-------
              Table 5-3 is a summary of SIC Code 281 TRI on-site releases and transfers off-
site for waste management for RY 1994.  In SIC Code 281, the most significant difference
between facility estimates and site surveyor estimates is in underground injection.  This
difference is attributed to errors made by two facilities surveyed. These two facilities perform
manufacturing process operations that are not typical in chemical manufacturing facilities. If
these two facilities are not considered in the sum of on-site releases and transfers off-site for
waste management, the total percent difference for total on-site releases and transfers off-site for
waste management quantities in SIC Code 281 releases drops to -1.1 percent.
                                       Table 5-3
 Summary of SIC Code 281 TRI On-Site Releases and Transfers Off-Site for
                Waste Management for RY 1994 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Disposal
Off-Site Treatment
Off-Site Recycling
Off-Site Energy Recovery
Total
Quantity of Chemicals as
Reported by the Facilities
15.4
126
1.73
15.0
0.00440
0.0644
29.9
9.97
1.18
0.304
200
Quantity of Chemicals as
Reported by the Site
Surveyors
16.1
126
1.65
38.0
0.00440
0.0641
30.0
10.1
1.48
2.00
225
Percent Difference*
-4.3%
0.53%
5.1%
-61%
0.0%
0.47%
-0.17%
-1.5%
-21%
-85%
-11%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no releases to this medium at the facilities surveyed in this SIC Code.
                                          5-5
-------
             Table 5-4 presents two forms of the 1994 on-site releases and transfers off-site for
waste management for SIC Code 281.  In the second column, on-site releases and transfers off-
site for waste management as reported by the facilities surveyed were scaled-up to represent the
total on-site releases and transfers off-site for waste management quantities for SIC Code 281 for
those facilities with less than 16 Form  Rs. The third column is the total on-site releases and
transfers off-site for waste management amount for SIC Code 281 as reported by all SIC Code
281 facilities with less than 16 Form Rs taken from the TRIS database. This comparison
examines how closely the surveyed facilities match the overall SIC Code 281 release profile.
(Site surveyors estimates are not presented on this table).  The percent difference in total on-site
releases and transfers off-site for waste management quantities between the scaled-up estimate
and the TRI database totals is -51%.
                                      Table 5-4
     1994 Reported TRI On-Site Releases and Transfers Off-Site for Waste
                 Management for SIC Code 281 (millions of Ibs.)
Medium
Fugitive and Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Scaled Quantity of
Chemicals as Reported
fcy the Facilities Surveyed
141
1.73
15.0
0.004
0.064
41.1
200
Quantity of Chemicals AS
Reported by All Facilities
ill the TRIS Database
90.8
25.4
153
68.2
31.4
41.1
410
Percent Difference*
36%
-94%
-90%
-100%
-100%
0.7%
-51%
*Percent Difference = (Sea - TRI)/TRI x 100, where Sea = Scaled Facility Estimate Total and TRI = Facility
Estimate Total as Reported to TRI.
                                          5-6
-------
              Table 5-5 is a summary of SIC Code 285 TRI on-site releases and transfers off-
 site for waste management for the reporting year 1994.  The most significant difference in on-site
 releases and transfers off-site for waste management activity estimates by facility and by site
 surveyor are for transfers to off-site recycling.  None of the facilities surveyed in SIC Code 285
 had releases to receiving streams, underground injection, or to land on site.  One of the surveyed
 facilities in this SIC Code  put release values under the wrong release type and grossly
 underestimated all on-site  releases and transfers off-site for waste management.  If this facility is
 not considered in the total  sum, the total percent difference between facility estimates and site
 surveyor estimates in SIC  Code 285 drops to -20%.
                                       Table 5-5
  Summary of SIC Code 285 TRI On-Site Releases and Transfer Off-Site for
                Waste Management for RY1994 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Disposal
Off-Site Treatment
Off-Site Recycling
Off-Site Energy Recovery
Total
Quantity of Chemicals as
Reported by the Facilities
2.23
0.654
0.00
0,00
0.00
0.0615
0.0874
0.101
2.75
2.67
8.55
Quantity of Chemicals as
Reported by the Site
Surveyors
3.15
0.533
0.00
0.00
0.00
0.0157
0.219
0.462
5.08
2.87
12.3
Percent Difference*
-29%
23%
NA
NA
NA
290%
-60%
-78%
-46%
-7.0%
-31%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no releases to this medium at the facilities surveyed in this SIC Code.
                                          5-7
-------
             Table 5-6 presents two forms of the 1994 on-site releases and transfers off-site for
waste management for SIC Code 285. In the second column, on-site releases and transfers off-
site for waste management quantities as reported by the facilities surveyed were scaled-up to
represent the total on-site releases and transfers off-site for waste management quantities for SIC
Code 285 for those facilities with less than 16 Form Rs.  The third column is the total on-site
releases and transfers off-site for waste management amount for SIC Code 285 as reported by all
SIC Code 285 facilities with less than 16 Form Rs taken from the TRIS database. This
comparison examines how closely the surveyed facilities match the overall SIC Code 285  release
profile.  (Site surveyors estimates are not presented on this table). The percent difference in total
on-site releases and transfers off-site for waste management releases and other waste
management quantities between the scaled-up estimate and the TRI database totals is -92%.
                                      Table 5-6
     1994 Reported TRI On-Site Releases and Transfers Off-Site for Waste
                 Management for SIC Code 285 (millions of Ibs.)
Medium
Fugitive and Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Scaled Quantity of
Chemicals as Reported
by the Facilities Surveyed
2.88
0.00
0.00
0.00
0.0615
5.61
8.55
Quantity of Chemicals as
Reported by All Facilities
in the TRIS Database
13.5
0.00
0.00
0.055
1.63
95.5
111
Percent Difference*
-79%
0.0%
0.0%
-100%
-96%
-94%
-92%
 *Percent Difference = (Sea - TRI)/TRI x 100, where Sea = Scaled Facility Estimate Total and TRI = Facility
 Estimate Total as Reported to TRI.
                                           5-8
-------
              Table 5-7 is a summary of SIC Code 30 TRI on-site releases and transfers off-site
for waste management quantities for the reporting year 1994. Many facilities in this SIC Code
reported fugitive emissions as stack emissions.  Even so, the sum of the fugitive and stack
emissions estimated by the facilities and site surveyors was in close agreement.  The greatest
percent difference in estimates by facility and by site surveyor are for discharges to POTWs,
where facility estimates were 100% less than site surveyor estimates. However, discharges to
POTWs account for much less than 0.1  percent of all quantities. None of the facilities surveyed
in SIC Code 30 had on-site releases to receiving streams, underground injection, or to land on
site. Total on-site releases and transfers off-site for waste management quantities estimated by
facilities and site surveyors were in close agreement.
                                      Table 5-7
  Summary of SIC Code 30 TRI On-Site Releases and Transfers Off-Site for
                Waste Management for RY1994 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Disposal
Off-Site Treatment
Off-Site Recycling
Off-Site Energy Recovery
Total
Quantity of Chemicals as
Reported by the Facilities
26.0
182
0.00
0.00
0.00
0.00
7.13
21.6
10.9
0.332
248
Quantity of Chemicals as
Reported by the Site
Surveyors
69.3
138
0.00
0.00
0.00
0.00145
5.84
23.5
17.2
0.347
254
Percent Difference*
-63%
31%
NA
NA
NA
-100%
22%
-8.1%
-37%
-4.3%
-2.4%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable.  There were no releases to this medium at the facilities surveyed in this SIC Code.
                                          5-9
-------
             Table 5-8 presents two forms of the 1994 on-site releases and transfers off-site for
waste management quantities for SIC Code 30. In the second column, on-site releases and
transfers off-site for waste management quantities as reported by the facilities surveyed were
scaled-up to represent the total on-site releases and transfers off-site for waste management
quantities for SIC Code 30 for those facilities with less than 16 Form Rs. The third column is the
total on-site releases and transfers off-site for waste management amount for SIC Code 30 as
reported by all SIC Code 30 facilities with less than 16 Form Rs taken from the TRIS database.
This comparison examines how closely the surveyed facilities match the overall SIC 30 release
profile.  The percent difference in total on-site releases and transfers off-site for waste
management quantities between the scaled-up estimate and the TRI database totals is 7.4%.
                                      Table 5-8
     1994 Reported TRI On-Site Releases and Transfers Off-Site for Waste
                  Management for SIC Code 30 (millions of Ibs.)
Medium
Fugitive and Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Scaled Quantity of
Chemicals as Reported
by the Facilities Surveyed
208
0.00
0.00
0.00
0.00
40.0
248
Quantity of Chemicals as
Reported fay All Facilities
in the TRIS Database
154
0.230
0.00
0.357
2.22
73.7
231
Percent difference*
35%
-100%
0.0%
-100%
-100%
-46%
7.4%
*Percent Difference = (Sea - TRI)/TRI x 100, where Sea = Scaled Facility Estimate Total and TRI = Facility
Estimate Total as Reported to TRI.
                                          5-10
-------
             Table 5-9 is a summary of SIC Code 26 TRI on-site releases and transfers off-site
for waste management for RY 1995. None of the facilities surveyed in SIC Code 26 had releases
to underground injection, off-site treatment, off-site recycling, or off-site energy recovery.
Facility and site surveyor estimates are in close agreement.  The close agreement between the
facility and surveyor estimates in SIC Code 26 can be attributed to the step-by-step procedures
listed in the NCASI Handbook of Chemical Specific Information for SARA 313 Form R
Reporting which most paper and paperboard facilities use as guidance for filling out Form Rs.
This manual  is distributed by NCASI, and has not been through EPA approval. However, it is
still a good source for documentation and calculations needed to complete the Form Rs.
                                      Table 5-9
  Summary of SIC Code 26 TRI On-Site Releases and Transfers Off-Site for
                Waste Management for RY 1995 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Disposal
Off-Site Treatment
Off-Site Recycling
Off-Site Energy Recovery
Total
Quantity of Chemicals as
Reported by the Facilities
12.6
69.9
2.03
0
0.189
0.00421
0.761
0
0
0
85
Quantity of Chemicals as
Reported by the Site
Surveyors
12.9
69.1
1.85
0
0.119
0.00421
0.767
0
0
0
85
Percent Difference*
-2.5%
1.2%
9.6%
NA
59%
0.0%
-0.74%
NA
NA
NA
0.8%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no releases to this medium at the facilities surveyed in this SIC Code.
                                         5-11
-------
             Table 5-10 presents two forms of the 1995 on-site releases and transfers off-site
for waste management for SIC Code 26.  In the second column, on-site releases and transfers off-
site for waste management as reported by the facilities surveyed were scaled-up to represent the
total on-site releases and transfers off-site for waste management for SIC Code 26 for those
facilities with less than 16 Form Rs. The third column is the total on-site releases and transfers
off-site for waste management amount for SIC Code 26 as reported by SIC Code 26 facilities
with less than 16 Form Rs taken from the TRIS database. This comparison examines how
closely the surveyed facilities match the overall SIC Code 26 release profile.  (Site surveyors
estimates are not presented on this table). The percent difference in total on-site releases and
transfers off-site for waste management between the scaled-up estimate and the TRI database
totals is -66%.
                                     Table 5-10
    1995 Reported TRI On-Site Releases and Transfers Off-Site for Waste
                  Management for SIC Code 26 (millions of Ibs.)
Medium
Fugitive and Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Scaled Quantity of
Chemicals as Reported
by the Facilities Surveyed
82.5
2.03
0.00
0.189
0.00421
0.761
85
Quantity of Chemicals as
Reported by All Facilities
in the TRIS Database
166
7.01
0.00
3.39
41.0
29.9
247
Percent Difference*
-50%
-71%
0.0%
-94%
-100%
-97%
-66%
*Percent Difference = (Sea - TRI)/TRI x 100, where Sea = Scaled Facility Estimate Total and TRI = Facility
Estimate Total as Reported to TRI.
                                          5-12
-------
              Table 5-11 is a summary of SIC Code 286 TRI on-site releases and transfers off-
site for waste management quantities for the reporting year 1995.  None of the facilities
surveyed in SIC Code 286 had on-site releases to underground injection. The close agreement
between the facility and surveyor estimates SIC Code 286 can be attributed to the relatively large
environmental staff and explicit corporate policies followed by the large organic chemical
companies visited.
                                     Table 5-11
  Summary of SIC Code 286 TRI On-Site Releases and Transfers Off-Site for
                Waste Management for RY1995 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Disposal
Off-Site Treatment
Off-Site Recycling
Off-Site Energy Recovery
Total
Quantity of Chemicals as
Reported by the Facilities
7.80
18.0
0.0958
0
0.000136
125
0.0910
36.4
5.11
127
320
Quantity of Chemicals as
Reported by the Site
Surveyors
9.15
16.7
0.123
0
0.000180
128
0.100
36.4
5.11
129
325
Percent Difference*
-15%
7.6%
-22%
NA
-24%
-2.2%
-9.3%
-0.04%
0.0%
-1.4%
-1.5%
'Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no releases to this medium at the facilities surveyed in this SIC Code.
                                         5-13
-------
             Table 5-12 presents two forms of the 1995 on-site releases and transfers off-site
for waste management for SIC Code 286. In the second column, on-site releases and transfers
off-site for waste management as reported by the facilities surveyed were scaled-up to represent
the total on-site releases and transfers off-site for waste management for SIC Code 286 for those
facilities with less than 16 Form Rs. The third column is the total on-site releases and transfers
off-site for waste management amount for SIC Code 286 as reported by SIC Code 286 facilities
with less than 16 Form Rs taken from the TRIS database. This comparison examines how
closely the surveyed facilities match the overall SIC Code 286 release profile. (Site surveyors
estimates are not presented on this table). The percent difference in total on-site releases and
transfers off-site for waste management between the scaled-up estimate and the TRI database
totals is 8.0%.
                                     Table 5-12
    1995 Reported TRI On-Site Releases and Transfers Off-Site for Waste
                 Management for SIC Code 286 (millions of Ibs.)
Medium
Fugitive and Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Scaled Quantity of
Chemicals as Reported
by the Facilities Surveyed
25.8
0.0958
0.00
0.00
125
169
320
Quantity of Chemicals as
Reported by All Facilities
in the TRIS Database
77.9
4.07
33.1
0.926
68.8
163
348
Percent Difference*
-67%
-98%
-100%
-100%
82%
3.7%
-8.0%
*Percent Difference = (Sea - TRI)/TRI x 100, where Sea = Scaled Facility Estimate Total and TRI = Facility
Estimate Total as Reported to TRI.
                                          5-14
-------
              Figure 5-1 presents estimates of total on-site releases and transfers off-site for
waste management calculated by facilities and site surveyors for each SIC Code surveyed for RY
1994 and RY 1995. The total on-site releases and transfers off-site for waste management were
in good agreement, calculated to be within +3% for most SIC Codes.  Estimates of total on-site
releases and transfers off-site for waste management calculated by facilities and site surveyors
for all SIC Codes surveyed in RY 1994 and RY 1995 differed by 4%.

              Figures 5-2, 5-3, and 5-4 present the relative percent differences in estimates of
on-site releases and transfers off-site for waste management between facilities and site surveyors
for each SIC Code. In all SIC Codes, fugitive emissions tend to be incorrectly reported as stack
emissions, leading to overestimates of stack emissions and underestimates of fugitive emissions.
Another trend in SIC Codes 25, 281, 285, and 30 for RY 1994 is the misreporting of chemical
transfers off-site for purposes of disposal, treatment, recycling,  or energy recovery.  Many
facilities do not record the actual fate of chemicals transferred off-site when filling out the Form
Rs. Most facilities check the off-site disposal or off-site treatment boxes without considering the
possibility of recycling or energy recovery. Facilities in these same SIC Codes tend to have
container residue that was overlooked. The container residue is usually treated, recycled, or
disposed of by the vendor collecting the drums, and not incorporated into the product as reported
by the facilities.

5.1.2         Comparison  of RY 1994 and RY 1995 On-Site Releases and Transfers Off-
              Site for Waste Management to RY 1987 and RY 1988 On-Site Releases and
              Transfers Off-Site for Waste Management

              Tables 5-13 through 5-16 contain the TRI on-site releases and transfers off-site for
waste management for the surveys conducted for RY 1995, 1994, 1988, and 1987 data,
respectively. Different SIC Codes were surveyed in each reporting year of the site survey
program, so caution should be exercised when comparing data from one reporting year to the
next. The tables present a comparison between the quantity of chemicals released on-site or
transferred off-site for waste  management as reported by the facilities and the quantity of
chemicals released on-site or transferred off-site for waste management as reported by the site
surveyor.  The percent difference between the estimates are also provided. The percent
differences for each reporting year are summarized on Figures 5-5 and 5-6.

                                          5-15
-------
                Figure 5-1. Comparison of Facility and Site Surveyor Estimates of Total On-Site Releases
                                       and Transfers Off-Site for Waste Management
       350
                                                                                                                 Facility
                                                                                                                 Site Surveyor
              SIC Code 25         SIC Code 281        SIC Code 285        SIC Code 30
               (RY 1994)          (RY 1994)          (RY 1994)          (RY 1994)

                                                     Industry Sector
SIC Code 26
 (RY 1995)
SIC Code 286
 (RY 1995)
Data for this figure can be found on Tables 5-1 through 5-12.
-------
                                    Ll-S
I
3
5'
O-
O
Percent Difference *
I
V
                                            Stream (NA)

                                       Underground Inje ction (NA)
                                          (Q
                                           c
                                        8  £
                                        ff !°
                                        2 O

                                        II

                                        11.
                                        3  to
                                        (D  O
                                        2- =
                                        5- a
                                                                          O
                                                                          8 S
                                                                          10 3,
                                                                          01 _|
                                                                          fi> O
                                                                          as
                                                                          to JiJ

                                                                          •5

                                                                         «< 2.
                                                                          (D W
                                                                          o. 5"
                                                                         II
                                                                          O H

                                                                          II
                                                                         (Q (A
                                                                            
-------
                         Figure 5-3.  Comparison of Estimates of Total On-Site Releases and Transfers Off-Site for
                              Waste Management in SIC Codes 285 and 30 Surveyed for Reporting Year 1994
            290.0%5
00
                                                290.0%
              80.0%	
              -80.0%
             -100.0%
                                Fugitive Air
                              Q Stack Air
                                Receiving Stream
                                Underground Injection
                                Land On-Site
                                PO7W
                                Off-Site Disposal
                                Off-Site Treatment
                                Off-Site Recycling
                                Off-Site Energy Recovery
                                      SIC Code 285
SIC Code 30
        * Percent Difference = (Fa-SS)/SS x 100, w here Fa = Facility Estimate and SS = Site Surveyor Estimate.
        NA - Not applicable. There w ere no releases to this medium at the facilities surveyed in this SIC Code.
        Data for this figure can be found on Tables Tables 5-5 and 5-6.
-------
               Figure 5-4. Comparison of Estimates of Total On-Site Releases and Transfers Off-Site for
                    Waste Management in SIC Codes 26 and 286 Surveyed for Reporting Year 1995
     120.0%
     100.0%
     -80.0%
    -100.0%
                           SIC Code 25
SIC Code 281
                                                                                                           H Fugitive Air
                                                                                                           n Stack Air
                                                                                                           gg Receiving Stream
                                                                                                           B Underground Injection
                                                                                                           m Land On-Site
                                                                                                           QPOTW
                                                                                                           B Off-Site Disposal
                                                                                                           H Off-Site Treatment
                                                                                                           B Off-Site Recycling
                                                                                                           B Off-Site Energy Recovery
* Percent Dfference = (Fa-SS)/SS x 100, w here Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There w ere no releases to this medium at the facilities surveyed in this SIC Code.
Data for this figure can be found on Tables 5-9 and 5-11.
-------
                                 Table 5-13
Summary of RY 1995 TRI On-Site Releases and Transfers Off-Site for Waste
           Management for SIC Codes 26 and 286 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Quantity of Chemicals as
Reported by the Facilities
20
88
2.1
NA
0.19
125
169
405
Quantity of Chemicals as
Reported by the Site
Surveyors
22
86
2.0
NA
0.12
128
171
410
Percent Bifference*
-7.6%
2.4%
7.7%
NA
59%
-2.2%
-1.2%
-1.2%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
                                 Table 5-14

Summary of RY 1994 TRI On-Site Releases and Transfers Off-Site for Waste
      Management for SIC Codes 25,281,285, and 30 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Quantity of Chemicals as
Reported by the Facilities
47
350
1.7
15
0.0044
0.77
98
510
Quantity of Chemicals as
Reported by the Site
Surveyors
93
300
1.7
38
0.0044
0.86
110
550
Percent Difference*
-49%
15%
5.1%
-61%
0.0%
-11%
-13%
-6.7%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
                                    5-20
-------
                                Table 5-15
Summary of RY 1988 TRI On-Site Releases and Transfers Off-Site for Waste
  Management for SIC Codes 28,291, and 34 Through 38 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Quantity of Chemicals as
Reported by the Facilities
470
850
30
0.00
60
550
530
2,490
Quantity of Chemicals as
Reported by the Site
Surveyors
480
740
3
0.00
70
750
420
2,463
Percent Difference*
-2.1%
15%
900%
NA
-14%
-27%
26%
1.1%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
                                 Table 5-16

Summary of RY 1987 TRI On-Site Releases and Transfers Off-Site for Waste
         Management for SIC Codes 20 Through 39 (millions of Ibs.)
Medium
Fugitive Air
Stack Air
Receiving Stream
Underground Injection
Land On Site
POTW
Off-Site Transfers
Total
Quantity of Chemicals as
Reported by the Facilities
800
1,800
9,600
3,200
2,400
2,200
2,600
22,500
Quantity of Chemicals as
Reported by the Site
Surveyors
800
1,900
9,900
3,200
2,700
2,000
2,700
23,000
Percent Difference*
0.0%
-5.3%
-3.0%
0.0%
-11%
10%
3.7%
-2.2%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
                                    5-21
-------
                                                 zz-s
o
0)
5T
-K   -7 "
?   if »
                                             Percent Difference'
I
O
§
Cr

s-
CD
en

s
co

Q.

£
•tv
o>  o
   CD

     O  X
     33
     ^ p
     V)  <
     il
     o  S
     3 >?
     - B"
     II
     3* sr
     O CD
CD  C/)
•<  7=f
CD  CD
     II
     CD S.
     '  C?
                                                                                                  (Q
                                                                                                        Ol
                                                                                                   O
                                                                                                   o

                                                                                                   •o
                                                                                                   fi>
                                                                                                   55'
                                                                                                   o
                                                                                                   =
                                                                                                    73
                                                                                                    
                                                                                                    (Q  CO

                                                                                                    (O  Q)
                                                                                                    01  3
                                                                                                    Q)  $2.
                                                                                                    to
                                                                                                    to
                                                                                                        CO
                                                                                                        I
                                                                                                        (A
                                                                                                        ff
                                                                                                        3
                                                                                                        Q)
                                                                                                       (Q
                                                                                                        
-------
to
LO
                        Figure 5-6. Comparison of On-Site Releases and Transfers Off-Site for Waste Management
                                                       from Reporting Years 1988 and 1987
             900.0%
              100.0%	
               -40.0%
                                1988 (SIC Codes 28, 291, 34-38)
1987 (SC Codes 25, 30, 281, 285)
         * Percent Difference = (Fa-SS)/SS x 100, w here Fa = Facility Estimate and SS = Site Surveyor Estimate.
         NA - Not applicable. There w ere no releases to this medium at the facilities surveyed in this SIC Code.
                                      B Fugitive Air
                                      rj Stack Air
                                      m Receiving Stream
                                      g Underground Injection
                                      H Land On-Site
                                      • POTW
                                      • Off-Site Transfers
         Data for this figure can be found on Tables 5-11 and 5-12.
-------
             The percent differences in facility and site surveyor total estimates for RY 1987,
RY 1988, RY 1994, and RY 1995 are presented in Table 5-17. The percent differences for each
reporting year were less than 7 percent.

                                     Table 5-17
           Percent Difference of Facility Estimated and Site Surveyor
    Estimated Total TRI On-Site Releases and Transfers Off-Site for Waste
 Management for RY 1995, RY 1994, RY 1988, and RY 1987 (millions of Ibs.)
TRI Reporting Year
1995
1994
1988
1987
Percent Difference*
-1.2%
-6.7%
1.1%
-2.2%
5.1.3
              *Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site
              Surveyor Estimate.
Analysis of Specific Releases
              Analyses of specific on-site releases and transfers off-site for waste management
are presented in this section.  These analyses only apply to facilities that correctly reported
chemical on-site releases and transfers off-site for waste management. Facilities that incorrectly
reported an on-site release or transfer off-site for waste management, incorrectly did not report an
on-site release or transfer off-site for waste management, or correctly did not report an on-site
release or transfer off-site for waste management are not included in the analyses in section 5.1.3.
On-site releases and transfers off-site for waste management in this section are analyzed on a
total facility basis. For example, if a facility underestimated the release of a chemical by 1,000,
but overestimated the release of another chemical by 1,000 Ibs, the errors would cancel and
would not be identified in this analysis.
                                          5-24
-------
5.1.3.1
Fugitive Air Releases
              A comparison of the percent difference between facility estimates and site
surveyor estimates for fugitive air releases is presented in Table 5-18. It is notable that in SIC
Code 30, over half the facilities estimates differed by more than 50 percent from the site surveyor
estimate.
                                       Table 5-18
     Comparison of the Percent Difference (PD)1 Between Facility and Site
                  Surveyor Estimates for Fugitive Air Emissions
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
21%
0%
7%
17%
43%
0%
% Facilities
Where
PD<1%
39%
13%
7%
17%
43%
0%
% Facilities
Where
PD<10%
47%
38%
12%
28%
43%
50%
% Facilities
Where
PD<50%
79%
67%
69%
40%
57%
83%
% Facilities
Where
PD>50%
21%
33%
31%
60%
43%
17%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Release Estimate, and SS = Site Surveyor Release Estimate.
Percentages are based on survey weighted data.
              In Figure 5-7, the facility fugitive air emissions estimates are again compared to
the site surveyor estimates, but now the difference in facility and site surveyor estimates are
classified according to whether the facility and the site surveyor estimates were equal, whether
the facility estimates were less than the site surveyor estimates, or whether the facility estimates
were greater than the site surveyor estimates.
                                          5-25
-------
                  Figure 5-7. Comparison of Facility Estimates and Site Surveyor Estimates
                                         for Fugitive Air Emissions
  50%
                                        24%
                            50%
                                   17%
                                                                69%
       SIC Code 281 (RY
SIC Code 285 (RY 1994)
                                                                                                      83%
  SIC Code 286 (RY 1995)
                          21%
                                                            17%
                                     34%
                             28%
51%
                                                               49%
        SIC Code 25 (RY 1994)
SIC Code 30 (RY 1994)
                                                                            29%
   29%

SIC Code 26 (RY 1995)
     I Facility Estimate Equal to Site Surveyor Estimate
     I Facility Estimate Greater Than Site Surveyor Estimate
             Facility Estimate Less Than Site Surveyor Estimate
-------
 5.1.3.2
Stack Air Releases
              A comparison of the percent difference between facility estimates and site
surveyor estimates for stack air releases is presented in Table 5-19. It is notable that in SIC Code
285, over half the facilities' estimated releases differed by more than 50 percent from the site
surveyor estimate.

                                      Table 5-19
     Comparison of the Percent Difference (PD)1 Between Facility and Site
                    Surveyor Estimates for Stack Air Emissions
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
10%
0%
0%
31%
38%
0%
% Facilities
Where
PD<1%
25%
22%
28%
35%
50%
14%
% Facilities
Where
PD<10%
55%
78%
28%
58%
75%
14%
% Facilities
Where
PD<50%
97%
92%
37%
75%
75%
57%
% Facilities
Where
PD>50%
3%
8%
63%
25%
25%
43%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Release Estimate, and SS = Site Surveyor Release Estimate.

              In Figure 5-8, the facility stack air emissions estimates are again compared to the
site surveyor estimates, but now the difference in facility and site surveyor estimates are
classified according to whether the facility and the site surveyor estimates were equal, whether
the facility estimates were less than the site surveyor estimates, or whether the facility estimates
were greater than the site surveyor estimates.

              In SIC Codes 25,281,285, and 286, facility estimates were higher than site
surveyors estimates more often than they were equal to or lower than site surveyor estimates.  In
SIC Codes 26 and 30, facility estimates were equal to, greater than, and less than site surveyor
estimates in approximately equal proportions. Site surveyors often encountered facilities that did
                                          5-27
-------
                             Figure 5-8. Comparison of Facility Estimates and Site Surveyor Estimates

                                                      for Stack Air Emissions
                                                                  3%
                                       29%
             71%
                    SIC Code 281 (RY
                              97%



                          SIC Code 285 (RY 1994)
                                                                                     57%
                                                                                                                        43%
SIC Code 286 (RY 1995)
to
oo
10%
            53%
                                        37%
                     SIC Code 25 (RY 1994)
                                                                             31%
                                                 32%
                                    37%



                          SIC Code 30 (RY 1994)
                                                                                                                24%
                                                                                         38%
                                                                                 38%
SIC Code 26 (RY 1995)
                 I Facility Estimate Equal to Site Surveyor Estimate

                 I Facility Estimate Greater Than Site Surveyor Estimate
                                      I Facility Estimate Less Than Site Surveyor Estimate
-------
not understand the definition of stack air releases and misclassified fugitive emissions as stack
emissions. This appears to be the primary reason most of the SIC Codes surveyed showed an
inclination to overestimate stack releases.
5.1.3.3
Discharges to Receiving Streams
              A comparison of the percent difference between facility estimates and site
surveyor estimates for receiving stream discharges is presented in Table 5-20.  SIC Codes 25,
285, and 30 did not have any surveyed facilities with discharges to receiving streams.  SIC Code
285 had two facilities which reported discharges to receiving streams. One facility overestimated
emissions by 90 percent, and the other facility underestimated emissions by 55 percent.
                                      Table 5-20
     Comparison of the Percent Difference (PD)1 Between Facility and Site
             Surveyor Estimates for Discharges to Receiving Streams
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
NA
42%
NA
NA
43%
0%
%Fa««itie$
Where
HK1%
NA
73%
NA
NA
43%
0%
% Facilities
Where
H>50%
NA
13%
NA
NA
14%
100%
'PD = The absolute value of the percent Difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Release Estimate, and SS = Site Surveyor Release Estimate.
NA - Not applicable. There were no releases to receiving streams at the facilities surveyed for this SIC Code.
              In Figure 5-9, the facility receiving stream release estimates are again compared to
the site surveyor estimates, but now the difference in facility and site surveyor estimates are
classified according to whether the facility and the site surveyor estimates were equal, whether
                                          5-29
-------
                               Figure 5-9. Comparison of Facility Estimates and Site Surveyor Estimates
                                                        for Receiving Streams
                                        42%
                                                                             50%
            58%
                                                     50%
                     SIC Code 281 (RY 1994)
                                          SIC Code 286 (RY 1995)
OJ
O
43%
                                              43%
                          14%
                     SIC Code 26 (RY 1995)
I Facility Estimate Equal to Site
 Surveyor Estimate

I Facility Estimate Less Than Site
 Surveyor Estimate

I Facility Estimate Greater Than Site
 Surveyor Estimate
-------
the facility estimates were less than the site surveyor estimates, or whether the facility estimates
were greater than the site surveyor estimates.
5.1.3.4
Underground Injection
              A comparison of the percent difference between facility estimates and site
surveyor estimates for underground injection quantities is presented in Table 5-21.
Only two facilities surveyed in SIC Code 281 had releases by underground injection. The
facility estimates for these quantities is between 10% and 50% less than the site surveyor
estimates. No facilities surveyed in SIC Codes 25,26, 30, 285, or 286 had underground injection
releases.
                                      Table 5-21
     Comparison of the Percent Difference (PD)1 Between Facility and Site
                  Surveyor Estimates for Underground Injection
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD-0
NA
0%
NA
NA
NA
NA
% Facilities
Where
PD<1%
NA
0%
NA
NA
NA
NA
% Facilities
Where
PB50%
NA
0%
NA
NA
NA
NA
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
NA - Not applicable. There were no underground injection amounts at the facilities surveyed for this SIC Code.
5.1.3.5
Releases and Other Waste Management Quantities to Land On Site
              A comparison of the percent difference between facility estimates and site
surveyor estimates for releases to land on site is presented in Table 5-22. Releases to land on-site
occurred at only one site surveyed in SIC Codes 281 and 286, and at three sites surveyed in SIC
                                          5-31
-------
Code 26. No sites surveyed in SIC Codes 25, 30, and 285 had releases to land on-site.  In Figure
5-10, the facility releases to land on-site are again compared to the site surveyor estimates, but
now the difference in facility and site surveyor estimates were equal, whether the facility
estimates were less than the site surveyor estimates, or whether the facility estimates were greater
than the site surveyor estimates.
                                      Table 5-22
     Comparison of the Percent Difference (PD)1 Between Facility and Site
                   Surveyor Release Estimates to Land On Site
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
NA
100%
NA
NA
33%
0%
% Facilities
Where
PD<1%
NA
100%
NA
NA
33%
0%
% Facilities
Where
PD<10%
NA
100%
NA
NA
33%
0%
% Facilities
Where
PB<50%
NA
100%
NA
NA
33%
100%
% Facilities
Where
PI>>50%
NA
0%
NA
NA
67%
0%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
NA - Not applicable. There were no releases to land on site at the facilities surveyed for this SIC Code.
5.1.3.6
Discharges to POTWs
              A comparison of the percent difference between facility estimates and site
surveyor estimates for discharges to POTWs is presented in Table 5-23.  The number of facilities
surveyed in SIC Codes 25, 26, 30, 281, 285, and 286 that reported discharges to a POTW are
two, zero, four, one, one, and seven, respectively.
                                          5-32
-------
                            Figure 5-10. Comparison of Facility Estimates and Site Surveyor Estimates
                                                         to Land On-Site
                          100%

                   SIC Code 281 (RY1994)
      100%

SIC Code 286 (RY 1995)
U)
U)
               67%
                                        33%
                   SIC Code 26 (RY 1995)
I Facility Estimate Equal to Site
 Surveyor Estimate

I Facility Estimate Less Than Site
 Surveyor Estimate

I Facility Estimate Greater Than Site
 Surveyor Estimate
-------
                                      Table 5-23
     Comparison of the Percent Difference (PD)1 Between Facility and Site
                   Surveyor Estimates for Discharges to POTWs
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
0%
0%
0%
NA
100%
17%
% Facilities
Where
PD<1%
73%
0%
0%
NA
100%
50%
% Facilities
Where
PB<10%
73%
27%
0%
NA
100%
67%
% Facilities
Where
PD<50%
73%
65%
0%
NA
100%
83%
% Facilities
Where
W>>50%
27%
35%
100%
NA
0%
17%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
NA - Not applicable. There were no discharges to POTWs at the facilities surveyed for this SIC Code.
              In Figure 5-11, the facility discharge to POTW estimates are again compared to
the site surveyor estimates, but now the difference in facility and site surveyor estimates are
classified according to whether the facility and the site surveyor estimates were equal, whether
the facility estimates were less than the site surveyor estimates, or whether the facility estimates
were greater than the site surveyor estimates.
5.1.3.7
Off-Site Transfers for Disposal
              A comparison of the percent difference between facility estimates and site
surveyor estimates for off-site transfers for disposal is presented in Table 5-24. SIC Code 285,
representing paint manufacturing facilities, had a high percentage of facilities which disagreed
with the site surveyors estimates because most facilities overlooked container residue and other
forms of off-site disposal.
                                           5-34
-------
               Figure 5-11. Comparison of Facility Estimates and Site Surveyor Estimates
                                      for Discharges to POTWs
35%
                       65%
     SIC Code 281 (RY
                        27%
                                     33%
     100%

SIC Code 285 (RY 1994)
                                                            17%
73%
     SIC Code 25 (RY 1994)
                                                            50%
SIC Code 286 (RY 1995)
Facility Estimate Equal to Site
Surveyor Estimate

Facility Estimate Less Than Site
Surveyor Estimate

Facility Estimate Greater Than Site
Surveyor Estimate
-------
                                       Table 5-24
    Comparison of the Percent Difference (PD)1 Between Facility Transfer
     Estimates and Site Surveyor Transfer Estimates for Off-Site Disposal
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
NA
15%
0%
0%
50%
50%
% Facilities
Where
PD<1%
NA
78%
0%
0%
100%
50%
% Facilities
Where
PD<10%
NA
78%
0%
0%
100%
100%
% FacUities
Where
PD«50%
NA
100%
4%
59%
100%
100%
% Facilities
Where
FD>50%
NA
0%
96%
41%
0%
0%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
NA - Not applicable. There were no off-site transfers for disposal at the facilities surveyed for this SIC Code.
              In Figure 5-12, the facility off-site transfers for disposal estimates are again
compared to the site surveyor estimates, but now the difference in facility and site surveyor
estimates are classified according to whether the facility and the site surveyor estimates were
equal, whether the facility estimates were less than the site surveyor estimates, or whether the
facility estimates were greater than the site surveyor estimates.
5.1.3.8
Off-Site Transfers for Treatment
              A comparison of the percent difference between facility estimates and site
surveyor estimates for off-site transfers for treatment is presented in Table 5-25.  SIC Code 285,
representing paint manufacturing facilities, had a total of two facilities with off-site transfers for
treatment.  These facilities overlooked many of these transfers. Container residue and bad
process batches made up the bulk of off-site treatment transfers. As shown in Table 5-21, these
two facility estimates were greater than 50% different than the site surveyor estimate.  SIC Code
25, representing furniture manufacturing facilities, had three facilities which reported off-site
transfers for treatment.  These transfers were mainly disposed or recycled by off-site vendors
                                           5-36
-------
                 Figure 5-12. Comparison of Facility Estimates and Site Surveyor Estimates
                                     for Transfers Off-Site for Disposal
                        15%
39%
                         46%
        SIC Code 281 (RY
   96%

SIC Code 285 (RY 1994)
                                                                                                       50%
                                                                             50%
SIC Code 286 (RY 1995)
                            41%
                                                                 50%
  59%
                                      50%
         SIC Code 30 (RY 1994)
SIC Code 26 (RY 1995)
      I Facility Estimate Equal to Site Surveyor Estimate         g Facility Estimate Less Than Site Surveyor Estimate
      I Facility Estimate Greater Than Site Surveyor Estimate
-------
instead of treated. Therefore, these facility estimates were greater than 50% different than the
site surveyors estimates.
                                      Table 5-25
    Comparison of the Percent Difference (PD)1 Between Facility Transfer
    Estimates and Site Surveyor Transfer Estimates for Off-Site Treatment
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD-0
0%
22%
0%
24%
NA
40%
% Facilities
Where
PD<1%
0%
38%
0%
27%
NA
80%
% Facilities
Where
PD<10%
0%
65%
0%
35%
NA
100%
% Facilities
Where
PD<50%
0%
65%
0%
73%
NA
100%
% Facilities
Where
PD>50%
100%
35%
100%
27%
NA
0%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
              In Figure 5-13, the facility off-site transfers for treatment estimates are again
compared to the site surveyor estimates, but now the difference in facility and site surveyor
estimates are classified according to whether the facility and the site surveyor estimates were
equal, whether the facility estimates were less than the site surveyor estimates, or whether the
facility estimates were greater than the site surveyor estimates.
5.1.3.9
Off-Site Transfers for Recycling
              A comparison of the percent difference between facility estimates and site
surveyor estimates for off-site transfers for recycling is presented in Table 5-26.
                                          5-38
-------
               Figure 5-13. Comparison of Facility Estimates and Site Surveyor Estimates
                                  for Transfers Off-Site for Treatment
56%
                        22%
                           2%
      SIC Code 281 (RY
                                                                              20%
                                                                                                     40%
      100%

SIC Code 285 (RY 1994)
                                                                               40%
SIC Code 286 (RY 1995)
                                                            24%
            100%

       SIC Code 25 (RY 1994)
                                     76%
SIC Code 30 (RY 1994)
   I Facility Estimate Equal to Site Surveyor Estimate
   I Facility Estimate Greater Than Site Surveyor Estimate
           I Facility Estimate Less Than Site Surveyor Estimate
-------
                                     Table 5-26
    Comparison of the Percent Difference (PD)1 Between Facility Transfer
    Estimates and Site Surveyor Transfer Estimates for Off-Site Recycling
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
3%
0%
0%
0%
NA
100%
% Facilities
Where
PD<1%
3%
0%
8%
0%
NA
100%
% Facilities
Where
PD<10%
50%
0%
24%
0%
NA
100%
% Facilities
Where
PD<50%
84%
100%
62%
100%
NA
100%
% Facilities
Where
PD>50%
16%
0%
38%
0%
NA
0%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
             In Figure 5-14, the facility off-site transfers for recycling estimates are again
compared to the site surveyor estimates, but now the difference in facility and site surveyor
estimates are classified according to whether the facility and the site surveyor estimates were
equal, whether the facility estimates were less than the site surveyor estimates, or whether the
facility estimates were greater than the site surveyor estimates. There is a general tendency
among SIC Codes 25, 281, and 285 for facility estimates to be less than site surveyor estimates.
Facilities in these SIC Codes tend to overlook solvent remaining in container residue that can be
recycled by some off-site vendors.
                                          5-40
-------
            Figure 5-14. Comparison of Facility Estimates and Site Surveyor Estimates
                                      for Off-Site Recycling
                                  24%
         100%

   SIC Code 281 (RY 1994)
                                                             76%
SIC Code 285 (RY 1994)
16%
                                    52%
                 81%
  SIC Code 25 {RY 1994)
                                                              48%
  SIC Code 30 (RY 1994)
I Facility Estimate Equal to Site
 Surveyor Estimate

I Facility Estimate Less Than Site
 Surveyor Estimate

I Facility Estimate Greater Than Site
 Surveyor Estimate
-------
5.1.3.10
Off-Site Transfers for Energy Recovery
             A comparison of the percent difference between facility estimates and site
surveyor estimates for off-site transfers for energy recovery is presented in Table 5-27.
                                      Table 5-27
              Comparison of the Percent Difference (PD)1 Between
                  Facility Transfer Estimates and Site Surveyor
                Transfer Estimates for Off-Site Energy Recovery
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
20%
0%
0%
0%
NA
40%
% Facilities
Where
PD<1%
39%
0%
0%
0%
NA
40%
% Facilities
Where
PD<10%
59%
0%
34%
100%
NA
80%
% FacUities
Where
PD<50%
77%
0%
100%
100%
NA
100%
% Facilities
Where
PD>50%
23%
100%
0%
0%
NA
0%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
             In Figure 5-15, the facility off-site transfers for energy recovery estimates are
again compared to the site surveyor estimates, but now the difference in facility and site surveyor
estimates are classified according to whether the facility and the site surveyor estimates were
equal, whether the facility estimates were less than the site surveyor estimates, or whether the
facility estimates were greater than the site surveyor estimates. Over half the facilities in SIC
Codes 25, 30, 281, and 286 had off-site transfer for energy recovery estimates less than the site
surveyor estimates.
                                          5-42
-------
              Figure 5-15. Comparison of Facility Estimates and Site Surveyor Estimates
                                    for Off-Site Energy Recovery
 34%
                                                            21%
                         66%
                                                                                                    40%
                                      79%
                                                                        60%
     SIC Code 281 (RY
SIC Code 285 (RY 1994)
SIC Code 286 (RY 1995)
19%
                      20%
          61%
     SIC Code 25 (RY 1994)
      100%

SIC Code 30 (RY 1994)
   I Facility Estimate Equal to Site Surveyor Estimate
   I Facility Estimate Greater Than Site Surveyor Estimate
           I Facility Estimate Less Than Site Surveyor Estimate
-------
5.2           On-Site Waste Management Activities

5.2.1          On-Site Waste Management Activities by SIC Code

              On-site waste management activity quantities as reported by the facilities and the
site surveyors were summed for all chemicals to get total facility on-site waste management
activity quantities. Total facility on-site activity quantities were scaled and summed for all
facilities to get total on-site activity quantities for each SIC Code. The total on-site activity
quantities for SIC Codes are presented by activity in Tables 5-28 through 5-33. In general,
facility and site surveyor estimates showed better agreement for on-site treatment than on-site
recycling for each SIC Code.

              Facilities in SIC Code 286 tend to be much larger than those in the other SIC
Codes surveyed. The process operations performed in these facilities and the many uses for
solvents in these processes create many opportunities  for on-site recycling, treatment, and energy
recovery, as shown by the large amount of chemicals in Table 5-33.
                                           5-44
-------
                                    Table 5-28
      Summary of SIC Code 25 TRI On-Site Waste Management Activity
                     Quantities for RY1994 (millions of Ibs.)
On-Site Waste
Management Activity
Recycling
Treatment
Energy Recovery
Quantity of Chemicals as
Reported by the Facilities
2.36
5.54
0.00
Quantity of Chemicals as
Reported by the Site
Surveyors
3.23
5.26
0.00
Percent Difference*
-27%
5.3%
NA
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no on-site activities of this kind at the facilities surveyed in this SIC Code.
                                    Table 5-29
     Summary of SIC Code 281 TRI On-Site Waste Management Activity
                     Quantities for RY 1994 (millions of Ibs.)
On-Site Waste
Management Activity
Recycling
Treatment
Energy Recovery
Quantity of Chemicals as
Reported by the Facilities
332
79.7
0.00
Quantity of Chemicals as
Reported by tbe Site
Surveyors
328
78.9
0.00
Percent Difference*
1.2%
1.0%
NA
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no on-site activities of this kind at the facilities surveyed in this SIC Code.
                                        5-45
-------
                                     Table 5-30
     Summary of SIC Code 285 TRI Oil-Site Waste Management Activity
                     Quantities for RY 1994 (millions of Ibs.)
On-Site Waste
Management Activity
Recycling
Treatment
Energy Recovery
Quantity of Chemicals as
Reported fay the Facilities
3.48
0.00
0.00
Quantity of Chemicals as
Reported by the Site
Surveyors
2.95
0.00
0.00
Percent Difference*
18%
NA
NA
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no on-site activities of this kind at the facilities surveyed in this SIC Code.
                                     Table 5-31
      Summary of SIC Code 30 TRI On-Site Waste Management Activity
                     Quantities for RY 1994 (millions of Ibs.)
On-Site Waste
Management Activity
Recycling
Treatment
Energy Recovery
Quantity of Chemicals as
Reported by the Facilities
160
32.6
0.00
Quantity of Chemicals as
Reported by the Site
Surveyors
160
33.1
0.00
Percent Difference*
-0.09%
-1.36%
NA
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no on-site activities of this kind at the facilities surveyed in this SIC Code.
                                         5-46
-------
                                     Table 5-32
      Summary of SIC Code 26 TRI On-Site Waste Management Activity
              Quantities for Reporting Year 1995 (millions of Ibs.)
On-Site Waste
Management Activity
Recycling
Treatment
Energy Recovery
Quantity of Chemicals as
Reported by the Facilities
0
258
65.6
Quantity of Chemicals as
Reported by the Site
Surveyors
0
287
75.9
Percent Difference*
NA
-10%
-14%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no on-site activities of this kind at the facilities surveyed in this SIC Code.
                                     Table 5-33

     Summary of SIC Code 286 TRI On-Site Waste Management Activity
              Quantities for Reporting Year 1995 (millions of Ibs.)
On-Site Waste
Management Activity
Recycling
Treatment
Energy Recovery
Quantity of Chemicals as
Reported by the Facilities
702
144
222
Quantity of Chemicals as
Reported by the Site
Surveyors
3,821
223
222
Percent Difference*
-82%
-35%
0.0%
*Percent Difference = (Fa-SS)/SS x 100, where Fa = Facility Estimate and SS = Site Surveyor Estimate.
NA - Not applicable. There were no on-site activities of this kind at the facilities surveyed in this SIC Code.
             The large percent difference in on-site recycling activities in SIC Code 286 is due

to one facility misreporting recycling activities. If this facility was excluded from the analysis,

the percent difference would be less than 1%.
                                         5-47
-------
5.2.2
Oil-Site Recycling
              A comparison of the percent difference between facility estimates and site
surveyor estimates for on-site recycling is presented in Table 5-34.  SIC Codes 281 and 30 had
the best agreement between facility on-site recycling estimates and site surveyor on-site
recycling estimates.
                                      Table 5-34
     Comparison of the Percent Difference (PD)1 Between Facility On-Site
      Recycling Estimates and Site Surveyor On-Site Recycling Estimates
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
49%
49%
0%
21%
NA
60%
% Facilities
Where
PD<1%
49%
49%
3%
100%
NA
80%
% Facilities
Where
PD<10%
49%
87%
3%
100%
NA
80%
% Facilities
Where
PD<50%
80%
100%
100%
100%
NA
80%
% Facilities
Where
PD>50%
20%
0%
0%
0%
NA
20%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
             In Figure 5-16, the facility on-site recycling estimates are again compared to the
site surveyor estimates, but now the difference in facility and site surveyor estimates are
classified according to whether the facility and the site surveyor estimates were equal, whether
the facility estimates were less than the site surveyor estimates, or whether the facility estimates
were greater than the site surveyor estimates.
                                          5-48
-------
                 Figure 5-16. Comparison of Facility Estimates and Site Surveyor Estimates
                                           for On-Site Recycling
      29%
                           49%
   22%
       SIC Code 281 (RY 1994)
                                                               20%
                                                                               20%
                                                                                                       60%
                                                                          20%
                                         80%
SIC Code 285 (RY 1994)
SIC Code 286 (RY 1995)
                                                                21%
51%
                                        79%
       SIC Code 25 (RY 1994)
 SIC Code 30 (RY 1994)
     I Facility Estimate Equal to Site Surveyor Estimate
     I Facility Estimate Greater Than Site Surveyor Estimate
            I Facility Estimate Less Than Site Surveyor Estimate
-------
5.2.3
On-Site Treatment
             A comparison of the percent difference between facility estimates and site
surveyor estimates for on-site treatment is presented in Table 5-35.
                                     Table 5-35
     Comparison of the Percent Difference (PD)1 Between Facility On-Site
     Treatment Estimates and Site Surveyor On-Site Treatment Estimates
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
0%
54%
100%
0%
13%
20%
% Facilities
Where
PEK1%
0%
63%
100%
0%
25%
80%
% Facilities
Where
PD<10%
100%
100%
100%
100%
38%
80%
% Facilities
Where
PD<50%
100%
100%
100%
100%
63%
80%
% Facilities
Where
PD>50%
0%
0%
0%
0%
37%
20%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
             In Figure 5-17, the facility on-site treatment estimates are again compared to the
site surveyor estimates, but now the difference in facility and site surveyor estimates are
classified according to whether the facility and the site surveyor estimates were equal, whether
the facility estimates were less than the site surveyor estimates, or whether the facility estimates
were greater than the site surveyor estimates.
5.2.4
On-Site Energy Recovery
             A comparison of the percent difference between facility estimates and site
surveyor estimates for on-site energy recovery is presented in Table 5-36. Only six facilities
surveyed in SIC Code 281, 286, and 26 performed on-site energy recovery.  All facility
                                         5-50
-------
                 Figure 5-17. Comparison of Facility Estimates and Site Surveyor Estimates
                                            for On-Site Treatment
    20%
                           55%
                                                                                                   12%
25%
        SIC Code 281 (RY
     100%

SIC Code 285 (RY 1994)
                                                                                                           25%
                                                                           63%
SIC Code 286 (RY 1995)
                                                                                    20%
                                                                                                    20%
               100%
          SIC Code 25 (RY 1994)
       100%

SIC Code 30 (RY 1994)
         60%

SIC Code 26 (RY 1995)
      I Facility Estimate Equal to Site Surveyor Estimate
      I Facility Estimate Greater Than Site Surveyor Estimate
           I Facility Estimate Less Than Site Surveyor Estimate
-------
estimates were equal to the site surveyor estimates except for one facility in SIC Code 26. No
facilities surveyed in SIC Codes 25, 285, or 30 performed on-site energy recovery.
                                      Table 5-36
              Comparison of the Percent Difference (PD)1 Between
                Facility On-Site Energy Recovery Estimates and
                Site Surveyor On-Site Energy Recovery Estimates
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
% Facilities
Where
PD=0
NA
100%
NA
NA
75%
100%
% Facilities
Where
PD<1%
NA
100%
NA
NA
75%
100%
% Facilities
Where
PD<10%
NA
100%
NA
NA
75%
100%
% Facilities
Where
PD<50%
NA
100%
NA
NA
100%
100%
% Facilities
Where
PD>50%
NA
0%
NA
NA
0%
0%
'PD = The absolute value of the percent difference, where percent difference = (Fa - SS)/SS x 100, Fa = Facility
Estimate, and SS = Site Surveyor Estimate.
NA - Not applicable. There were no on-site activities of this kind at the facilities surveyed in this SIC Code.
              In Figure 5-18, the facility on-site energy recovery estimates are again compared
to the site surveyor estimates, but now the difference in facility and site surveyor estimates are
classified according to whether the facility and the site surveyor estimates were equal, whether
the facility estimates were less than the site surveyor estimates, or whether the facility estimates
were greater than the site surveyor estimates.
                                          5-52
-------
                            Figure 5-18. Comparison of Facility Estimates and Site Surveyor Estimates
                                                   for On-Site Energy Recovery
                         100%

                   SIC Code 281 (RY 1994}
      100%

SIC Code 286 (RY 1995)
U)
            25%
                                      75%
                   SIC Code 26 (RY 1995)
Facility Estimate Equal to Site
Surveyor Estimate

Facility Estimate Less Than Site
Surveyor Estimate

Facility Estimate Greater Than Site
Surveyor Estimate
-------
5.3           Production Ratio/Activity Index

              The production ratio/activity index is a pollutant specific measure that relates the
changes in business activity between subsequent reporting years. The production ratio/activity
index can be estimated using several methods. The methods are presented below:
              TCM - the ratio of the amount of the chemical manufactured in the current
              reporting year to the previous reporting year;
              TCPV - the ratio of production volume in the current reporting year to the
              previous reporting year;
              TCU - an activity index of the amount of the toxic chemical used in the current
              reporting year to the previous reporting year;
              HR - an activity index of the amount of operating hours for an activity in the
              current reporting year to the previous reporting year;
              WT - an activity index or production ratio based on a weighted average of data
              from several processes; and
              OTH - any other estimation method.
              Figure 5-19 and Table 5-37 present the distribution of use for each method that
was reported by the facilities, by SIC Code.  Site surveyors reviewed the method used by each
facility to determine whether it was the most appropriate. Table 5-38 presents the frequency that
site surveyor's agreed with the facility's choice of method. As shown on the table, facilities
which used a method not listed (as noted by the "other" category), could have used a better
method to determine the production ratio.  Table 5-39 presents the distribution of the most
appropriate method as observed by site surveyors.

              As shown on Table 5-38, the site surveyor disagreed most often with the "other"
basis of estimate. Most production ratios can be accurately accounted for using the amount of
chemical manufactured or used from one year to the next, or the change in production volume.
Facilities would have more accurate activity indices if one of these three bases was used for ratio
estimation.
                                           5-54
-------
                                                Table 5-37

                    Method of Estimate Used by Facilities to Calculate
                                           Production Ratio
SIC Code
25 (RY 1994)
.281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
Percent of
FacUitles
Using TCM
0%
5.4%
2.9%
0%
4.8%
19.6%
Percent of
Facilities
Using TCPV
52.1%
53.6%
67.7%
14.9%
78.6%
57.4%
Percent of
Facilities
Using TCU
4.8%
31.7%
14.0%
52.3%
9.5%
11.5%
Percent of
Facilities
Using HR
21.5%
0.0%
0%
0%
0%
1.6%
Percent of
Facilities
Using WT
0%
0%
0%
5.7%
4.7%
6.6%
Percent of
Facilities
Using OTH
21.6%
9.3%
15.4%
27.1%
2.4%
3.3%
TCM - the ratio of the amount of the chemical manufactured in the current reporting year to the previous reporting year.
TCPV - the ratio of production volume in the current reporting year to the previous reporting year.
TCU - an activity index of the amount of the toxic chemical used in the current reporting year to the previous reporting year.
HR - an activity index of the amount of operating hours for an activity in the current reporting year to the previous reporting year.
WT - an activity index or production ratio based on a weighted average of data from several processes.
OTH - any other estimation method.
                                                    5-55
-------
                          Figure 5-19.  Method of Estimate used by Facilities to Calculate PR/AI
         1.0%
                9-2%     5.3%
                                                 15.4%
                                                              2.9%
                                          14.0%
     31.4%
                                    53.1%
             SIC Code 281 (RY 1994)
                              52.1%
                                             27%
    21.
              4.8%
                                                                                       1.6%
                                                                                           6.6%
                                            3.3%
                                                                                                                19.6%
                                                                                  11.5%
                                                                           67.7%
SIC Code 285 (RY 1994)
                                                                     15%
                                                                       52%
              SIC Code 25 (RY 1994)
SIC Code 30 (RY 1994)
SIC Code 286 (RY 1995)
                                                                                              4.7%2.4%   4.8%
                                                                                       9.5%
                                                                                                                 .6%
SIC Code 26 (RY 1995)
The data for this Figure is shown in Table 5-37.
                                                   ]TCM   i|TCPV  i)TCU
                                                   IHR    • WT    «OTH
-------
                    Table 5-38

Percent of Time Surveyor Agreed with Facility Basis of
             Production Ratio Estimate
SIC Code
25 (RY 1994)
25 (RY 1994)
25 (RY 1994)
25 (RY 1994)
281 (RY 1994)
281 (RY 1994)
281 (RY 1994)
281 (RY 1994)
281 (RY 1994)
285 (RY 1994)
285 (RY 1994)
285 (RY 1994)
285 (RY 1994)
30 (RY 1994)
30 (RY 1994)
30 (RY 1994)
30 (RY 1994)
26 (RY 1995)
26 (RY 1995)
26 (RY 1995)
26 (RY 1995)
26 (RY 1995)
Facility Basis of Estimate
TCPV
TCU
HR
OTH
TCM
TCPV
TCU
HR
OTH
TCM
TCPV
TCU
OTH
TCPV
TCU
WT
OTH
TCM
TCPV
TCU
WT
OTH
Percent of Time Surveyor
Agreed with Basis
61%
100%
100%
51%
100%
99%
100%
100%
15%
100%
100%
100%
0%
89%
100%
0%
48%
100%
91%
100%
100%
0%
                       5-57
-------
                                     Table 5-38 (Continued)

               Percent of Time Surveyor Agreed with Facility Basis of
                                   Production Ratio Estimate
SIC Code
286 (RY 1995)
286 (RY 1995)
286 (RY 1995)
286 (RY 1995)
286 (RY 1995)
286 (RY 1995)
Facility Basis of Estimate
TCM
TCPV
TCU
HR
WT
OTH
Percent of Time Surveyor
Agreed with Basis
100%
100%
100%
0%
100%
50%
TCM - the ratio of the amount of the chemical manufactured in the current reporting year to the previous reporting year.
TCPV - the ratio of production volume in the current reporting year to the previous reporting year.
TCU - an activity index of the amount of the toxic chemical used in the current reporting year to the previous reporting year.
HR - an activity index of the amount of operating hours for an activity in the current reporting year to the previous reporting year.
WT  - an activity index or production ratio based on a weighted average of data from several processes.
OTH - any other estimation method.
                                              Table 5-39

         Method of Estimate That Should Have been Used by Facilities to
                                  Calculate Production Ratio



SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
Percent of
Facilities
that Should
Use TCM
0%
5.8%
2.9%
0%
7.0%
19.4%
Percent of
Facilities
that Should
Use TCPV
31.7%
53.2%
83.1%
13.3%
72.0%
58.0%
Percent of
Facilities
that Should
Use TCU
31.0%
39.5%
14.0%
73.6%
16.3%
14.5%
Percent of
Facilities
that Should
UseHR
21.5%
0.0%
0%
0%
0%
0%
Percent of
Facilities
that Should
UseWT
4.8%
0%
0%
0%
0%
0%
Percent of
Facilities that
Should Use
OTH
11.0%
1.5%
0%
13.1%
0%
1.6%
TCM - the ratio of the amount of the chemical manufactured in the current reporting year to the previous reporting year.
TCPV - the ratio of production volume in the current reporting year to the previous reporting year.
TCU - an activity index of the amount of the toxic chemical used in the current reporting year to the previous reporting year.
HR - an activity index of the amount of operating hours for an activity in the current reporting year to the previous reporting year.
WT - an activity index or production ratio based on a weighted average of data from several processes.
OTH - any other estimation method.
                                                   5-58
-------
5.4           Source Reduction Activities

              The following discussion reviews how accurately facilities indicate source
reduction activities on Form Rs. Starting in reporting year 1991, EPA required facilities to
include on their Form R reports information describing source reduction activities that were
implemented to reduce the quantity of Section 313 chemicals in waste. This information offers
users of the data insight into how often industrial facilities reduce pollution at the sources. To
assess the accuracy of source reduction entries in the TRI database, analyses in this section
address three questions:

              •      Are the source reduction activities that facilities indicate on Form Rs
                     legitimate?
              •      Why do facilities make errors when claiming source reduction?
              •      Do facilities consistently report source reduction activities on Form Rs?

              It should be noted that this section focuses only on source reduction activities that
facilities indicate on "Form Rs." Form As do not include fields for reporting source reduction.

5.4.1         Errors in Classifying Source Reduction

              To identify errors commonly made by facilities and reasons why facilities made
these errors, site surveyors determined during each visit whether facilities indicate source
reduction activities that were consistent with  definitions of source reduction presented in the
EPCRA Section 313 reporting instructions. In cases where facilities did not claim source
reduction activities, site surveyors generally did not determine whether facilities overlooked
source reduction activities. Accordingly, the most recent site survey data are sufficient for
evaluating whether source reduction activities currently loaded in the TRI database are
legitimate, but the data are not sufficient for determining the total number of source reduction
activities that should have been reported.
                                            5-59
-------
              Table 5-40 summarizes how often source reduction activities were used on
EPCRA Section 313 chemicals and how often these claims were made in error.  For reference,
Table 5-41 indicates the source reduction activities most commonly used on EPCRA Section 313
chemicals in the selected SIC Codes.  The data in these tables suggest that facilities in the
furniture manufacturing industry (SIC Code 25), organic chemicals industry (SIC Code 286) and
plastics manufacturing industry (SIC Code 30) claim source reduction much more frequently
than facilities in the inorganic chemical  manufacturing industry (SIC Code 281), paper industry
(SIC Code 26), and paint manufacturing industry (SIC Code 285). Modifications to spray
application and surface coating processes account for a majority of the source reduction activities
claimed by furniture manufacturers. Employee training and improved maintenance account for a
majority of the source reduction activities claimed by organic chemical manufacturers.  No
specific group of source reduction activities were as prevalent for the other industries.

              The data also indicate that, of the source reduction activities claimed by rubber
and plastic manufacturing facilities, nearly three fourths were claimed in error. Many rubber and
plastic manufacturing facilities claimed  that a reduction in the number of toxic chemical supplies
was source reduction even though the reduction in suppliers did not decrease the amount of toxic
chemicals purchased or used. The frequency  of errors was notably lower for furniture
manufacturers and inorganic chemical manufacturers, and no errors were identified in the source
reduction activities claimed by paint manufacturers, organic chemical manufacturers, and
paperboard facilities.

5.4.2         Sources of Errors Made When Claiming Source Reduction

              Specific reasons for erroneously classifying source reduction activities differ from
one facility to the next.  In general, however,  most errors resulted from facilities not
understanding exactly what activities constitute source reduction.  Site surveyors noted several
examples supporting this hypothesis:
                                           5-60
-------
                                                    Table 5-40
                              Errors in Source Reduction Activity Classifications
Reporting
Year
1994
1995
SIC
Code
25
281
285
30
26
286
Frequency with which facilities claim source
reduction activities
Total number of source
reduction activities
claimed by the selected
facilities
48
24
30
21
3
36
Estimated percent of
Form Rs submitted by
facilities in SIC Code
with source reduction
claimed3
33%
14%
21%
32%
5.9%
30.1%
Frequency with which facilities make errors when
claiming source reduction
Number of source
reduction activities
claimed in error by the
selected facilities
8
6
0
14
0
0
Estimated percent of
source reduction
activities that are
claimed, but in error8
27%
22%
0%
78%
0%
0%
1 Percents in this column were calculated using the weighting factors discussed in Section 2.6.3.
-------
               Table 5-41
Most Common Source Reduction Activities
     Claimed by the Selected Facilities
SIC
Code
25
281
285
30
26
286
Source
Reduction
Code
W72
W73
W74
W39
7 others
W21
6 others
W25
W39
W82
W52
3 others
W42
W13
W39
W14
W52
5 others
W32
W58
W13
5 others
W36
W51
W19
Description
Modified spray systems for coating applications
Substituted materials used for coating applications
Improved application techniques for surface coating
Miscellaneous spill and leak prevention
Many different descriptions
Ensuring materials are used before reaching their shelf-life
Many different descriptions
Instituted programs to exchange unwanted materials
Miscellaneous spill and leak prevention
Modified composition of products
Modified equipment, layout, or piping
Many different descriptions
Substituted raw materials
Improved maintenance scheduling and recordkeeping
Miscellaneous spill and leak prevention
Changed production schedule to minimize changeovers
Modified process equipment, layout, or piping
Many different descriptions
Improved practices for loading and unloading chemicals
Process modifications
Employee training and improved maintenance
Many different descriptions
Spill and leak detection program
Recirculation within processes
Reuse of materials
Percent of Chemicals
at Selected Facilities
that used this Code
16.9%
10.0%
9.5%
9.0%
8.5%
0.5%
6.2%
4.0%
4.0%
4.0%
3.0%
13.3%
10.9%
7.1%
7.1%
5.5%
5.5%
12.6%
0.7%
5.9%
23.3%
9.6%
5.5%
5.5%
5.5%
                  5-62
-------
              •      An organic chemical manufacturing facility installed a new pollution
                     control mechanism to remove ammonia from a waste stream. The facility
                     considered the new device as source reduction, but the site surveyor noted
                     that the facility already accounted for the new device as "treatment" and
                     should not have claimed the device as source reduction.

              •      Due to decreased demand for a particular product, a paint manufacturing
                     facility purchased less of a glycol ether solvent. The facility claimed this
                     reduced usage of raw materials as source reduction, but the site surveyor
                     did not consider decline in production demand as source reduction.

              •      A furniture manufacturing facility claimed source reduction after installing
                     a new software system to track purchases of Section 313 chemicals.
                     Although this system helped the facility make more accurate threshold
                     determinations, the site surveyor noted that the software did not reduce
                     amounts of hazardous chemicals that were purchased.  Therefore, the site
                     surveyor concluded that installing the new software was not source
                     reduction.


              Because errors in claiming source reduction resulted primarily from facilities

misinterpreting definitions, EPA can help minimize similar errors in future reporting years by

preparing revised reporting instructions that clarify which activities should, and should not, be

classified as source reduction.
5.4.3          Feedback from Facilities


              The accuracy of source reduction data depends to a great extent on how facilities

choose to implement the reporting requirements. Although the survey instrument did not include
explicit fields for documenting feedback from facilities regarding source reduction, site

surveyors noted several relevant comments made by facility contacts:
                     Some facilities noted that source reduction codes do not inform those who
                     access TRI data of the extent to which emissions are reduced by source
                     reduction. These facilities, therefore, saw little benefit from claiming
                     source reduction activities on their Form Rs.

                     Some facilities found the list of source reduction codes cumbersome,
                     noting that the list contains too many codes or that definitions of specific
                     codes are too vague.
                                          5-63
-------
              •      Some facilities chose to not claim on their Form Rs legitimate source
                     reduction activities because the activities caused only marginal reductions
                     in overall releases.
              •      Facilities seem confused about whether source reduction activities should
                     be reported only for the first year in which they were implemented or for
                     every year thereafter. This confusion leads to inconsistent reporting
                     practices among the facilities that claim source reduction.  This confusion
                     stems from discrepancies in Agency guidance. The agency guidance
                     document states that source reduction activities should be reported only in
                     the first year of implementation while the Pollution Prevention Act states
                     that each annual report will identify the source reduction practices used
                     with respect to each chemical for which the report is submitted.

              These comments from facilities stress that, for a variety of reasons, facilities often
choose to not report legitimate source reduction activities. Therefore, the total number of source
reduction activities currently logged in the TRI database may understate the extent of pollution
prevention efforts adopted by industrial facilities.

5.4.4         Overall Accuracy of Source Reduction Data

              Site surveyors found that facilities in the selected industries frequently
misinterpreted definitions of source reduction and should not  have claimed roughly 30 percent of
all source reduction activities reported to TRI in 1994 and 1995. Observations made by site
surveyors and feedback provided by facilities both suggest that many facilities did not claim
legitimate source reduction activities on their Form Rs, but the current site survey data are
insufficient for evaluating how often this occurs.  Accordingly, the quality of source reduction
data in the  TRI database is compromised by omissions and erroneous submissions of source
reduction information, and TRI data, therefore, may not be an accurate measure of the extent of
pollution prevention efforts for many industries.

              Because the primary cause of errors in reporting source reduction  seems to be due
to facilities misinterpreting definitions, EPA can help improve the accuracy of source reduction
data by preparing TRI reporting instructions that clearly define which activities are,  and are not,
considered to be source reduction.
                                           5-64
-------
6.0           PREPARATION OF THE FORM R

              Site surveyors interviewed facility personnel during each site visit to obtain
general information regarding completion of the Form R reports and to identify trends among the
surveyed facilities. The information obtained during these interviews included quantitative
information such as facility size (the number of employees at the facility), time to complete Form
Rs, the types of personnel primarily responsible for preparing the Form R reports, and the types
of references used by these personnel. In addition, qualitative feedback was obtained on the
Form R Instructions, the Automated Form R (AFR), the TRI Hotline, use of the Form A
beginning in RY 1995, and suggestions for additional guidance that EPA should develop to assist
facilities in release and other waste management quantities estimation and Form R preparation.

6.1           Facility Personnel and References

Table 6-1  identifies the percentage of facilities visited in a particular size range (based on number
of employees) for each SIC Code group. As can be seen from the table, most of the inorganic
chemical and paint manufacturing facilities had fewer than 50 employees, while the furniture,
paper, organic chemicals, and plastics related industries had between 50-499 employees on
average.  In general, the size of the facilities visited for the RY 1994 and RY 1995 analysis were
smaller than those visited in the RY 1987 and RY 1988 site visits.

              Table 6-2 lists the types of personnel identified by the facility as being primarily
responsible for preparing the Form R reports for each SIC Code group included in this analysis.
As can be seen on the table, facility and corporate environmental staff most often completed the
reports for each of the source categories visited. A brief description of each staff type identified
in the table follows:
                    Facility Environmental - In most cases, this is a full-time position for an
                    on-site employee whose primary responsibility is dealing with
                    environmental issues.
                                          6-1
-------
                     Corporate Environmental - This person would have environmentally-
                     related responsibilities for more than one individual facility and may or
                     may not be physically located at the facility.

                     Facility Staff- This is an employee whose responsibilities extend beyond
                     the environmental arena. This staff type was primarily found to be
                     preparing the Form R reports at facilities that had fewer than 50
                     employees.

                     Consultant/Contractor - This includes personnel contracted outside the
                     company to prepare the facility's Form R report.

                     Safety Personnel - This staff type is similar to Facility Environmental.
                     This person may have responsibilities including complying with
                     Environmental Health and Safety issues as well.

                     Other - This is anyone who filled out the Form R that does not belong to
                     one of the previously described staff types.
                                        Table 6-1
                     Number of Employees at Visited Facilities
Percentage of Facilities with a given Number of Employees"
Employee
Range
10-49 employees
50-499
employees
>500 employees
RY 1987
SIC
Code
20-39
17
62
21
RY 1988
SIC
Code
28, 291
45
48
7
SIC
Code
34-38
7
42
51
RY1994
SIC
Code
25
4
82
14
SIC
Code
281
77
23
0
SIC
Code
285
61
39
0
SIC
Code
30
19
81
0
RY 1995
SIC
Code
26
0
70
30
SIC
Code
286
40
60
0
"The 1987 and 1988 data are raw percentages of the facilities actually surveyed and are not scaled up to represent the entire SIC Code. The
1994 and 1995 data are scaled data which are weighted to represent the entire SIC Code.
                                            6-2
-------
                                        Table 6-2
                    Types of Personnel Completing the Form R
Percentage of Facilities using a Particular Staff Type to Prepare their form &•
Staff Type
Facility Environmental
Corporate Environmental
Facility Staff
Consultant/Contractor
Safety Personnel
Other
3987
SIC
Code
20-39
47
29
11
10
2
1
1988
SIC
Code
28, 291
29
28
37
1
5
0.00
SIC
Code
34-58
38
11
21
15
15
0.00
1994
SIC
Code
25
31
31
25
12
8
0.00
SIC
Code
281
53
17
38
12
14
6
SIC
Code
285
24
25
67
18
1
8
SIC
Code
30
43
14
36
13
9
0.00
199S
SIC
Code
2$
80
10
10
0
0
0
SIC
Code
28$
60
0
50
0
0
0
 Totals may equal more than 100 percent due to facility personnel identifying themselves as more than one staff type. The 1987 data are raw
percentages of the facilities actually surveyed and are not scaled up to represent the entire SIC Code. The 1988,1994, and 1995 data are scaled
data which are weighted to represent the entire SIC Code.
              Table 6-3, Figure 6-1, and Figure 6-2 identify the references most commonly used
by facilities to prepare their Form R reports. As can be seen on Table 6-3, more than 90 percent
of the facilities visited used the TRI Reporting Form R instructions for RY 1994 and RY 1995 as
compared to less than half who used the reference for preparing their RY 1988 Form R reports.
In addition, a higher percentage of facilities are now using Trade Association Materials, Privately
Sponsored Seminar Materials, computer programs, and EPA Sponsored Training Workshops
than were used in the past. SIC Code 26, sampled for RY 1995, relies heavily on guidance from
NCASI, a pulp and paper research organization, in completing Form Rs and documenting release
calculations. EPA Sponsored Training Workshops have also contributed to increased use of
EPA's compilation of air pollutant emission factors document, AP-42. The increased use in
workshop or other training materials and computer programs is due to the greater availability of
such resources than were available in the past.
                                            6-3
-------
                                                Table 6-3
                   Common References Used to Compile the Form Rs
Percentage of Facilities using a Particular Reference*
Reference
TRI Reporting Form R
Instructions
The Emergency Planning and
Community Right-to Know
Act, Section 3 13
Estimating Releases and
Waste Treatment Efficiencies
for TRI
EPCRA Section 313 Release
Reporting Guidance,
Estimating Chemical
Releases
Compilation of Air Pollution
Emission Factors, AP-42
Toxic Chemical Release
Reporting Proposed Rule, 52
FR 21152, June 4, 1987
Industry Trade Association
Materials
Privately Sponsored Seminar
Materials
EPA Sponsored Training
Workshop
Computer Programs
1987
SIC
Code
20-39
90
26
24
15
11
10
10
9
N/A
N/A
1988
SIC
Code
28, 291
44
9
17
0
12
3
10
5
N/A
3
SIC
Code
34-38
44
13
9
8
3
7
9
7
N/A
3
1994
SIC
Code
25
91
N/A
22
10
15
N/A
12
16
13
30
SIC
Code
281
93
N/A
8
9
17
N/A
13
7
32
23
SIC
Code
285
96
N/A
22
1
33
N/A
24
26
27
15
SIC
Code
30
100
N/A
11
0
5
N/A
9
7
38
39
1995
SIC
Code
26
90
N/A
20
10
20
N/A
90
0
20
40
SIC
Code
286
100
N/A
10
10
60
N/A
20
10
10
40
"Totals may equal more than 100% as facilities often used more than one reference. The 1987 data are raw percentages of the facilities actually
surveyed and are not scaled up to represent the entire SIC Code.  The 1988, 1994, and 1995 data are scaled data which are weighted to
represent the entire SIC Code N/A means the reference was not listed in the questionnaire in this reporting year and was not specifically listed
by any of the facilities.
                                                     6-4
-------
                                                                      S-9
t
§

I

S1


Q.
O




I
05
CO
                             Form R Instructions



        Estimating Releases and Waste Treatment

        Efficiencies for the Toxic Chemical Release

                   Inventory Form



EPCRA Section 313 Release Reporting Guidance,

   a through q, Estimating Chemical Releases
                                                      AP-42
                                  Trade Association Guidance
                       Privately Sponsored Seminar Materials
                                      Computer Programs
                  The Emergency Planning and Community

                       Right-to-Know Act, Section 313
              Toxic Chemical Release Reporting Proposed

                   Rule, 52 FR 21152, June 4, 1987
                    I
                    o
                  •< 3
                  £ 3
                                                                                                                        (O

                                                                                                                         I
                                                                                                                         O)
o
o


3
o
•3

7J
CD
                                                                                                                         §
                                                                                                                         O
                                                                                                                         CD
                                                                                                                         (A
                                                                                                                         CD
                                                                                                                         a
                                                                                                                         O
                                                                                                                         o

                                                                                                                        •D_

                                                                                                                         cb~
                                                                                                                         •n
                                                                                                                         o
                                                                                                                                           

                                                                                                                                           31
                                                                                                                         00
                                                                                                                         -J
                                                                                                                         fl)
                                                                                                                         3
                                                                                                                         a.

                                                                                                                         3
                                                                                                                                           oo
                                                                                                                                           oo
-------

                                                                9-9
o
OJ
3
I
CL
§
5-

o>

CO
                          Form R Instructions



      Estimating Releases and Waste Treatment

      Efficiencies tor the Toxic Chemical Release

                 Inventory Form


Title III Section 313 Release Reporting Guidance,

   a through q, Estimating Chemical Releases
                                    AP-42
                 Trade Association Guidance
        Privately Sponsored Seminar Materials
                     Computer Programs
        EPA Sponsored Training Workshop
     r+ 
-------
6.2          Amount Of Time Needed To Prepare Form R Reports

             Tables 6-4, 6-5, and 6-6, and Figures 6-3, 6-4, and 6-5 show the number of hours
required to collect the necessary data and complete all the Form Rs for facilities surveyed in RY
1988, RY 1994, and RY 1995, respectively. Section 8 of the Form R was added after RY 1987
and RY 1988 Form Rs were completed.  Thus, one would expect an increase in the amount of
time needed to collect the necessary data and complete the Form Rs in the 1994 and 1995
reporting years.  Even so, as was the case with the RY 1988, in RY 1994 the majority of facilities
reported taking less than 24 hours to complete all of their Form Rs.  There was even an observed
decrease in the percentage of facilities reporting taking over 100 hours to complete their Form Rs
in RY 1994. Facilities surveyed in RY  1995 tended to be larger and had more Form Rs than the
facilities surveyed in previous years. Thus, the amount of time needed to fill out all Form Rs
was greater in RY 1995, as shown in Table 6-6.

             In order to take the number of Form Rs filled out by the facilities surveyed into
consideration when calculating the average time required by facilities  to fill out Form Rs, an
analysis was done which divided the maximum number of hours in the range checked by the
number of Form Rs filled out by the facility. The number of hours needed to fill out each Form
R is presented by SIC Code and reporting year in Table 6-7.
                                          6-7
-------
                                       Table 6-4
     Number of Hours Required to Complete all the Form Rs for RY 1988
Time Estimate
^8 hours
9-24 hours
25-40 hours
4 1-1 60 hours
>160 hours
Percentage of Facilities
SIC Codes 28 and 291"
32
6
6
25
15
SIC Codes 34-38
71
5
24


 Totals do not equal 100% because not all facilities reported the time estimate. These data are scaled data which are weighted to represent the
entire SIC Code.
                                       Table 6-5
     Number of Hours Required to Complete all the Form Rs for RY 1994
Time Estimate
^20 hours
2 1-50 hours
5 1-1 00 hours
100-200 hours
>200 hours
Percentage of Facilities
SIC Code 25'
42
48
6
0
0
SIC Code 281
63
19
12
6
0
SIC Code 285
61
38
0
1
0
SIC Code 30
56
42
2
0
0
* Totals do not equal 100% because not all facilities reported the time estimate. These data are scaled data which are weighted to represent the
entire SIC Code.
                                           6-8
-------
                                                          6-9
o
CO
5T
f
 o
 01
 3
 O-
 CD

 S1

 Q.
Percent of Facilities
                            S   8
                                                                                                                     (Q
                                                                                                                     C
                                                                                                                     0>
                                                                                                                     CO
                                                                                                                      (D
                                                                                                                      Z
                                                                                                                      (D

                                                                                                                  S  8-
                                                                                                                  §  8.
                                                                                                                  "S  s1
                                                                                                                  ro  o
                                                                                                                   
                                                                                                                   -•* -n
                                                                                                                   fi) O

                                                                                                                   Q. 3

                                                                                                                   ^	
                                                                                                                    i  3
                                                                                                                       (O
                                                                                                                       00
                                                                                                                       00
                                                                                                                       0)
                                                                                                                       o
                                                                                                                       o
                                                                                                                       o
                                                                                                                       a
                                                                                                                       A
-------
                                      01-9
                                   Percent of Facilities
0>
Q.
O.

c
•a


5
 "O
S° to

is 5
 ft)  O
 3  g
                                                                                                 71



                                                                                                 (O
                                                                                                 to
                                                                                                  o

                                                                                                  o
                                                                                                  O
                                                                                                  Q.
                                                                                                  (D
-------
                                   Table 6-6
     Number of Hours Required to Complete all the Form Rs for RY 1995
Time Estimate
s8 hours
9-20 hours
2 1-40 hours
41-100 hours
> 100 hours
Percentage of Facilities
SIC Code 26
20
10
40
20
10
SIC Code 286
20
0
10
60
10
These data are scaled data which are weighted to represent the entire SIC Code.
                                   Table 6-7
          Average Number of Hours Needed to Complete a Form R
SIC Code
25 (RY 1994)
281 (RY 1994)
285 (RY 1994)
30 (RY 1994)
26 (RY 1995)
286 (RY 1995)
RY1 994 Overall
RY 1995 Overall
Time Estimate (Hours)
12.9
9.9
11.5
13.4
9.4
8.8
11.7
9.0
These data are scaled data which are weighted to represent the entire SIC Code.
                                     6-11
-------
f
O
                                                             21-9
o-
CD


C

Q.
g

5"

O)

O)
                                    Percent of Facilities
                   CD


                   a

                   SL
I

Q.

CD


I
CD
                             _^   K>   W
                             o   o   o

                             vP   vg   ^S
                             0s*   0s-   o^
?   #   *   *   *   f
                                                                                                         •n

                                                                                                        to'
                                                                                                         c

                                                                                                         3
                                                                                                         o>
                                                                                                         0,
                                                                                                         (D

                                                                                                         •z.
                                                                                                         a>
                                                                                                         (D
                                                                                                         a
                                                                                                         (D
                                                                                                         a

                                                                                                     O  o

                                                                                                     o   o

                                                                                                    -£   °
                                                                                                    ^   3


                                                                                                     -I
                                                                                                     o>  g.
                                                                                                     B)   (D

                                                                                                     2   •"
                                                                                                     Q.  o
                                                                                                     ro
                                                                                                                           7J

                                                                                                                           _i.
                                                                                                                           
-------
              It is recognized that the average time needed per Form R is a function of the
maximum number of hours in the lowest range checked. (The lowest range is different for
facilities surveyed in RY 1994 and RY 1995.)  However, the time estimates listed represent an
average range, and are significantly lower than the 43 hours needed per Form R listed in the 1995
EPCRA Section 313 reporting instructions.

6.3           Use of the Hotline

              For RY 1994 and RY 1995, 33 percent and 50 percent, respectively, of the
facilities visited reported calling the hotline. Figure 6-6 shows the percentage of facilities calling
the hotline for each of the SIC Codes included in this analysis.

              As can be seen by Figure 6-6, personnel at approximately half of the inorganic
chemical (SIC Code 281) and paint and allied products (SIC Code 285), paperboard (SIC Code
26), and organic chemical (SIC Code 286), facilities called the hotline, compared to
approximately one-fourth for the furniture (SIC Code 25) and plastics manufacturing facilities
(SIC Code 30).

              Most of the respondents in RY 1994 and RY 1995 (86 percent) indicated that the
hotline response was helpful.  However, several facilities stated that they had difficulty in getting
through to speak to an operator, and in some cases the answers provided were not consistent.
The majority of facilities stating that the hotline was not helpful reported being unable to get
through to an operator.

6.4           Comments on the Form R Instructions

              Surveyors also interviewed facility personnel in an effort to gain general
comments on the Form R instructions.  One commenter suggested that tabs or a "quick reference
guide" be included in the instructions to allow for easy navigation through the separate sections
of the report. Other areas of the report which were stated as being unclear included how to
                                          6-13
-------
submit a revised form and how to distinguish or document off-site transfers for repackaged
materials.
             Table 6-8 shows the number of respondents who identified a particular aspect of
the chemical specification information as being unclear.

                                    Table 6-8

    Comments on the Form R Chemical Specific Instructions for RY 1994
Subject Area
Toxic Chemical Identity
Mixture Component Identity
Activities and Uses of the Toxic
Chemical
Releases to the Environment On-Site
Transfers in Waste to Off-Site
Locations
On-Site Waste Treatment Methods
and Efficiency and On-Site Energy
Recovery and Recycling Methods
Source Reduction and Recycling
Activities
Number of Respondents stating a Subject Area
was unclear
RY 1994
SIC
Code
25
0
1
1
1
1
0
1
SIC
Code
281
1
0
2
1
1
1
1
SIC
Code
285
0
0
1
0
0
1
0
SIC
Code
30
1
0
0
0
0
0
1
RY 1995
SIC
Code
26
1
0
1
0
0
2
2
SIC
Code
286
0
0
1
0
0
3
2
             The general comments received on the Form R instructions are summarized for
each SIC Code as follows:
                                       6-14
-------
SIC Code 25 - Furniture Manufacturing (RY 1994)

•      One facility needed more information on estimating pollutant releases for
       a material when only a range of concentration is provided on the MSDS.

•      The definitions for material usage type (manufactured, processed,
       otherwise used) are unclear.  In addition, more clarification is needed on
       estimating production ratio.
SIC Code 281 - Inorganic Chemicals Manufacturing (RY 1994)

•      Several facilities requested more details on identifying and quantifying
       treatment, recycling, and repackaging activities.  More examples would be
       helpful.

•      More guidance is needed on estimating future releases needed for Section
       8.
SIC Code 285 - Paint Manufacturing (RY 1994)

•      Need assistance in estimating releases associated with wastewater
       treatment plant discharges.
SIC Code 30 - Rubber and Plastics Manufacturing (RY 1994)

•      Several facilities had questions on how to determine production
       ratio/activity index.
SIC Code 26 - Paper and Paperboard Manufacturing (RY 1995)

•      Many facilities reported that the guidance for ammonia threshold and
       release calculations are unclear.

•      One facility would like clarification on the definitions of energy recovery,
       treatment and recycling.
SIC Code 286 - Organic Chemicals Manufacturing (RY 1995)

•      The definitions for material usage type (manufactured, process, otherwise
       used) are unclear.
                            6-15
-------
6.5          Comments on the Automated Form R (AFR)


             Approximately half of the facilities surveyed for RY 1994 and RY 1995 used the

AFR to help them prepare their Form R's. The majority of those that used the form stated that it

was helpful, with only a handful stating that it was not helpful. This information is shown

graphically in Figure 6-6.


             The types of feedback provided on the AFR is summarized as follows:
                    In general, the AFR is easy to use and does help minimize errors.  In
                    particular, it's nice that the common data (such as facility name) is
                    maintained from year to year and only the release information needs to be
                    updated.  Several commenters stated that it seems to be getting better with
                    each release version.

                    Several commenters who attempted to use the AFR could not get it
                    running on their systems.  This was more common with Windows NT
                    users. Windows NT does not seem to properly interface with the AFR.

                    There were numerous instances of individuals having difficulties in
                    printing their reports once finished.

                    One commenter stated that it took 6 weeks for a phone call requesting
                    assistance in using the AFR to be returned.

                    The AFR could use more range checks, error checks, and look up tables.
                    In addition, transcription errors cannot be caught using the AFR.
                                         6-16
-------
             Figure 6-6.  Percent of Facilities Calling the Hotline by
                        Industry for RY 1994 and RY 1995
70%
 0%
        SIC Code 25
        (RY1994)
SICCodeSO
 (RY1994)
SIC Code 26
 (RY1995)
SIC Code 286
 (RY1995)
                                     Industry
          Figure 6-7.  Percent of Facilities using the Automated Form R
                             for RY 1994 and RY 1995
              D'd not
             Use APR
               46%
             Used APR (found it
                  helpful)
                   51%
                            Used APR (found it
                               not helpful)
                                  3%
                                   6-17
-------
6.6           Comments on Use of the Form A

              In general, facilities like the new Form A and found it easy to use. However,
results show that some facilities incorrectly used the Form A when they should have used the
Form R.  Facilities must use the Form R if they manufacture, process, or otherwise use at least 1
million pounds of the toxic chemical. Some facilities did not realize this upper limit existed.
Also, facilities must use the Form R if their total annual reportable amount is greater than 500
pounds. The annual reportable amount includes those quantities released, including disposed or
treated, recovered at the facility from recycle operations, combusted at the facility for the
purpose of energy recovery, and transferred off-site for purposes of recycle, energy recovery
treatment, or disposal. Some facilities interpreted annual  reportable amount to be releases only,
and therefore incorrectly used the Form A even when their actual reportable amount was greater
than 500 pounds. Some facilities did comment that since they have to estimate their releases and
other waste management quantities regardless of which form they use, they might as well
complete the full Form R.
                                          6-18
-------
7.0           CONCLUSIONS

              Overall, the agreement between facility and surveyor estimates was good.
General trends noted in the RY 1994 data were that the total releases claimed by the facility for
all SIC Codes surveyed were less than the total releases claimed by the site surveyors.  RY 1995
data showed that the total releases and other waste management quantities claimed by the facility
approximately equaled those quantities claimed by the site surveyors.  The percent difference in
total releases and other waste management quantities between the facility calculations and the
surveyor calculations ranged from 0.8% different in SIC Code 26 (paper and paperboard
manufacturing) to 31% different in SIC Code 285 (paint manufacturing).  The relatively high
percent difference for total releases and other waste management quantities in SIC Code 285 is
due to two facilities which did not understand on-site recycling quantities, and, therefore, did not
report them.  If these two facilities are excluded from the analysis, the percent difference is 20%.
Total aggregate releases and other waste management quantities calculated by facilities and site
surveyors for all SIC Codes surveyed in RY 1994 and RY 1995 differed by 4%. Engineering
calculations and mass balances were the most common methodologies used by facilities to
determine releases.

              Facilities in all SIC Codes  tended to overestimate stack emissions and
underestimate fugitive emissions. Many facilities were confused by the definitions of these
emissions and misclassified fugitive emissions  as stack emissions. Surveyors also identified
frequent errors to the classification of transfers  to off-site disposal/treatment/energy
recovery/recycling.

              Significant differences were observed between the quantity of chemicals reported
released and otherwise managed by surveyed facilities, scaled to the entire SIC Code, and the
quantity of chemicals released and otherwise managed by all facilities in that SIC Code reported
in the TRIS database.  While the results of the site survey program are useful to identify trends in
the data, common errors, and the relative accuracy of the data, this finding suggests the absolute
magnitude of releases and other quantities managed as waste or errors in these estimates at  the
surveyed facilities should not be used to represent the  entire  SIC Code.
                                           7-1
-------
              TRI site survey results apply to aggregate data only, that is, some of the apparent
data accuracy comes from the cancellation effects of under- and over-reporting of releases. The
accuracy referred to here is not "true accuracy" but a measure of how well facilities used best
available data and estimation techniques. The results of the site survey, therefore, should not be
used to evaluate the accuracy of the data from an individual facility or a single release stream.

              Evaluation of the threshold calculations performed revealed that for 90% of the
chemicals across all SIC Codes surveyed threshold determinations were correctly determined.
Five percent were incorrectly reported as exceeding thresholds, and five percent were incorrectly
omitted from Form R reporting. Comparison of the 6 SIC Codes surveyed for RY 1994 and  RY
1995 reveal that SIC Codes 25 (furniture manufacturing), 26 (paper manufacturing), and 30
(rubber and plastic manufacturing) facilities had the best accuracy for determining thresholds, as
98% of the toxic chemicals thresholds were correctly determined. SIC Code 281 (inorganic
chemical manufacturing) facilities were the least accurate in correctly determining thresholds, as
84% of the toxic chemicals were correctly determined.

              Evaluation of on-site activities revealed that organic chemical manufacturing
facilities (SIC Code 286) do much more on-site treatment and recycling than facilities in the
other SIC Codes surveyed.  In general, the site surveyors  agreed with the facilities releases and
other waste management estimates for most on-site activities.

              The increasing number of trade association conferences and amount of EPA and
trade association guidance has increased the quality of TRI reporting. Some industries, like the
pulp and paper industry, have developed their own guidance manuals for facilities in their
association to use in filling out the TRI reports. Other industries, such as the paint manufacturing
industry, seem to know less about TRI reporting, estimation techniques, and the documentation
requirements.  In such industries, additional guidance on the amount and types of documentation
needed for accurate TRI reporting, and the methodologies used to estimate releases and other
waste management activities would help reduce the number of errors in threshold determinations
and release estimates.
                                           7-2
-------
8.0          RECOMMENDATIONS

             This section presents several recommendations for the EPCRA Section 313
program based on the results and conclusions of the RY 1994 and RY 1995 data quality
assessment based on site surveys. Improvements in reporting guidance and in the reporting
instructions, as well as facilities' experience in completing Form R reports for the previous
reporting years will continue to improve the data quality in the TRI database. Recommendations
for continued improvement of the TRI database are listed below.

8.1          Additional Guidance Concerning Form R Instruction and Documentation

             General recommendations noted by site surveyors for all SIC Codes include
introducing and explaining the Question/Answer document and sixteen guidance documents
currently available from EPA in the front of the TRI instructions. Many facilities and trade
associations did not read the entire TRI instruction booklet and, therefore, were not aware these
documents existed. Specific comments from facilities in each of the SIC Codes visited are as
follows:

             SIC Code 25 - Furniture Manufacturing (RY 19941

             •      Better definitions are needed in order to distinguish between manufacture,
                    process, and otherwise use activities.
             •      There were several comments with the general sentiment that the EPA and
                    State and Local agencies should attempt to consolidate environmental
                    reporting and standardize acronyms and definitions for terms such as
                    "fugitive" emission sources. One specific suggestion was for a "Dummies
                    guide to environmental reporting."
                                          8-1
-------
SIC Code 30 - Rubber and Plastics Manufacturing (RY 1994)
       EPA should develop "a good, clean web-site" for TRI data. The
       commenter stated that the TRI CD-ROM was helpful and possibly could
       serve as the  structure for a web site.

       More information was requested on determining production ratio.  This
       was specifically requested for those EPCRA Section 313 Chemicals
       produced as by-products or where the production ratio is determined by
       something other than the annual production ratio of the final product.
SIC Code 281 - Inorganic Chemicals Manufacturing (RY 1994)
       One commenter stated that much of the information reported as part of the
       Form R could be obtained by EPA from other sources such as Title V Air
       Pollution permits.

       One general comment from several commenters was that more
       information is needed on estimating fugitive emissions from sources such
       as wastewater treatment and mixing tanks.

       Facility Personnel also requested more guidance on classifying and
       quantifying recycling and source reduction activities.
SIC Code 285 - Paint Manufacturing (RY 1994)
       An industry-specific guidance manual was requested, including examples
       specific to chemicals and release types associated with paint
       manufacturing.

       One consultant who worked on the Form R preparation felt that the
       estimation release guidance could be geared to a more technical audience.

       The guidance is not clear on what "working losses" are for storage tanks.
       Another commenter felt that the range of loss factors for paint mixing was
       too wide.
                             8-2
-------
              SIC Code 26 - Paper and Paperboard Manufacturing CRY 1995)


              •      Some facility contacts said they would have benefited from published
                    treatment or removal efficiencies as well as published partition
                    coefficients.

              •      Many facilities requested more definitive ammonia reporting guidance.


              SIC Code 286  - Organic Chemicals Manufacturing (RY 1995)
                    One general comment from some facilities was that greater availability of
                    the guidance manuals was needed.  Facility contacts suggested industry
                    specific guidance, targeting (and listing) specific SIC Codes be posted on
                    the Internet.
8.2          Additional Guidance Concerning Threshold Determinations


                    Although the nature and extent of threshold determinations varies from

one industry to the next, some general lessons can be learned from the mistakes identified by the

site surveyors. Table 8-1 lists common errors made by facilities when determining thresholds

and offers several recommendations to avoid making such errors in the future. These

recommendations may also be useful to EPA when developing future releases of TRI reporting

instructions.
8.3          Additional Guidance Concerning Release and Other Waste Management
             Quantity Estimates
             Table 8-2 lists common errors made by facilities in all SIC Codes surveyed when

estimating releases and other quantities managed as waste, and offers several recommendations

to avoid making such errors in the future.
                                          8-3
-------
                                     Table 8-1
    Recommendations for Avoiding Errors in Threshold Determinations
            Error Observed in
         Determining Thresholds
  Recommendation for Avoiding Error in
       Future TM Reporting Years
Facility did not document results of threshold
determinations.
Reporting instructions should emphasize that
documentation requirements apply to both
threshold determinations and release
estimates.
Facility assumed Section 313 chemicals
exceeded thresholds, rather than calculating
annual usages and comparing these to
reporting thresholds.
Facilities should be informed that assuming
thresholds are exceeded, rather than
calculating annual usages for Section 313
chemicals, is a common source of errors in
TRI reporting. Reporting instructions should
encourage facilities not to assume thresholds
are exceeded, even for chemicals used in very
large or very small quantities.
Facility overlooked Section 313 chemicals
that were purchased in mixtures.
Facilities should carefully review the most
recent MSDS for every mixture brought on
site to identify all Section 313 chemicals used
during a reporting year.
Facility considered only raw materials used
for production and overlooked chemicals used
for other purposes.
Facilities should take a systematic approach to
identify all chemicals and mixtures used in
production and non-production capacities,
including catalysts, underground injection
well treatment chemicals, wastewater
treatment chemicals, and the like.
Facility environmental staff was unaware that
certain Section 313 chemicals were used at the
plant.
Facilities should implement measures, such as
chemical usage logs or hazardous chemical
inventories, to ensure that environmental staff
are aware of all Section 313 chemicals used in
industrial applications.
Facility did not account for EPA's most recent
releases of threshold determination guidance.
EPA should enhance outreach efforts to
ensure that all facilities are aware of revised
reporting guidelines well in advance of
submission deadlines.
                                         8-4
-------
                                    Table 8-2

                   Recommendations for Avoiding Errors in
                    Identifying Release Types and Sources
             Observed Error
 Recommendation for Avoiding Error in
       Future TRI Reporting Years
Fugitive emissions from general indoor air
reported as stack emissions when released
from a single building vent
Stack and fugitive releases need to be better
defined, especially regarding general room
air that is simply moved to one release point
on top of the building (without air pollution
control devices). Note: Many state
definitions are exactly opposite the TRI
definition in this instance.
Use of outdated SOCMI emission factors
Instructions should note that SOCMI factors
have been updated.
Overlooked stack emissions from storage
tanks, or reporting these emissions as
fugitives.
Instructions should emphasize this potential
release source and briefly discuss the
definition of loading, working, and breathing
losses from tanks (and the methodology to
calculate them).
Overlooked container residue
Instructions should emphasize that even a
"RCRA empty" drum is expected to contain a
residual (possibly up to two inches) and that
it must be considered for TRI reporting.
Also, note that on-site drum rinsing and
disposal of the rinsate will result in a release.
Overlooked coincidental manufacturing
Instructions should indicate that if coal
and/or fuel oil are used in boilers/burners
there is a potential for coincidental
manufacture (and release) of various EPCRA
Section 313 Chemicals (such as H2SO4, HC1,
and HF).
Incorrectly reporting release disposition for
off-site transfers
Instructions should emphasize that facilities
should attempt to determine the type of
receiving facility that is accepting the
transfers.
                                         8-5
-------
                             Table 8-2 (Continued)
                  Recommendations for Avoiding Errors in
                    Identifying Release Types and Sources
             Observed Error
 Recommendation for Avoiding Error in
      Future TRI Reporting Years
Questions over catastrophic releases
EPA should provide guidance as to what
quantity and frequency designate a release for
Section 8.8.
Questions over on-site recycling
EPA should provide a definition of recycling
and include examples of streams that can be
considered as being recycled in Sections 7
and 8. An example would be used solvents.
Specifically, is a "used" solvent that is
collected and processed in the next batch
considered recycled/reused, or does it have to
be separated first (e.g., distilled or filtered).
This is not clear in the current guidance.
Questions over source reduction
EPA should consider shortening the list of
codes for source reduction and should
provide definitions for each code.
Questions over on-site treatment
EPA should provide definitions for each code
and clarify whether separation technologies
should be considered (for example, many
facilities were unsure if a water scrubber that
simply moved a EPCRA Section 313
Chemical from the air to a water stream
should be considered as treatment).
Questions over energy recovery
EPA needs to define situations for energy
recovery.  Examples that came up included
pulp mills which use Kraft recovery boilers.
The main purpose is to generate stream, but
at the same time waste material that contains
EPCRA Section 313 Chemicals is being
burned and destroyed.  Some people argued
that the BTU value was high enough to claim
energy recovery. Other people argued that
the intent for throwing certain wastes  into the
boiler was to destroy (and treat) the EPCRA
Section 313  Chemicals.
                                        8-6
-------
                              Table 8-2 (Continued)
                   Recommendations for Avoiding Errors in
                     Identifying Release Types and Sources
             Observed Error
  Recommendation for Avoiding Error in
       Future TRI Reporting Years
Overlooking On-Site Recycling: One facility
noted that although they employ considerable
on-site recycling activities (to recover/reuse
expensive raw materials that contain EPCRA
Section 313 Chemicals); they do not report
them for TRI because claiming recycling
would require them to register as a hazardous
waste treatment facility for their state.
There needs to be a consistent definition
between states and TRI for recycling. EPA
and state agencies should discuss this
situation and provide appropriate guidance.
Questions over Section 8 amounts.
Facilities would like a simple formula for
releases in each block of Section 8. (e.g.,
Section 8.1 = 5.1 + 5.2 + 5.3 + 6.2 (disposal
only)). This will cut down on errors and
double counting.
There needs to be clarification of the
treatment definitions in Sections 7 and 8 of
the Form R.
The definitions in the two sections are
currently different, and this can cause
problems when reporting. Confusion occurs
when: 1) chemicals go through a treatment
system but are not destroyed  - facilities need
direct guidance to claim "0" efficiency, and
then what to put in Section 8 (0 or NA); 2)
facilities may report the amount sent to
treatment vs the amount treated.  This is
confusing because facilities are supposed to
report the amount sent to energy recovery
and the amount sent to recycling, but not the
amount sent to treatment (they should
correctly report the amount treated instead).
There needs to be clarification on how to
calculate production ratio for "otherwise
used" chemicals.
Facilities often used quantities purchased or
released from year to year rather than an
activity index, even though the guidance
specifically states not to do this.
                                        8-7
-------
-------
Appendix A
-------
-------
                          SURVEY INSTRUMENT
CEB.TRI\1030.nh
-------
-------
                                          FacilitylD:  _______ HJ_U
                                 1997
                      (TRI REPORTING YEAR 1995)
                          TRI DATA QUALITY
                         SURVEY INSTRUMENT
CEB.TRI\1030.nh
-------
-------
                                                               Facility ID:
                                      FACILITY FACT SHEET
Date of Visit:	  through.






Facility Name:	
Facility Address:
City:
State: 	   Zip Code:
Mailing Address (if different from street address):
Telephone:	         Fax:






Facility Contact: 	






Site Surveyors:  	
Pre-visit Telephone Contact:
Reviewers:
CEB.TRIM030.nh                                    A-l
-------
                                                             Facility ID:  |_|_|_|-|_l_l_|-|_l_l_l
                                            PRE-VISIT
                                     TELEPHONE CONTACT
TECHNICAL REVIEW
\.       How many Form R (reporting year 1995) chemical reports were submitted for this facility?
2.       How many 313 chemicals were identified by this facility, but not reported, for reporting year 1995?
3.       Did the facility submit any revised Form R chemical reports for reporting year 1995?



        YES....D                             NO...D    (StiptoQ.5)


4.       List the chemicals which had revised chemical reports.
5.      Did the facility submit any withdrawal requests to EPA for the reporting year 1995?



        YES....D                             NO...D   (ShptoQ.7)


6.      List the chemicals which had withdrawal requests.



        	                Approved  D          Denied D



        	                Approved  D          Denied CD



        	                Approved  D          Denied d



        	                Approved  CH          Denied U


7.      How many full-time equivalent employees did the facility have in 1995?                   	
        NOTE: If there were less than 10 fall-time equivalent employees in 1995, do not visit this facility.

        Terminate discussion with facility at this time.
CEB.TRIM030.nh                                    A-2
-------
                                                               FacilitylD:  |_|J_|-|_LL|-|_LI_I

8.      Briefly describe the industrial processes performed at this facility in 1995.
9.      Has the facility's process operations significantly changed since 1995 (including equipment, chemicals,
        feedstock, etc.)?

        YES	D                     NO....D   (SkiptoQ.n)

10.     Briefly describe any process changes.
11.     Has the facility implemented any new treatment, disposal, energy recovery, recycling or source reduction
        activities since 1995?

        YES	D                      NO	D   (SkiptoQ.13)

12.     Briefly describe any new treatment, disposal, energy recovery, recycling or source reduction activities.
LOGISTICS

13.     Will the facility be operating under typical conditions at the time of the visit?

        YES....D              NO	D
CEB.TRI\1030.nh                                    A-3
-------
                                                               Facility ID:  |_|_|_|

14.      What personal protective equipment will be needed to participate in a facility tour?

                       Hard Hat               D
Safety Boots
Safety Glasses
Respirator
Other:
D
a
a

15.      How long is a typical tour? If unknown, how many square feet does the facility occupy? 	
        (Consider this information when planning the type and duration of tour that would be most useful).

16.      Hotel recommendation: 	
17.      Directions to facility:
18.      Time to meet:
19.1     Is a confidentiality agreement required to be completed for this facility?

        YES...D        NO....D   (Skip to Q. 20)

19.2     Has a confidentiality agreement been completed?

        YES...D        NO....D

20.      Will the person who completed the Form R and all supporting materials be available during the site visit?

        YES...D        NO....D        Alternate Contact:  	
CEB.TRI\1030.nh                                    A-4
-------
                                                               Facility ID:  |_|_|_|-|_|_|




 21.     Describe the type and quantity of supporting material available for the Form R calculations.
CEB.TRA1030.nh                                     A-5
-------
                                                             Facility ID: |_|_|_|-'
                                           SECTION 1.0
                                     REPORT PREPARATION
1.1     Who prepared the release estimates in the facility's Form R chemical reports? (Check all that apply)

       Facility Environmental Staff	LJ

       Corporate Environmental Staff	LJ

       Facility Staff	D

       Consultant/Contractor	LJ

       Safety Department Staff	D

       Other, specify	LJ

1.2     Check all EPA documents and other references used to estimate releases and control efficiencies.

                       NONE	D

       TRI Reporting Form R and Instructions, 1995 Version
       (EPA 745-K-95-051)	D

       Estimating Releases and Waste Treatment Efficiencies for the TRI ("Green Book")
               EPA 560/4-88-002	D

       Title III Section 313 Release Reporting Guidance
               EPA/560-4-88-004 a through q, Estimating Chemical Releases	D

       Compilation of Air Pollution Emission Factors, AP-42	I—I

       Industry Trade Association Materials/Seminars	I—I

       Privately Sponsored Seminar Materials	LJ

       EPA-Sponsored Training Workshops  	LJ

       MSDSs  	D
CEB.TRI\1030.nh                                    A-6
-------
                                                               FacilitylD:  |_|_|_|-|_|_|_|-|_|_|_|

        Computer Programs (list)	O



        Other	D
 1.3     What is your estimate of the time needed to fulfill the reporting requirements of Section 313 for 1995?
        Please include familiarization with the regulation and reporting instructions, completion and internal
        review of the reporting forms, and documentation of all information in your reports.

                <, 8 Hours	D


                9 - 20 Hours	D


                21 - 40 Hours	D


                41 - 100 Hours	D


                > 100 Hours	D

 1.4     Did you find the 1995 Form R reporting instructions useful?

                YES	D      NO....D

 1.5     Did you feel any section of the instructions provided with the Form R were unclear?

                YES....D      NO	D (Go to Q. 1.6)    NA....D (Skip to Q.1.6)

 1.5.1    Check the appropriate section below and briefly explain the difficulty encountered.

                Facility Reporting Determination	d




                Part I. Facility Identification Information	CD
CEB.TRI\1030.nh                                    A-7
-------
                                                              FacilitylD:  L|_L|-|_|_|_|-|_|_|_|

                Part II. Chemical Specific Information (Circle number of all that apply)	EH
                               1.   Toxic Chemical Identity
                               2.   Mixture Component Identity
                               3.   Activities and Uses of the Toxic Chemical
                               4.   Maximum Amount On-Site
                               5.   Releases to the  Environment On-Site
                               6.   Transfers in Waste to Off-Site Locations
                               7.   On-Site Waste Treatment Methods and Efficiency and On-Site Energy
                                   Recovery and Recycling Methods
                               8.   Source Reduction and Recycling Activities

1.6             Did you call the Emergency Planning and Community Right-to-Know Hotline?

                       YES....D               NO....D    (SkiptoQ.1.7)

1.6.1            Did you find the operator's response helpful?

                       YES....D               NO....D

                If no, explain	

1.7             Have you ever received any assistance from EPA Regional or headquarters staff to prepare the
                Form R reports?

                       YES....D               NO....D

1.8             Has EPA or your state ever contacted you with questions about any of the reported estimates
                (excluding computer generated notices)?

                       YES....D               NO....D

1.9             Has the facility received any Notices of Significant Error, Notices of Noncompliance, or Notices
                of Technical Error from EPA or the state for any 1995 reports?

                       YES....D               NO....D

1.10            Does the facility use any computer software to track toxic chemicals brought on site, used, or
                identified in MSDSs?

                       YES....D               NO....D

                If yes, identify: 	
CEB.TRI\1030.nh                                    A-8
-------
                                                               FacilitylD: |_|_L|-|_LI_|-|.

 1.11            Did you use the Automated Form R (APR) electronic reporting to submit your Form Rs?


                       YES....D               NO....D    (Skip to Q.I. 12)

 1.11.1           Did you feel the AFR helped to reduce any errors on the Form R?


                       YES....D               NO....D

 1.11.2           Describe any comments on the use of the AFR.
1.12            Are there additional guidance manuals that EPA should develop to provide more clarification on
                Form R reporting?


                        YES....D              NO....D
1-13            If a Form R was completed and the total annual reportable amount was less than 500 pounds,
                why did the facility not complete the short form for the alternate threshold reporting?
CEB.TR]\1030.nh                                    A-9
-------
                                                      FacilitylD:  |_|_|_|-|_|J_|-|J_|_|
                                      SECTION 2.0
                          INTRODUCTION AND FACILITY TOUR
                           (313 CHEMICALS PRESENT ON-SITE)
2.1           List all chemicals reported on the facility's Form R Chemical Reports.
        Chemical Name                    CAS #                Not a Section 313 Chemical
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
                                                                       D
CEB.TRI\1030.nh                              A-10
-------
                                                            FacilitylD:  |_|_|_|.|_|_|_|-|_|_|J
2.2            List all Section 313 chemicals not reported on the facility's Form R chemical reports, but
               documented by the facility.

                       NONE....D

       Chemical Name                             CAS #                         Not a Section
                                                                                313 Chemical
                                                                                    D
                                                                                    D
                                                                                    D
                                                                                    D
                                                                                    D
                                                                                    D
                                                                                    D
2.3            List ALL other Section 313 chemicals not reported or documented, but identified by the
               surveyor during the site visit.

                      NONE....D

         Chemical Name                                          CAS #
CEB.TRlM030.nh                                  A-11
-------
2.4
                                                FacilitylD:  |_|_|_|-|_LI_|-|_|_|

List all mixtures identified during the facility tour which may contain Section 313 chemicals.
(a)
Mixture Name























(b)
Identify Section
313 Chemical
Present























(c)
Concentration of
Chemical1























(d)
Amount of Mixture
Used in 19952























(e)
Amount of Section
3 13 Chemical
Used2























'If concentration of chemical is below de minimis (0.1 wt.% for carcinogens, 1.0 wt.% for all others), do not
include mixture in threshold determination.
2 Complete columns d and e during threshold determination.
CEB.rRI\1030.nh
                                 A-12
-------
                                                            Facility ID:
 Process Diagram(s):
                             (identify release points and chemicals)
CEB.TRI\1030.nh                                   A-13
-------
                                                             Facility ID:
 Treatment Unit, Disposal, Energy Recovery, Recycling or Source Reduction Operations):
                              (identify release points and chemicals)
CEB.TRI\1030.nh                                   A-14
-------
                                                                FacilitylD:   _U-H-LI-H---!
  Facility Tour Notes:
CEB.TIU\1030.nh                                     A-15
-------
                                                                    Facility ID:
  Facility Tour Notes (Cont'd):
CEB.TRI\1030.nh                                       A-16
-------
Chemical Name: 	                             Facility ID:

CAS#|  I  III J-LLI-U
                                          SECTION 3.0
                          REVIEW OF THRESHOLD DETERMINATION
3.1    How is this chemical employed at the facility? (Check all that apply)

        Facility  Reviewer                                                        Facility Reviewer
3.1.1    D      D        Manufacture


                           a.     Produced at the facility                              LJ     LJ


                           b.     Imported by the facility                              LJ     LJ


                           c.     For on-site use/processing                            LJ     CD


                           d.     For sale/distribution                                 D     d


                           e.     By-product                                        D     D


                           f.     Impurity1 (% =	)                               D     D



3.1.2    LJ      LJ        Process (incorporative activity)


                           a.     Chemical reactant (raw materials,
                                 intermediates, etc.)                                  LJ     LJ


                           b.     Formulation component                              LJ     LJ


                           c.     Article component                                  LD     LJ


                           d.     Repackaging                                       LJ     LJ
'If impurity is present below de minimis concentrations (0.1% for carcinogens, 1% for others), it is exempt from
reporting.


CEB.TRI\1030.nh                                  A-17
-------
Chemical Name: 	                               Facility ID:
         Facility  Reviewer                                                            Facility Reviewer
3.1.3     CD       CU        Otherwise Use (nonincorporative activity)

                             a.     Chemical processing aid (added to reaction
                                   mixture)                                              CD      CU

                             b.     Manufacturing aid (process lubricants,
                                   coolants, etc.)                                         CD      CD

                             c.     Ancillary use (cleaners, degreasers, lubricants)           CD      CD

3.1.4     D       D        Exempt Uses

                             a.     Used in laboratory activities                            CD      CD

                             b.     Structural component                                  CD      CD

                             c.     Routine janitorial/facility grounds maintenance           CD      CD

                             d.     Personal employee use                                 CD      CD

                             e.     Motor vehicle maintenance                             CD      CD

                             f.     Intake water component                               CD      CD

                             g.     Contained in an article                                 CD      CD

3.2             Was the chemical reported by the facility?

                YES....D (GotoQ.3.3)   NO....CD

3.2.1            If no, why did the facility decide this chemical was not reportable?

                a.       Below threshold	CD

                b.       Exempt	CD

                c.       Overlooked chemical	LD

                d.       Other (specify)	CD


CEB.TRI\1030.nh                                    A-18
-------
Chemical Name: 	                               Facility ID:
CEB.TRI\1030jih                                   A-19
-------
Chemical Name: 	                               Facility ID:
3.3             Are all uses of the chemical exempt from reporting according to the surveyor or are all uses of
                the chemical a non-aerosol form of sulfuric acid or hydrochloric acid?

                YES....D (GotoQ.3.10)   NO....D (Continue)

3.4             Does documentation which supports the threshold determination exist?
                (Documentation is defined as any type of data available at the facility in any form which
                can be used to recalculate the estimate)

                YES....D      NO....D

3.4.1            If no, why not?

                a.      Documentation cannot be located	CD


                b.      Documentation was not retained by facility	CD


                c.      Facility unaware that documentation is required	CD


                d.      Facility overlooked the chemical (Skip to Q.3.6)	CD


                e.      Other (specify) 		CD

3.5             What was the basis of estimate used by the facility for the amount manufactured, processed, or
                otherwise used in 1995? Check all that apply.
                a.       Purchase/inventory records	CJ


                b.       Emission Factors 	CD


                c.       Mass balance	CD


                d.       Assumed threshold exceeded (no calculations completed)  	LJ


                e.       Process recipes	CD


                f.       Monitoring data	CJ


                g.       Production data  	CD


                h.       Other (specify)	CD



CEB.TRI\1030.nh                                    A-20
-------
Chemical Name: 	                               Facility ID:
CEB.TRI\1030.nh                                    A-21
-------
Chemical Name:  	                                Facility ID:  |_|_|_|-|_|_|_|-|_|_|_|


3.6             How much chemical did the facility manufacture, process, or otherwise use in 1995?2

                                                Facility                          Reviewer

                a.      Manufactured           	Ibs                 	Ibs

                b.      Processed               	Ibs                 	Ibs

                c.      Otherwise used          	Ibs                 	Ibs

                d.      Facility did not estimate these quantities  	LJ

3.7             Was the reviewer's estimate of the amount of chemical manufactured, processed, or otherwise
                used recalculated using available documentation or recreated using other facility data?

                a.      Recalculated, with no error	U


                b.      Recalculated, within a factor of 2	L_l


                c.      Recalculated, within a factor of 10	LI


                d.      Recalculated, greater than a factor of 10	EH


                e.      Recreated, with no error	CH


                f.      Recreated, within a factor of 2	I—I


                g.      Recreated, within a factor of 10	I—I

                h.      Recreated, greater than a factor of 10	I—I


                i.      Facility did not estimate these quantities  	I—I


3.8             Was a threshold exceeded for this chemical in 1995?

                YES....CD  (This chemical should have been reported. Continue)


                NO	LJ  (This chemical should not have been reported. Skip to Q 3.10)
2Record calculations and assumptions for the threshold determination on the worksheet in Section 6.0.


CEB.TRI\1030.nh                                     A-22
-------
Chemical Name: 	                               Facility ID:
3r9	What was the maximum amount of this chemical on-site at any time?*

	Pacttity	
	Reviewer
	a. Maximum on site	             Ibs	             Ibs


3.10            This chemical was:

                a. Correctly reported	  (Go to Section 4.0) CD


                b. Correctly not reported	 (Skip to next chemical) D


                c. Incorrectly reported	(Go to Q.3.11) CD


                d. Incorrectly not reported	(Go to Q.3.12)CH


3.11            Why was this chemical incorrectly reported?

                a.       Facility reported, although amount used was below threshold  	CD


                b.       Facility incorrectly assumed threshold was exceeded	D


                c.       Chemical activity was misclassified	LJ


                d.       Threshold quantity was miscalculated  	LJ


                e.       Chemical was exempt	CD


                f.       Chemical has been delisted/modified	CD


                g.       Other (specify)	 CD


                                                                                     (Skip to next chemical)
3Record calculations and assumptions for the maximum on-site quantity on the worksheet in Section 6.0.


CEB.TRlM030.nh                                   A-23
-------
Chemical Name: 	                               Facility ID:  |_|_|_|-|_|_|_|-|_|_|_|






3.12            Why was this chemical incorrectly not reported?





                a.       Chemical activity was overlooked	Q





                b.       Chemical activity was misclassified	LJ





                c.       Threshold quantity was miscalculated  	LJ





                d.       Other (specify)	D






                                                                                    (Continue to Section 4.0)




3.13            If the facility completed a short form for this chemical, are the releases less than 500 pounds?





                YES....UJ   (Skip to the next chemical and document the release calculations)





                NO	D  (Go to Section 4.0)
CEB.TRI\1030.nh                                    A-24
-------
                                              FacilitylD:  _-----_----
                               SECTION 4.0




                       REVIEW OF RELEASE TYPES
CEB.TRI\1030.nh                          A-25
-------
                                                                                                           FacilitylD:  |_|_|_|-|_|_|_|-|_|_|_|

                                         Section 4.1  Sources of Chemical Releases and Transfers

In the reviewers opinion, document whether or not the facility should have included releases from the following sources (1):
Chemical Name
cAs#LLLL_l-LLI-LI
SOURCE

B. Pumps/valves/flanges
C. Volatilization from process areas
D. Volatilization from treatment areas
E. Storage tank/stock pile losses
F. Accidental spills/releases
G. Waste treatment discharge streams
H. Storm water runoff
I. Process discharge streams
J. Housekeeping practices/clean-up wastes (i.e., solvent)
K. Container residue
L. Treatment sludges, recycling or energy recovery by-products
M. Combustion by-products
N. Other
Fugitive Air














Stack Air














Receiving
Stream














Under-
Ground
Injection














Land On-
Site














POTW














Off-Site
Transfer














NOTE:  COMPLETE ALL ROWS AND COLUMNS.
If a Form R was completed:
(1)   Y   =    Yes, release source should be included in release estimate that surveyor
              calculates in Section 5 and facility identified this release type.
     N   =    No, release should be included in release estimate but facility overlooked
              this release type.
     NA  =    This source was not present at the facility for this chemical
If a Form R was not completed (overlooked chemical):
     Y   =    Release source should be included
     NA  =    Release source was not present at the facility for this chemical
        CEB.TRI\1030.nh
                                                                          A-26
-------
                                                     SECTION 4.2 (a) and (b)
                                                             CODE LIST
Qi
Q2
Q3
              Y          =   Yes
              N          =   No
              NA         =   Facility overlooked this chemical
              FE         =   Fugitive air
              PS         =   Stack air
              RS         =   Receiving stream
              UI         =   Underground injection
              LA         =   Land on site
              PW         =   POTW
              TOSD      =   Off-site transfer (disposal)
              TOST      =   Off-site transfer (treatment)
              TOSR      =   Off-site transfer (recycling)
              TOSE      =   Off-site transfer (energy recovery)
              NA         =   Facility does not have a release to this
                              medium
              Y          =   Yes
              N          =   No
              NA         =   Facility overlooked this chemical
Q4
              FES       =   Releases are only fugitive releases and
                              are not released to a stack.
              SFE       =   Releases are to a stack and not released
                              as fugitives.
              VPC       =   Chemical is a volatile organic chemical
                              (VOC) and was not reported as an air
                              release.
              ACID      =   Mineral acids, which were neutralized,
                              were included.
              POTW     =   Wastewater discharge is to a POTW
                              and not a receiving stream.
              RECS      =   Wastewater discharge is to a receiving
                              stream and not to a POTW.
              ONLAND  =   Releases are to an on-site landfill, not
                              to an off-site landfill.
              OFFLAND =   Releases are to an off-site landfill, not
                              to an on-site landfill.
              NOER     =   Off-site energy recovery  does not take
                              place in a legitimate energy recovery
                              system.
              NOCOMB  =   Toxic chemical does not  have a heating
                              value high enough to sustain
                              combustion.
              NR        =   Site visit concluded that chemical is not
                              released to this medium.
              OTH       =   Other	

              NA        =
                                                                                                       Facility correctly identified release type
                                                                                                       or facility overlooked chemical
CEB.TRI\I030.nh
                                                                      A-27
-------
                                                             CODE LIST (Continued)
        Q5
        Q7
 Y
 Yl
 Y2
 Nl
 N2
 N3
 N4
 N5
 N6
 NA

^f—

-*f2-
-NA-

 1
 2
 3
 4
 NA
              -N-
Yes
Yes, but facility incorrectly identified release type
Yes, but documentation is unclear or incomplete
Documentation cannot be located
Documentation was not retained by facility
Facility unaware that documentation required
Facility overlooked chemical
Facility overlooked this release type
Other	
Facility does not have a release for this medium
Facility unable to locate data
Facility did not retain data
Monitoring data not used

Facility derived factors
EPA published emission factors
Trade association factors
Other	
Emission factors not used
                          -Yes
\£7 1 ~
Ml _
JN1 "
NO «E

MA m

Y
N_
MA _


Ol 1 Y ~

N_
~

MA D
rJA ™
Q12 Y
N
NA =
Q13 MP =
CW =
WTS =
SC =
SPS =
OTH =
1CS
r1 •!•* • • l j ... *
raciiity misiiucrpretca ac minimis rule
OfVuii-
\JUIGI
r> •!•* J * u J ... » i_


XT f -ft. 1 1 J *
No, facility overlooked treatment
No on-site treatment of this chemical for this release
medium occurred
V/10
ICS
XT
iNO
•XT t A £C * * A J
INO treatment efficiencies were not used
Yes
No
Facility does not have a release for this medium
Spent metal plating bath
Cleaning waste
Waste treatment sludge
Spent catalyst
Spent process solvent
Other:

                                                                                         NA   =
                                                                                      Facility does not have recycling releases
Note: This code list refers to the questions for the Section 4.2(a) and (b) table on page 27 and 28.
        CEB.TRr\1030.nh
                                                                           A-28
-------
                                                                                                          FacilitylD: |_|_|_|-|_|JJ-|_|_|_|
                                     Section 4.2a  Review of Release Types (On-Site Releases)
Chemical Name
CAS#| | | | | |-| | |-| |
1. Did the facility identify a release type on the Form R?
2. Enter surveyor's release types.
3. Did the facility correctly identify the release type?
4. If Q.3 is NO, identify the reason that the release type was incorrectly
identified, otherwise enter NA.
5. Is documentation on the facility's release estimate available for review?
IF Q.5 IS NO OR NA, SKIP TO QUESTION 12
6. If monitoring data were used, is it available for review?
7. If emission factors were used, what is the source of the factors?
8. Was each air or waste stream counted only once in release estimates? (1)
9. Were all air or waste streams containing > 1% or aO.1% (carcinogens) of
the chemical included in release calculations?
10. Was on-site treatment of this chemical included in release estimates?
1 1 . Were treatment efficiencies reported consistent with measurement data,
vendor specs, or EPA-published efficiencies? (2)
12. Does the facility have information available to estimate the amount of this
chemical released during 1 995?
Fugitive Air
§5.1













Stack Air
§5.2













Receiving
Stream
§5.3













Underground
Injection
§5.4













Land On Site
§5.5













(+Hf no, document all streams double counted in release calculations in Section 6.0
(2^ If no, document inconsistency of treatment efficiencies used in Section 6.0
       CEB.TRI\1030.nh
                                                                    A-29
-------
                                                                                                              Facility ID:
                                            Section 4.2b Review of Release Types (Off-Site)
Chemical Name
CAS# _LJJ_-LLI-LI
1 . Did the facility identify a release type on the Form R?
2. Enter surveyor's release types.
3. Did the facility correctly identify the release type?
4. If Q.3 is NO, identify the reason that the release type was incorrectly
identified, otherwise enter NA.
5. Is documentation on the facility's release estimate available for review?
IF Q.5 IS NO OR NA, SKIP TO QUESTION 12
6. If monitoring data were used, is it available for review?
7. If emission factors were used, what is the source of the factors?
8. Was each air or waste stream counted only once in release estimates? (1)
9. Were all air or waste streams containing k 1% or >0.1% (carcinogens) of
the chemical included in release calculations?
10. Was on-site treatment of this chemical included in release estimates?
1 1 . Were treatment efficiencies reported consistent with measurement data,
vendor specs, or EPA-published efficiencies? (2)
12. Does the facility have information available to estimate the amount of this
chemical released during 1995?
13. If appropriate, characterize the recycling stream (use multiple codes if
necessary).
POTW
§6.1














Off-Site
Transfer
(disposal)
§6.2














Off-Site
Transfer
(treatment)
§6.2














Off-Site
Transfer
(recycling)
§6.2














Off-Site
Transfer
(energy
recovery)
§6.2














(1) If no, document all streams double counted in release calculations in Section 6.0
(2) If no, document inconsistency of treatment efficiencies used in Section 6.0
       CEB.TRI\1030.nh
                                                                       A-30
-------
                                                                                                       Facility ID:  |_|_|_|-|_|_|_|

           Section 4.2c  Review of Release Types (On-Site Treatment, Energy Recovery or Recycling)
Chemical Name
CAS#LLLLU-LIJ-LI
1 . Did the facility identify an on-site treatment, energy recovery, or recycling method in §7 on the Form
R?(l)
2. Enter surveyor's identification of on-site methods. (2)
3. Did the facility correctly identify the on-site method? (1)
4. If Q.3 is NO, identify the reason that the method was incorrectly identified, otherwise enter NA. (3)
5. For on-site treatment in §8.6B, did the facility only report the quantity of chemical destroyed during
treatment? (4)
6. For on-site recycling in §8.4B, did the facility report the quantity of chemical recovered from
recycling? (4)
7. If appropriate, characterize the recycling stream (use multiple codes if necessary): (5)
8. Describe the type of recycling unit: (6)
On-Site
Treatment
(§7A or 8.6B)








On-Site
Energy
Recovery
(§7B or 8.2B)








On-Site
Recycling
(§7Cor
8.4B)








(1)    Y          =  Yes
      N          =  No
      NA         =  Facility overlooked this chemical

(2)    TR         =  On-site treatment
      ER         =  On-site energy recovery
      REC        =  On-site recycling
      NA         =  Facility does not use this on-site method

(3)    OFFLAND  =  Releases are to an off-site landfill, not an on-site
                     landfill
      NOER      =  Off-site energy recovery does not take place in a
                     legitimate energy recovery system
      NOCOMB   =  Toxic chemical does not have a heating value high
                     enough to sustain combustion
      NR         =  Toxic chemical is not recycled
      OTH        =  Other: 	
      NA         =  Facility correctly identified on-site method or facility
                     overlooked chemical.
(4)    Y          =  Yes
      N          =  No
      NA         =  Facility did not identify this on-site method

(5)    MP         =  Spent metal plating bath
      CW         =  Cleaning waste
      WTS        =  Waste treatment sludge
      SC         =  Spent catalyst
      SPS         =  Spent process solvent
      OTH        =  Other: 	
      NA         =  Facility did not estimate recycling releases

(6)    Identify type of on-site recycling unit used. See §7.C of
      FormR.
CEB.TRI\1030.nh
                                                                A-31
-------
                                              Facility ID:
                               SECTION 5.0




                     REVIEW OF RELEASE ESTIMATES
CEB.TRI\1030.nh                          A-32
-------
                                                                                                                FacilitylD:  |_|_|J-|_|_|J-|_|_|_|
                                      Section 5.1  Review of Release Estimates (On-Site Releases)

For each on-site release identified in Section 4.2a (Question 2), complete the following table:
Chemical Name
CAS#|_| | | | |-| | |-| |

1 . Enter facility's release estimate (in Ibs) (1)
2. What method(s) did the facility use to estimate their release? (2)
3. Based on data available to the facility, is this the most accurate
method to determine a release estimate? (3)
IF Q.3 IS YES, SKIP TO QUESTION 6
4. What is a better method(s) which could be used to calculate a more
accurate release estimate? (2) (4)

5. Enter the reviewer's release estimate using a more accurate
method(s) (5)

6. Enter the reviewer's release estimate using the same method(s) as the
facility. (5)

Fugitive Air
§5.1

ABC





ABC


ABC

Stack Air
§5.2

ABC





ABC


ABC
Receiving
Stream
§5J

ABC





ABC


ABC
Underground
Injection
§5.4

ABC





ABC


ABC

Land On Site
§5.5

ABC





ABC


ABC
(1)  Range Codes:
    A   =  1-10 Ibs
    B   =  11-499 Ibs
    C   =  500-999 Ibs
    Nl  =  Release estimate was not
           included on Form R but should
           have been, skip to Question 4
    N2  =  Facility overlooked this chemical,
           skip to Question 4
    NA =  Facility does not have a
           release to this medium, do not
           continue with this medium
N3 =  Release estimate was included but should not   (3)
      have been, do not continue with this medium
      but enter facility release (i.e, N3,100)
                                                 (4)
(2) M   = Monitoring data or direct measurements
   C    = Mass balance calculations
   E    = Published emission factors
   OC  = Engineering calculations ("minor calcs")
   OJ   = Engineering judgement ("guess")
   OH  = Hazardous waste manifests
   O    = Other	
   NA  = Facility did not estimate release
       Y =Yes
       N =No

       Document why this method is more
       accurate in Section 6.0
(5)
       NA = Facility did not estimate release
       [Note: Enter the number that was calculated. Only enter
       a range, if a range is the most accurate quantity that can
       be calculated.]
Document release calculations in Section 6.0
       CEB.TRI\1030.nh
                                                                        A-33
-------
                                                                                                                    Facility ID:
                                             Section 5.2  Review of Release Estimates (Off-Site)

For each off-site release identified in Section 4.2b (Question 2), complete the following table:
Chemical Name
cAs#LL_LLI- -LI-LI
1. Enter facility's release estimate (in Ibs) (1)
2. What method(s) did the facility use to estimate their release? (2)
3. Based on data available to the facility, is this the most accurate
method to determine a release estimate? (3)
IF Q.3 IS YES, SKIP TO QUESTION 6
4. What is a better method(s) which could be used to calculate a more
accurate release estimate? (2) (4)

5. Enter the reviewer's release estimate using a more accurate
method(s) (5)

6. Enter the reviewer's release estimate using the same method(s) as the
facility. (5)
POTW
§6.1
ABC




ABC
ABC
Off-Site
Transfer
(disposal)
§6.2
ABC




ABC
ABC
Off-Site
Transfer
(treatment)
§6.2
ABC




ABC
ABC
Off-Site
Transfer
(recycling)
§6.2
ABC




ABC
ABC
Off-Site Transfer
(energy recovery)
§6.2
ABC




ABC
ABC
(1)  Range Codes:                          (2) M
    A  =  1-10 Ibs                            C
    B  =  11-499 Ibs                          E
    C  =  500-999 Ibs                         OC
    Nl  =  Release estimate was not             OJ
           included on Form R but should        OH
           have been, skip to Question 4         O
    N2  =  Facility overlooked this chemical,     NA
           skip to Question 4
    NA =  Facility does not have a
           release to this medium, do not
           continue with this medium
= Monitoring data or direct measurements
= Mass balance calculations
= Published emission factors
= Engineering calculations ("minor calcs")
= Engineering judgement ("guess")
= Hazardous waste manifests
= Other	                       (5)
= Facility did not estimate release
(3)     Y = Yes
        N =No

(4)     Document why this method is more
        accurate in Section 6.0

NA = Facility did not estimate release
        [Note:  Enter the number that was calculated. Only enter
        a range, if a range is the most accurate quantity that can
        be calculated. 1
        Document release calculations in Section 6.0
        CEB.TRl\!030.nh
                                                                           A-34
-------
                                                                                                               Facility ID:
                                                  Section 5.3  Review of Form R §8 Data
                                                  (On-Site Releases or Off-Site Transfers)
Chemical Name
cAs#LLLLLI-LLI-LI
1. Enter facility's estimate from §8, Column B, on
the Form R. {Enter NA if facility did not estimate)
2. Enter facility's basis of estimate. (1)
3. Calculate the quantity released or transferred
using the method in footnote (2).
4. Are the facility's estimate (Q.I) and the quantity
released or transferred from Q.3 the same? (3)
5. If Q.4 is NO, provide notes or an explanation
detailing any differences in the calculation of
Section 8 data.
Quantity Released
§8.16





Quantity Used for Energy
Recovery Off Site
§8JB





Quantity Recycled Off Site
§8.5B





Quantity Treated Off Site
§8.7B





(1)  TECH     =   Used the following technique:

      - Form R §8.IB = [§5.1 + §5.2 + §5.3 + §5.4 + §5.5 + §6.2 (disposal only) - §8.8]
      - Form R §8.3B = §6.2 (energy recovery only) - §8.8
      - Form R §8.5B = §6.2 (recycled only) - §8.8
      - Form R §8.7B = §6.1 + §6.2 (treated only) - §8.8
                                                                   (3)  Y
                                                                       N
                                                                   NA
       Yes
       No
Facility did not estimate
    NOBASE  =
    OTH
     Data for Section 8 was estimated, basis not provided
     Other:	
    NA
=    Facility did not estimate; do not continue with this medium.
(2)  [Note: Use the best release estimate from Section 5.1 and 5.2 of this survey
    to calculate these quantities.]
    Document the calculations in Section 6.0.
       CEB.TRI\1030.nh
                                                                        A-35
-------
                                                                          FacilitylD: |__|_|- _U-|-|-U-I
                                Section 5.4 Review of Form R §8 Data
                       (On-Site Treatment, Energy Recovery, or Recycling)
For the on-site treatment or energy recovery method(s) identified in Section 4.2c (Question 2), complete the following table.
Only recreate on-site recycling estimates that were provided by the facility. Do not estimate on-site recycling releases NOT
identified by the facility.
Chemical Name
CAS#|_J_U_-I_U-U
1. Enter the facility's estimate of quantity from §8,
Column B, of the Form R. (1)
2. What method did the facility use to estimate the
amount treated, sent to energy recovery, or recovered from
recycling? (2)
3. Based on data available to the facility, is this the most
accurate method to estimate the amount treated, sent to
energy recovery, or recovered from recycling? (3)
IF Q.3 IS YES, SKIP TO QUESTION 6
4. What is a better method which could be used to
calculate a more accurate estimate? (2)
5. Enter the reviewer's estimate using a more accurate
method. (4)
6. Enter the reviewer's estimate using the same method.
(4)
On-Site Treatment
(§7Aor8.6B)







On-Site Energy
Recovery
(§7B or 8.2B)







On-Site Recycling
(§7C or 8.4B)







(1)
(2)
           Estimate was not included on Form R but should have been, skip to Question 4.
           Facility overlooked this chemical, skip to Question 4.
           Facility does not have this on-site method, do not continue with this medium.

           Monitoring data or direct measurements
           Mass balance calculations
           Published emission factors
           Engineering calculations
           Engineering judgement
           Hazardous waste manifests
           Other  	___^___
(3)
(4)
Y      =   Yes
N      =   No
NA    =   Facility does not have this on-site method

Document calculations in Section 6.0.
NA = Facility did not estimate release.
        CEB.TRI\1030.nh
                                                      A-36
-------
                                                                      FacilitylD:  |_|_|_|-|_L|_|-L|_|_|
                              Section 5.5  Review of Form R §8 Data
              (Production Ratio/Activity Index and Source Reduction Activities)
Chemical Name
cAs#U_--U-__l-U
Production Ratio/Activity Index
1. Enter the facility's estimate from §8.9
of the Form R. (Enter NO if facility did
not estimate)
2. Enter facility's basis of estimate (1).
3a. Is this estimate based on a variable
that most directly affects the quantities of
the toxic chemical generated as "waste"
quantities? (2)
3b. If Q.3a is NO, enter surveyor's choice
for alternate basis.
Source Reduction Activities
4a. Enter the source reduction activity
codes from Section 8.10 of the Form R.
4b. Provide a text description of the
source reduction activity.
4c. Is this activity "source reduction"
(i.e., not recycling, treatment, energy
recovery, or disposal) (2)?
Production Ratio/Activity Index (§8.9)
and Source Reduction Activities (§8.10)


3a.
3b.
Activity #1
4a-l.
4b-l.
4c-l.
Activity #2
4a-2.
4b-2.
4c-2.
Activity #3
4a-3.
4b-3.
4c-3.
(1)
   TCM
   TCPV
   TCU
   HR
   WT
   OTH

   NA
(2)
   Y
   N
Ratio of amount of the toxic chemical manufactured in 1995 to 1993
Ratio of production volume in 1995 to 1993
An activity index of the amount of toxic chemical used in 1995 to 1993
An activity index of the amount of operating hours for an activity in 1995 to 1993
An activity index or production ratio based on a weighted average of data from several processes
Other:	

The manufacture or use of the chemical began in 1995.
Yes
No
       CEB.TRI\1030.nh
                                 A-37
-------
                                              Facility ID:
                                SECTION 6.0




                       CALCULATION WORKSHEETS
CEB.TRI\1030.nh                           A-38
-------
                                                 Facility ID:
                 THRESHOLD DETERMINATION WORKSHEET
Chemical Name:

Description of Use










TOTALS
Amount
Manufactured











Amount Processed












Amount Otherwise
Used











Calculations:
CEB.TRI\1030.nh
A-39
-------
                                                         FacilitylD: UJ_H-U-|-|-I-U

                         MAXIMUM AMOUNT ONSITE WORKSHEET
Chemical Name:
INSTRUCTIONS:—Calculate the maximum amount of the chemical onsite at any one time during the
                   reporting year. Keep in mind the following:

	•	All storage areas where this chemical may be kept;

	•	The amount of chemical being used at any time; and

	•	The amount of chemical in each waste stream.
Storage Areas:
                                                                                  Total:
Chemical in Use:
                                                                                  Total:
Chemical in Waste Streams:
                                                                                  Total:
       CEB.TRI\1030.nh
                                              A-40
-------
                                                             FacilitylD: |_U_|-|_U_H_U-



                                                                                  Total On-Site:
CEB.TRI\1030.nh                                   A-41
-------
                                                        FacilitylD: |_L|_|-|_|_|_|-|_|_|_|

                             RELEASE ESTIMATE WORKSHEET
Chemical Name: 	             Release Type
CAS# _____-__-_               SIPage# 	      Question^
INSTRUCTIONS:   Record all calculations for release estimates below, in the appropriate sections.  Be sure
                   to identify if calculations use the same method as the facility or a more accurate method.
      CEB.TRI\1030.nh                             A-42
-------
Appendix B
-------
-------
Introduction

The TRI Site Visit study has two primary objectives. The first objective is to estimate the number
of facilities that have made an error in Form R filing. The second objective is to estimate the key
reasons for errors among those facilities with one or more filing errors.  Rather than sampling
from all facilities that have filed a Form R, the sample design focused on three major industry
groups ~ 1) Chemical manufacturing facilities, refined specifically to SIC codes 286 (Industrial
Organic Chemicals), 282 (Plastics Materials and Synthetic Resins, Synthetic Rubber, Cellulosic
and Other Manmade Fibers), 281 (Industrial Inorganic Chemicals), and 285 (Paints, Varnishes,
Lacquers, Enamels and Allied Products); 2) Rubber and Plastic Products which are covered by
SIC code 30; and 3) Furniture and Fixtures which are covered by SIC code 25. Facilities
reporting multiple SIC codes will be considered in scope only if all reported SIC codes fall within
one of the SIC code-defined industries selected for study. As discussed below, these three
industry groups were used as the primary stratifier of the facility sample.

The target sample size is 90 completed facility site visits.  This total sample size is allocated to the
three primary strata. Because the chemical group is the most important of the three groups, EPA
has decided to  have  an allocation of 40 site visits from this group, and 25 from each of the
remaining two  groups. While the sampling work has been conducted for all three groups, the
initial set of site visits are expected to be restricted to the chemical sector.
-------
-------
Sampling Frame

The sampling frame for the sample was constructed from the 1994 TRI data files (TRIPQUICK).
Those firms with more than fifteen Forms R were eliminated from the sampling frame at Radian's
request because the site visit to verify the Form R information would be too resource intensive.
For SIC codes 25 and 30 none of the facilities were eliminated for this reason, for SIC code 28 six
percent of the facilities were removed. The population size of facilities in each SIC code group (
after removal of facilities with more than fifteen Forms R) is:

Chemical               1274

Furniture                535

Rubber and Plastics     1872
Use of Two-Stage Cluster Sampling

It would be quite costly to draw a national stratified random sample of facilities, because the
facilities would be fairly well scattered across the U.S.  In other words, the cost traveling to each
of the selected facilities would be high.  The more cost-effective approach used here entailed
selecting a two-stage cluster sample of facilities.  The specifics of the sample selection are
outlined below.

The most readily available geographic identification variable in the TRI data file is the three-digit
Zip code sectional center of the facility. Three-digit Zip codes cover relatively small areas in
urban locations and considerably larger areas in rural locations.  Using the TRI data file
(eliminating facilities above the Form R threshold), it was possible  to sort all of the eligible
facilities by three-digit Zip code.  Two-stage cluster sampling involves the following steps:

1) Order all of the three-digit Zip codes in ascending Zip code order to introduce some implicit
geographic stratification.

2) Determine if any three-digit Zip codes are too small and must be combined with one or more
geographically adjacent Zip codes to form a cluster of sufficient size. This was done so that each
Zip code cluster included at least two facilities from each of the three SIC codes. Some areas of
the U.S. do not have any furniture manufacturing facilities. In order to avoid forming clusters
containing a large number of Zip codes, the rule was modified in these areas of the country to
allow for the formation of clusters with no furniture manufacturing facilities.  A total of 178
clusters were formed.

3) After creating the sampling frame of clusters, it was determined that the cluster consisting of
Zips 286-287 in North Carolina contained 61 (11.4%) of the 535 furniture manufacturing facilities
in the U.S. This cluster was included in the  sample with certainty since it contained more than 10
percent of the total industry facilities, and it was determined that 3 site visits with furniture'
-------
manufacturing facilities should take place in this cluster based on its size.

4) A measure of size was assigned to each of the remaining 177 clusters. The composite measure
of size, MOSj, for the i-th cluster was computed using the equation
where N^ is the number of facilities in the i-th cluster in SIC code group j, and Fj is the overall
desired sampling fraction for SIC code group j.

5) A systematic probability proportional to size (PPS) sample of 40 noncertainty clusters was
drawn from the sampling frame of 177 clusters.  The first step in the systematic PPS sample was
to cumulate the measure of size, MOSt over the 177 clusters. The total cumulative sum of MOS;
for the 177 clusters was then divided by 40, the sample size of clusters, to form a PPS selection
interval, K. A random start, R, between one and K was then generated. Forty selection numbers
were  computed as R, R+K, R+2K, R+3K,	,R+39K.  A cluster was selected if the selection
number fell into its sequence of numbers; that is, the selection number was greater than the
cumulative sum of all previous clusters, but less than or equal to the cumulative sum including the
designated cluster.

6) Overall, the sample consists of 41 clusters ~ the 40 noncertainty clusters plus the Zip 286-287
cluster in North Carolina.

Sampling Facilities Within Clusters

After drawing the first-stage sample of geographic clusters, the second-stage sample of facilities
was drawn.  For each of the 40 clusters, one chemical facility was selected as the primary
selection. A minimum of one reserve chemical facility was also drawn from each of the 40
clusters.  For rubber and plastics, it was first necessary to draw a random subsample of 25 of the
40 clusters.  For each of the 25 clusters, one primary rubber and plastics facility was randomly
drawn, and a minimum of one reserve rubber and plastics facility was also drawn. For furniture, it
was first necessary to draw a random subsample of 22 of the 40 clusters containing furniture
manufacturing facilities. For each of the 22 clusters, one primary furniture facility was randomly
drawn, and a minimum of one reserve furniture facility was also drawn. Recall that 3 primary
furniture facilities were also drawn from the cluster consisting of Zips 286-287 in North Carolina,
making a total of 25 primary selections.

Expected 95-Percent Confidence Intervals

The above design will yield about 40 completed site visits for the chemical group, and 25 from the
other two groups.  If we assume that 50% of facilities in the chemical group made a filing  error,
then the 90-percent confidence interval would be approximately plus or minus 13.0 percentage
-------
points for the chemical group. In other words, the chances are 9 out of 10 that the true
population percent of facilities with a filing error would be between 37% and 63%. Continuing
with this assumption, 20 of the 40 facilities in the chemical group would have a filing error.  For
estimates of key reasons for filing errors, the 90-percent confidence intervals could be as large as
plus or minus 18 percentage points.

For each of the other two groups, if we again assume that 50% of facilities in each of the two
groups made a filing error, then the 90-percent confidence interval would be approximately plus
or minus 16.5 percentage points for each group.  In other words, the chances are 9 out of 10 that
the true population percent of facilities with a filing error would be between 33.5% and 66.5%.
Continuing with this assumption, 12 or 13 of the 25 facilities in each of the two groups would
have a filing error.  Using this smaller number of facilities as the basis for estimates of key reasons
for filing errors, the 90-percent confidence intervals could be as  large as plus or minus 23.7
percentage points for each group.
-------
-------
Appendix C
-------
-------
Dear:

Your facility has been randomly selected by the U.S. Environmental Protection Agency
(EPA) to participate in a data quality survey of facilities that submitted 1994 reporting
year Form Rs for the Toxic Chemical Release Inventory (TRI) under Section 313 of the
Emergency Planning and Community Right-to-Know Act (EPCRA). The purpose of the
survey is to evaluate the quality of the data submitted on the Form R chemical reports and
to provide feedback to EPA that can be used to improve the reporting instructions,
guidance, or the reporting form. Eastern Research Group, Inc. (ERG) is working with
EPA to conduct survey visits to approximately 90 facilities randomly selected from the
1994 TRI database in the following SIC Codes: 25,281,282,285,286, and 30.

Participation in this program is voluntary.  However, we believe your participation will
provide a valuable evaluation of the threshold calculations and release and transfer
estimates for your facility, which will assist you in future Form R reporting.

ERG technical staff members will visit your facility at a date between April and May
1997. Most visits will last 1 to 2 days, depending on the number of chemical reports you
submitted and the number of Section 313 chemicals on site. Site surveyors will spend the
majority of that time reviewing the methodology and data that your facility used to make
threshold calculations and release and transfer estimates for the 1994 reporting year.
During the site visit, site surveyors will need your assistance to tour the plant in enough
depth to allow them to understand your manufacturing processes and to identify potential
release points. The site surveyors will make every effort to minimize disruptions to your
schedule  while they are on site.

You will be contacted by telephone within the next week to arrange a date for the site
visit. The name and location of your facility will not be released to EPA in order to ensure
confidentiality of your facility's information. All data collected on this project will be
tabulated and reported to EPA in summary form only. At the project conclusion, facility
identification information will be destroyed.
-------
-------
Appendix D
-------
-------
0

                                             DAt-rA
                     C  S€«l*»h»«iJ4"l  ^fo«nx*.i^l ^ C
                  s  7   .M^cr*naf C   /^ocoo^-c f   7 ,
                     L.            ^   *-    M*P*
-------
Tmi.ttM.oy>
   CovP*r4C€
          Of-
                                                      TO
-------
    A ^IXTU/LC d»K)rmi^iHr  1W
           OX.IPC  i* oseo -TO
i    10,000  VWs.
   I^TTM  1^,000 \b$.

        v\oS.
                    A-$ OPP»*FO TT> \
                    -.«.... .^  .-_ » «. . . ^> A t <
-------
-------
                                                  Facility ID:  | _ | _

                            RELEASE ESTIMATE WORKSHEET
Chemical Name:   \\i\-Lf\es	               Release Type   Pu.«y,' f/^g.   /r?r

CAS# |_|L|2|2|fi|-Ul2H2|               SIPage* 	      Question *
INSTRUCTIONS:  Record all calculations for release estimates below, in the appropriate sections.  Be
                  sure to identify if calculations use the same method as the facility or a more accurate
                  method.
                                 \ 1 2 OOO

                    t\
                                                 I — ^ Assume y //s 0
                   of
                                             yr
                                                          It
                                                            s.
      4 August 1997 - Revision 3
      CEB2\1108-02.ip                            H-39
-------
                                Facility ID:  |_L|_|-|_|_|

          RELEASE ESTIMATE WORKSHEET


f\\t\e^e^                     Release Type    S~ntc<
Chemical Name:       y  gflg-                     Release Type
CAS#  |_|1|2P£-S2Z               SIPage* _      Question #
INSTRUCTIONS:  Record all calculations for release estimates below, in the appropriate sections.  Be
                  sure to identify if calculations use the same method as the facility or a more accurate
                  method.
                                                                               yeir

            processed   « T-   mro   Xy/die, --c«?i\T-«fni^a

                                    Its

                                    Iks   p*-tf>Au-CT

                                    Ibs   -Puifivyt
       4 August 1997 - Revision 3
       CEB2M 108-02. ip                           H-39
                         ce     1/2000         «      y^n^ -^Oo     t  A a

            c^i    /A^»'   V
                                           een    c« ^»xi /«-f efe;  P^cilt^    e^/via^ecJ   'V^O |l,s.

                                                                             P"'

                                                                             J ;'

                    po->5» fc/e.   -fc?   re 00,1 ei le.   fl^e
-------
                                                 Facility ID: | _ |.


                            RELEASE ESTIMATE WORKSHEET
Chemical Name:   Le-*o-
CAS#  |!|H|i|2|_|-|-l---
                                          Release Type


                                          SIPage*
                                                                       Question #
INSTRUCTIONS:  Record all calculations for release estimates below, in the appropriate sections.  Be
                  sure to identify if calculations use the same method as the facility or a more accurate

                  method.
     O 1
                                 KVO
                              +
                                                    reuse

                             V
                               O.
                                                  /L-   I
                                                                                L-/yr.
O.
              ^V\  / aq, ooo cx?o L~\/\  k<\  \ /"
              ^v  C -  F: - Al^J I "
                                                                                  .   fo
 4 August 1997 - Revision 3
 CEB2M 108-02 ip
                                            H-39
-------