&EPA
EPA821-R-09-007
Technical Support Document for the
Annual Review of Existing Effluent
Guidelines and Identification of Potential
New Point Source Categories
U.S. Environmental Protection Agency
Engineering and Analysis Division
Office of Water
1200 Pennsylvania Avenue, NW
Washington, D.C. 20460
October 2009
-------
CONTENTS
Page
1. INTRODUCTION 1-1
1.1 Introduction References 1-2
2. DEVELOPMENT OF TRIRELEASES2007 2-1
2.1 TRI 2-1
2.1.1 Utility of TRI 2-3
2.1.2 Constraints and Limitations of TRI 2-3
2.2 Overview of TRI Databases 2-4
2.3 TRIRawData2007 2-5
2.4 TRICalculations2007 2-6
2.4.1 Modifications to TRI-Reported Data 2-8
2.4.2 POTW Removals 2-9
2.4.3 TWFs 2-9
2.4.4 Metal Compounds 2-22
2.4.5 Determination of "Basis of Estimate" of Reported TRI Releases 2-22
2.5 TRIReleases2007 2-23
2.5.1 NAICS/Point Source Category Crosswalk 2-24
2.5.2 Development of 2007 TRI Rankings 2-24
2.6 Results of the Preliminary Analysis of the TRIReleases200 7 Database 2-25
2.6.1 Metals Analysis 2-27
2.7 Development of TRIReleases2007 References 2-31
3. DMRLOADS2007: DEVELOPMENT AND CATEGORY RANKINGS 3-1
3.1 Overview of DMRLoads2007 3-1
3.1.1 NPDES Permitting and Reporting Requirements 3-1
3.1.2 Overview of PCS and ICIS-NPDES 3-3
3.1.3 PCS and ICIS-NPDES Data Structure 3-5
3.2 DMRLoads2007: Database Development and Methodology 3-7
3.2.1 Data Sources used in the Development of DMRLoads2007 3-7
3.2.2 PCSLoadCalculator2007 3-9
3.2.3 ICIS-NPDES Pollutant Loading Tool 3-25
3.2.4 DMRLoadsAnalysis2007 3-42
3.2.5 DMRNutrients2007 3-50
3.2.6 DMRLoads2007 3-51
3.2.7 Database Corrections 3-52
3.3 Results of the Preliminary Analysis 3-59
3.4 Data Quality Review 3-59
3.4.1 Mercury Discharges Reported Using PRAM 50092 3-65
3.4.2 Facility Reviews 3-65
3.5 DMRLoads2007 References 3-72
4. IDENTIFICATION OF POINT SOURCE CATEGORIES 4-1
4.1 Background on NAIC Sand SIC Codes 4-2
4.2 SIC Code to Point Source Category Crosswalk 4-3
i
-------
CONTENTS (Continued)
Page
4.2.1 SIC Codes Related to Existing Point Source Categories 4-4
4.3 NAICS Code to Point Source Category Crosswalk 4-10
4.3.1 NAICS Codes Related to Existing Point Source Categories 4-11
4.4 Potential New Point Source Categories 4-20
4.4.1 Direct Discharges 4-21
4.4.2 Indirect Discharges 4-21
4.5 Identification of Point Source Category References 4-21
5. Toxic WEIGHTING FACTORS 5-1
5.1 TWF Background and Development 5-1
5.2 New Toxic Weighting Factors Developed During the 2009 Annual Review 5-1
5.3 Chemicals without Toxic Weighting Factors 5-2
5.4 Toxic Weighting Factor References 5-8
6. QUALITY REVIEW 6-1
6.1 Overview of Quality Review Steps 6-1
6.1.1 Completeness Checks 6-2
6.1.2 Accuracy Checks 6-3
6.1.3 Reasonableness Checks 6-3
6.2 Quality Review of the DMRLoads200 7 Database 6-4
6.3 Quality Review of the TRIReleases2007 Database 6-5
6.4 Quality Review References 6-5
7. RESULTS OF 2009 SCREENING-LEVEL ANALYSIS 7-6
7.1 Preliminary Results of the Screening-Level Review 7-6
7.2 Prioritization of Categories 7-10
7.2.1 Categories for Which EPA is Currently Developing or Revising ELGs7-10
7.2.2 Categories for Which EPA Recently Promulgated or Revised ELGs ... 7-11
7.2.3 Discharges Not Categorizable 7-11
7.2.4 Categories with One Facility Dominating the TWPE 7-12
7.2.5 Combining the Final DMR and TRI Rankings 7-12
7.3 Identification of Categories With Existing Effluent Guidelines for Further Review
7-14
7.4 Results of 2009 Screening-Level Analysis References 7-18
11
-------
LIST OF TABLES
Page
2-1 TRI 2007 Tables Downloaded from EPA 2-5
2-2 Tables Imported or Created in TRICalculations2007 2-6
2-3 TWF Modifications 2-10
2-4 Dioxins Congeners and Their Toxic Weighting Factors 2-11
2-5 EPA Facility-Specific Dioxin Congeners Distribution Corrections 2-12
2-6 Chemical Composition of Creosote and TWF 2-13
2-7 Definition of Polycyclic Aromatic Compounds 2-14
2-8 PAC Concentrations in Petroleum Products 2-16
2-9 PAC Concentrations in Crude Oils (mg/kg) 2-17
2-10 Supply and Disposition of Crude Oil and Petroleum Products 2-18
2-11 Products for Which PAC Composition Is Not Available 2-19
2-12 Calculation of Toxic Weighting Factor for Petroleum PACs 2-19
2-13 Calculation of Toxic Weighting Factor for Wood Preserving PACs 2-20
2-14 Calculation of Toxic Weighting Factor for Pulp, Paper, and Paperboard PACs 2-22
2-15 Tables Created in TRIReleases2007 2-23
2-16 Point Source Category Rankings 2-25
2-17 Metals Discharged by TWPE in TRI 2007 2-28
2-18 Facilities Reporting Discharges of Metals with the Highest TWPE in TRI 2007 2-29
3-1 States and Territories Included in DMRLoads2007 3-3
3-2 Data Types in PCS and ICIS-NPDES Used for DMRLoads2007 Development 3-6
3-3 PCS CNVRT Module Output 3-10
3-4 Example Calculation for DL Option Loads 3-16
3-5 NODI Code Descriptions 3-16
3-6 PCSLoadCalculator2007 Output 3-19
iii
-------
LIST OF TABLES (Continued)
Page
3-7 Results of NODI Code Excluded from EST=YES Revision Analysis for PCS 3-23
3-8 Actual Number of Days per Monitoring Period 3-28
3-9 Example Type 1 Flow Correction 3-29
3-10 Example Type 2 Flow Correction 3-30
3-11 Measurement Value Selection Priorities and Calculations 3-36
3-12 Flow Value Selection Priorities 3-38
3-13 ICIS-NPDES Load Calculator Module Output 3-42
3-14 Tables Imported or Created in DMRLoadsAnalysis2007 3-43
3-15 TWF Assignment for Chemical Mixtures 3-46
3-16 Comparison of Alternative Load Calculation Methods 3-47
3-17 Results of DL Sensitivity Analysis 3-49
3-18 Conversion Factors for Nitrogen Compounds 3-50
3-19 Tables Imported or Created in DMRLoads2007 3-51
3-20 Case-by-Case Point Source Category Reassignments in DMRLoads2007 3-55
3-21 Comparison of Load and Concentration Ranges for Common Parameters 3-60
3-22 DMR 2007 Point Source Category Rankings by TWPE 3-61
3-23 Summary of DMRLoads2007 Facility Review 3-67
4-1 SIC Codes Counted in Multiple Point Source Categories 4-5
4-2 Facility-Level Point Source Category Assignment SIC Codes 4-6
4-3 Example NAICS/Point Source Category Crosswalk Development 4-11
4-4 NAICS Codes Counted in Multiple Point Source Categories 4-12
4-5 Facility-Level Point Source Category Assignment NAICS Codes 4-14
5-1 Revised Boron TWF 5-2
5-2 Chemicals with no TWFs in TRIReleases2007 5-2
5-3 Chemicals with no TWFs in DMRLoads2007 5-3
iv
-------
LIST OF TABLES (Continued)
Page
6-1 Number of Facilities in Categories Selected for Preliminary Category Review 6-2
7-1 TRIReleases2007 Point Source Category Rankings 7-6
7-2 DMRLoads2007 Point Source Category Rankings 7-8
7-3 Point Source Categories That Have Undergone a Recent Rulemaking or Review 7-11
7-4 Point Source Categories with One Facility Dominating the TWPE Discharges 7-13
7-5 Final DMRLoads2007 and TRIReleases2007 Combined Point Source Category
Rankings 7-15
-------
LIST OF FIGURES
Page
2-1 Relationship Between the Three TRI2007 Databases 2-5
2-2 Basic Structure of the TRIReleases200 7 Database 2-24
3-1 Relationship Between Data Sources and Database Development Tools for the
Development of DMRLoads2007 3-8
3-2 Example PCS STATS Code 3-12
3-3 Flow Diagram for PCS Load Calculator Routine 3-13
3-4 Relationship Diagram for Convert Module Output 3-32
3-5 Example STATS Code in ICIS-NPDES CONVERT_DMR Table 3-34
3-6 DMRLoadsAnalysis2007 Inputs Used to Create DMR2007 Table 3-44
3-7 DMRLoads2007 Database Structure 3-52
VI
-------
Section 1 - Introduction
1. INTRODUCTION
Under the Clean Water Act (CWA), EPA establishes national technology-based
regulations known as effluent guidelines and pretreatment standards to reduce pollutant
discharges from categories of industry discharging directly to waters of the United States or
discharging indirectly through publicly owned treatment works (POTWs). The CWA sections
301(d), 304(b), 304(g), and 307(b) require EPA to annually review these effluent guidelines and
pretreatment standards.
This document supports EPA's 2009 review of its existing effluent guidelines and
pretreatment standards. It also presents EPA's evaluation of categories of indirect dischargers
without pretreatment standards to identify potential new categories for pretreatment standards, as
required under CWA sections 304(g) and 307(b). Currently there are 56 point source categories
that have effluent guidelines and pretreatment standards, which include over 450 subcategories.
Additionally, CWA section 304(m) requires EPA to biennially publish an effluent guidelines
program plan and provides for public notice and comment on such plan. Therefore, this
document also supports the Preliminary 2010 Effluent Guidelines Program Plan (Preliminary
2010 Plan). Included in the Preliminary 2010 Plan is a solicitation for comments and data on
industry categories currently not subject to effluent guidelines that are discharging non-trivial
amounts of toxic or non-conventional pollutants.
EPA's annual review of effluent guidelines and pretreatment standards has several
components. First, EPA reviews all industrial categories subject to existing effluent limitations
guidelines and standards to identify potential candidates for revision, as required by the CWA
sections 304(b), 301(d), 304(g) and 307(b). The findings of this review are discussed in Section 7
of this report and are called the screening level analysis. Second, EPA reviews direct discharging
industries not currently subject to effluent limitations guidelines and standards to identify
potential candidates for effluent limitations guidelines development, as required by section
304(m)(l)(B) of the CWA. Finally, EPA reviews indirect discharging industries not currently
subject to pretreatment standards to identify potential candidates for pretreatment standards
development, as required by section 307(b).
In conducting a screening level analysis, EPA uses readily available information from the
Toxics Release Inventory (TRI), the Permit Compliance System (PCS), and the Integrated
Compliance Information System for the National Pollutant Discharge Elimination System (ICIS-
NPDES) to estimate the magnitude and relative toxicity of discharges from these industrial
wastewater discharges. Section 1 discusses how EPA uses Standard Industrial Classification and
North American Industrial Classification System codes to relate these discharge data to the 56
point source categories. EPA estimates the relative toxicity of these pollutant discharges in terms
of toxic-weighted pound equivalents (TWPE). EPA estimates TWPE based on toxic weighting
factors, which are discussed in detail in Section 5. EPA also uses available data to estimate
discharges of pollutants in pounds, such as nutrients. For its 2009 review, EPA used information
as reported to TRI, PCS, and ICIS-NPDES for 2007. EPA used 2007 data because these were the
most recent TRI data available at the time it began the 2009 annual review. EPA used 2007 PCS
and ICIS-NPDES data to reflect the same reporting year. EPA's 2009 screening level review is
similar to that used for previous annual reviews (U.S. EPA, 2004; U.S. EPA, 2006; U.S. EPA,
2008). EPA used the 2009 review to confirm the identification of the three industrial categories
prioritized for further review in the Final 2008 Effluent Guidelines Program Plan (73 FR 53218,
-------
Section 1 - Introduction
September 15, 2008) and to list the industrial categories currently regulated by existing effluent
guidelines that cumulatively comprise 95 percent of the reported hazard (reported in units of
TWPE) for preliminary category reviews.
This report describes the development of the databases that EPA used in conducting its
2009 screening-level analysis. This report is a companion report for the Technical Support
Document for the Preliminary 2010 Effluent Guidelines Program Plan (U.S. EPA, 2009). It also
presents the results of the 2009 screening-level analysis. The remainder of this report is divided
into the following sections:
• Section 2 - Development of TRIReleases2007;
• Section 3 - Development of DMRLoads2007;
• Section 1 - Identification of Point Source Categories;
• Section 5 - Toxic Weighting Factors (TWFs);
• Section 6 - Quality Review; and
• Section 7 - Results of 2009 Screening-Level Analysis.
Section 7 provides a summary of the TWPE calculated from the TRI data and PCS/ICIS-
NPDES data. EPA used the combined TWPE from the 2007 TRI and PCS/ICIS-NPDES data to
prioritize its review of industry sectors that offer the greatest potential for reducing hazard to
human health or the environment.
1.1 Introduction References
1. U. S. EPA. 2004. Technical Support Document for the 2004 Effluent Guidelines Program
Plan. EPA-821-R-04-014. Washington, DC. (August). EPA-HQ-OW-2003-0074-1346
through 1352.
2. U. S. EPA. 2006. Technical Support Document for the 2006 Effluent Guidelines Program
Plan. EPA-821-R-06-018. Washington, DC. (December). EPA-HQ-OW-2004-0032-
2782.
3. U. S. EPA. 2008. Technical Support Document for the 2008 Effluent Guidelines Program
Plan. EPA-821-R-08-015. Washington, DC. (August). EPA-HQ-OW-2006-0771-1701.
4. U. S. EPA. 2009. Technical Support Document for the Preliminary 2010 Effluent
Guidelines Program Plan. EPA-821-R-09-006. Washington, DC. (October). EPA-HQ-
OW-2007-0571 DCN 06703.
1-2
-------
Section 2 - Development of TRIReleases2007
2. DEVELOPMENT OF TRIRELEASES2007
As discussed in Section 1, EPA annually reviews promulgated effluent limitations
guidelines and pretreatment standards (ELGs) by investigating available information on
industrial pollutant discharges. EPA identified that the Toxics Release Inventory (TRI) contains
readily available and relevant data on industrial pollutant discharges, specifically that TRI data
has information on industrial pollutant discharges to surface waters ("direct discharges") and to
publicly owned treatment works (POTWs) ("indirect discharges"). Consequently, EPA was able
to use TRI data for its review of: (1) promulgated effluent guidelines ("direct discharges"); (2)
promulgated pretreatment standards ("indirect discharges"); and (3) direct and indirect industrial
pollutant discharges not currently subject to effluent guidelines or pretreatment standards. As
discussed in Section 7, EPA combined the toxic-weighted pound equivalent (TWPE) calculated
from the TRI data and the discharge monitoring report data contained in the Permit Compliance
System (PCS) and the Integrated Compliance Information System for the National Pollutant
Discharge Elimination System (ICIS-NPDES) (see Section 3 for information about PCS and
ICIS-NPDES). EPA used this combined TWPE to prioritize its review of industry sectors that
offer the greatest potential for reducing hazard to human health or the environment.
This section discusses the methodology EPA used to create TRIReleases2007, a database
created by the EPA to analyze 2007 TRI data. It also presents the unweighted annual pollutant
load (i.e., pounds) and the relative toxicity of these discharges using toxic weighting factors
(TWFs) (i.e., TWPE) for all facilities reporting discharges to TRI for the year 2007 and for the
point source categories that these facilities represent. Tables A-l and A-2 in Appendix A present
the annual pollutant load and TWPE from TRIReleases2007 on a six-digit North American
Industrial Classification System (NAICS) code and chemical basis, respectively. This section is
organized in the following subsections:
• Section 2.1 - General TRI information;
• Section 2.2 - Overview of TRI databases;
Section 2.3 - TRIRawData2007;
• Section 2.4 - TRICalculations2007;
• Section 2.5 - TRIReleases2007;
• Section 2.6 - TRIReleases2007; and
• Section 0 - TRIReleases2007 References.
2.1 TRI
TRI is the common name for Section 313 of the Emergency Planning and Community
Right-to-Know Act. Each year, facilities that meet certain criteria must report their releases and
other waste management activities of listed toxic chemicals (i.e., the quantities of toxic
chemicals recycled, collected and combusted for energy recovery, treated for destruction, or
disposed by the facility). A separate report must be filed for each chemical that exceeds the
reporting threshold. For the 2009 annual review of effluent guidelines, EPA used data for
reporting year 2007, because they were the most recent data available at the time the review
began. The TRI list of chemicals for reporting year 2007 includes more than 600 chemicals and
chemical categories. Prior to 2006, facilities were required to identify their operations using
Standard Industrial Classification (SIC) codes. Starting with reporting year 2006, EPA modified
2-1
-------
Section 2 - Development of TRIReleases2007
the requirement for facilities to report SIC codes, so that facilities now use the NAICS code (73
FR 324666).
A facility must submit a TRI report if it meets the following three criteria (U.S. EPA,
2001):
1. NAICS Code Determination: Most facilities in NAICS codes 11, 21, 22, 31
through 33, 42, 48 through 49, 51, 54, 56 and 81, and federal facilities are
potentially subject to TRI reporting. EPA generally relies on facility claims
regarding the NAICS code identification. The primary NAICS code determines if
TRI reporting is required. The primary NAICS code is associated with the
facility's revenues, and may not relate to their pollutant discharges (73 FR 12045,
March 6, 2008).
2. Number of Employees: Facilities must have 10 or more full-time employees or
their equivalent. EPA defines a "full-time equivalent" as a person who works
2,000 hours in the reporting year (there are several exceptions and special
circumstances that are well defined in the TRI reporting instructions).
3. Activity Thresholds: If the facility is in a covered NAICS code and has 10 or
more full-time employee equivalents, it must conduct an activity threshold
analysis for every chemical and chemical category on the current TRI list. The
facility must determine whether it manufactures, processes, or otherwise uses
each chemical at or above the appropriate activity threshold. Reporting thresholds
are not based on the amount of release. All TRI thresholds are based on mass, not
concentration. Thresholds for persistent bioaccumulative toxic (PBT) chemicals
are lower than for non-PBT chemicals.
In TRI, facilities report annual releases to the environment of each toxic chemical or
chemical category that meets reporting requirements. TRI requires facilities to report on-site
releases to air, receiving streams, disposal to land, underground wells, and several other
categories. Facilities must also report the amount of toxic chemicals in wastes transferred to off-
site locations, including discharges to POTWs and other off-site locations, such as commercial
waste disposal facilities.
For this review, EPA focused on facility reports of chemical discharges directly to a
receiving stream or transfers to a POTW. For discharges directly to a stream ("direct
discharges"), EPA took the annual loads directly from the reported TRI data for calendar year
2007. For transfers of chemicals to POTWs ("indirect discharges"), EPA first adjusted the TRI
pollutant loads to account for pollutant removal at the POTW prior to discharge to the receiving
stream (see Section 2.4.2 for more details).
TRI does not require facilities to sample and analyze wastestreams to determine the
quantities of toxic chemicals released. Facilities may estimate releases based on mass balance
calculations, published emission factors, site-specific emission factors, or other approaches.
Facilities must indicate the basis of their release estimate using a reporting code. According to
TRI's reporting guidance, facilities should use one-half the detection limit to estimate mass
releases of chemicals that are measured below their detection limit and are reasonably expected
-------
Section 2 - Development of TRIReleases2007
to be present. Nondetects of dioxin and dioxin-like compounds, however, may be reported as
zero.
TRI allows facilities to report releases as specific numbers or as ranges, if appropriate.
Specific estimates are encouraged if data are available to ensure the accuracy; however, EPA
allows facilities to report releases in the following ranges: 1 to 10 pounds, 11 to 499 pounds, and
500 to 999 pounds. For this review, if a facility reported releases in a range, EPA used the mid-
point of each reported range to represent a facility's releases.
2.1.1 Utility of TRI
The data collected in TRI are particularly useful for the 304(m) review process for the
following reasons:
• TRI includes data from all 50 states and U.S. territories;
• TRI includes transfers to POTWs, not just direct discharges;
• TRI includes discharge data from manufacturing NAICS codes and some other
industrial categories which may handle significant quantities of toxic chemicals;
and
• TRI includes releases of many chemicals, not just those already identified as
problems and limited in facility discharge permits.
2.1.2 Constraints and Limitations of TRI
TRI provides comprehensive data for direct and indirect discharging facilities. However,
EPA identified the following constraints and limitations to using TRI for the screening-level
analysis:
• Small establishments (less than 10 employees) are not required to report, nor are
facilities that do not meet the reporting thresholds. Therefore, facilities reporting
to TRI may not provide a complete picture of the industry.
• Release reports are, in part, based on estimates, not measurements, which may
result in inaccurately reported releases. For example, TRI encourages facilities to
report some compounds as present at one-half the detection level if a facility
suspects that the compound has the potential to be present, even if measured data
show the compound is below its detection level. As a result, many companies are
conservative and adopt this approach. For facilities with large flows, this can
result in large estimates of pounds or TWPE of pollutant released with no
measurements to support that the compound was ever present above the detection
level.
• Certain chemicals (polycyclic aromatic compounds (PACs), dioxin and dioxin-
like compounds, and metal compounds) are reported as a class, not as individual
compounds. Because the individual compounds in the class have widely varying
toxic effects, the potential toxicity of chemical releases can be inaccurately
estimated.
• Facilities are identified by NAICS code, not point source category. For some
NAICS codes, it may be difficult or impossible to identify the point source
category that is the precise source of the toxic wastewater releases (see Section 1
for additional information).
2-3
-------
Section 2 - Development of TRIReleases2007
• The list of chemicals covered by TRI is not all-inclusive and changes over time.
• Only facilities in certain NAICS codes are required to report; therefore, some
sources of water pollutant discharges are not included.
• A facility is not required to report releases if the releases do not exceed the
reporting threshold.
• Information in TRI does not represent national estimates because not all facilities
are required to report to TRI.
Despite TRI's limitations and constraints, EPA has determined that it is appropriate to be
used for an initial screening-level review and prioritization of the pollutant loadings discharged
by industrial categories. EPA will further evaluate the prioritized categories in a second level of
review which may include additional data collection and verification of data reported in TRI.
2.2 Overview of TRI Databases
EPA developed the end-user database, TRIReleases2007, in three steps:
1. Downloaded relevant data from TRI to create TRIRawData2007 (see Section 2.3).
2. Estimated relative toxicity of discharges, set up groupings of facilities (by NAICS
code and discharge type), and made corrections and adjustments to create
TRICalculations2007 (see Section 2.4).
3. Grouped the pollutant discharges in TRICalculations2007 by NAICS code, point
source category, and other groupings to create TRIReleases2007 for rankings and
other analyses (see Section 2.5).
Figure 2-1 shows how these three databases are related and the following sections
describe the creation and particulates of each database in greater detail.
2-4
-------
Section 2 - Development of TRIReleases2007
EPA TRI Web Site
Files
File Type 1
File Type 2B
File Type 3A
File Type 3B
File Type 4
(Downloaded)
TRIRawData2007
Tables
>US_l_2007_v07
US_2b_2007_v07
US_3a_2007_v07
US_3b_2007_v07
US 4 2007 v07
(Copied)
Tables
. TRI Raw Data 2007 Tables
Priority Pollutants
PSC Codes
NAICS/PSC Crosswalk
TWFs
POTW Removals
Dioxin Chemicals
TRI Master List
TRI Master Facility List
(Linked)
Tables
. TRI Calculations 2007 Tables
Counts of Facilities by NAICS
Counts of Facilities by PSC
NAICS Code Rankings
Point Source Rankings
Figure 2-1. Relationship Between the Three TRI 2007 Databases
2.3 TRIRawData2007
EPA created TRIRawData2007 using the 2007 TRI data for all of the United States,
which are available from the EPA Web site (www.epa.gov/tri). Table 2-1 lists the relevant TRI
2007 files that EPA imported into the Microsoft Access™ database.
Table 2-1. TRI 2007 Tables Downloaded from EPA
Table Name
"File Type 1 : Facility,
Chemical, Releases and Other
Waste Management Summary
Information"
"File Type 2B: Detailed On-
Site Waste Treatment Methods
and Efficiency"
Description of File Contents
Facility information (Part I on Form R and Form A), as well as most chemical
information (Part II on Form R and Form A). Data elements are reported
individually. The information is also disaggregated based on Waste Management
code (i.e., Management "M" code reported on TRI Form R), and aggregated up to
On-site Releases, Off-site Releases, Other On-site Waste Management, and
Transfers Off Site for Further Waste Management categories.
Facility information (Part I on Form R and Form A) and On-site Waste
Treatment Methods and Efficiency data (Part II, Section 7 A on Form R).
2-5
-------
Section 2 - Development of TRIReleases2007
Table 2-1. TRI2007 Tables Downloaded from EPA
Table Name
"File Type 3A: Details of
Transfers Off Site"
"File Type 3B: Details of
Transfers to POTW"
"File Type 4: Details of Facility
Information"
Description of File Contents
Facility information (Part I on Form R and Form A) as well as details of
individual transfers off-site (Part II, Section 6.2 on Form R).
Facility information (Part I on Form R and Form A) as well as a list of POTWs
(Part II, Section 6. l.B on Form R).
Facility information (Part I on Form R and Form A) for all facilities that have
ever reported to the TRI program. The "reporting year" field at the beginning of
each record identifies the last year the facility reported to the TRI program.
Source: http://www.epa.gov/tri/tridata/tri07/data/index.htm.
2.4 TRICalculations200 7
As the second step in developing TRIReleases2007, EPA created TRICalculations2007
by copying raw data tables from TRIRawData2007', omitting unrelated data (e.g., air emissions
and source reduction activities), and performing the following actions:
• Corrected NAICS code classification for certain facilities and chemicals and
corrected certain reported chemical quantities (Section 2.4.1);
• Estimated POTW removals for indirect discharges (Section 2.4.2);
• Estimated the mass-based and toxic-equivalent pollutant loadings (Section 2.4.3);
• Combined releases of parent metals and their associated compounds (Section
2.4.4); and
• Determined basis of TRI release and transfer estimates (Section 2.4.5).
To perform the calculations listed above, EPA imported tables from previous versions of
EPA's TRICalculations databases containing Chemical Abstract Service (CAS) numbers, TWFs,
and POTW removal rates. Table 2-2 lists the database tables that EPA imported or created in
TRICalculations2007.
Table 2-2. Tables Imported or Created in TRICalculations2007
Table Name
"All Water Releases"
"Dioxin Chemicals"
"Dioxin Distributions"
"Manual Data
Changes"
"Manual Dioxin
Distribution Changes"
Created or
Imported?
Created using VBA
code
Imported from
TRIReleases2005
Created using VBA
Code
Created
Created a
Description
Lists calculated TWPE for every chemical discharge reported to
TRI in 2007 for which EPA has calculated a TWF. EPA
developed this table using data from TRIRawData2007 and TWF
tables. This table serves as an intermediate table between the
TRIRawData2007 tables and the TRI Master List Table.
Lists the 17 dioxin congeners and the TRI congener number
associated with each.
Lists the dioxin distributions (see Section 2.4.3) of all facilities
reporting dioxin discharges in 2007. EPA developed this table
using data from TRIRawData2007.
Documents any changes that EPA made to the data from
TRIRawData2007.
Documents changes that EPA made to the dioxin distributions
from TRIRawData2007.
2-6
-------
Section 2 - Development of TRIReleases2007
Table 2-2. Tables Imported or Created in TRICalculations2007
Table Name
"Manual Load
Changes"
"Manual NAICS Code
Changes"
"NAICS Code
Changes"
"NAICS Code Changes
(for facilities with no
NAICS Codes)"
"NAICS_Codes"
"OCPSF Pesticides"
"Parent Metals and
Compounds"
"Pesticides Chemical
List"
"POTW Removals"
"Priority Pollutants"
"PSC Codes"
"PSC/NAICS
Crosswalk"
"PSC/NAICS
Crosswalk without M
PM_fixed"
"Pulp and Paper Dioxin
Distribution"
"Pulp and Paper
Phases"
"TRI Raw Data 2007
Tables"
"Point Source Category
Codes"
Created or
Imported?
Created a
Created a
Created a
Created a
Created
Created using VBA
code
Imported from
TRIReleases2005
Imported from
TRIReleases2005
Imported from
TRIReleases2005
Imported from
TRIReleases2000
Imported from
TRIReleases2005
Created
Created
Imported from
TRIReleases2005
Imported from
TRIReleases2005
Imported from
TRIRawData2007
Imported from
TRIReleases2005
Description
Documents changes that EPA made to the magnitude of the
discharges in the data from TRIRawData2007.
Documents changes that EPA made to the NAICS codes in the
data from TRIRawData2007 '.
Documents changes that EPA makes every year to the NAICS
codes of the discharges in the data from TRIRawData2007 '.
Documents the NAICS codes that EPA assigned to facilities with
no NAICS codes in the data from TRIRawData2007 '.
Lists U.S. Economic Census definitions of the NAICS codes.
EPA developed this table using information from the U.S.
Economic Census Web site (www.census.gov).
Lists all pesticide discharges reported for 2007 classified under
the Pesticide Chemicals Category.
Links parent metals to the appropriate metal compound groups
(e.g., nickel and nickel compounds).
Lists all chemicals classified as pesticides under the Clean Water
Act.
Lists all 612 TRI chemicals and chemical compounds and their
chemical-specific average POTW percent removal. See "POTW
Percent Removals Used for the TRIReleases2002 Database"
(Codding, 2005) (see Section 2.4.2),
Lists priority pollutants (CAS number and chemical name).
Defines all codes for point source categories.
EPA used the "SIC/Point Source Category Crosswalk" table from
TRIReleases2000 and the NAICS/SIC Crosswalk developed for
the 2002 U.S. Economic Census to develop this table (see Section
1).
Identifies NAICS codes that could have discharges subject to the
Metal Products and Machinery (MP&M) ELGs. EPA created this
table from information in the MP&M rulemaking.
Contains the dioxin distribution used for facilities in the Pulp,
Paper, and Paperboard Category that do not report a facility-
specific distribution. EPA developed this table using information
obtained from the pulp and paper industry (Matuzko et al., 2006).
Lists the NAICS code placeholders used to identify facilities in
the Pulp, Paper, and Paperboard Category by regulatory phase.
Copy of all original TRI tables stored in the TRIRawData2007
database and deleted information not needed for the 2009 annual
review.
Lists point source categories and corresponding point source
category codes.
2-7
-------
Section 2 - Development of TRIReleases2007
Table 2-2. Tables Imported or Created in TRICalculations2007
Table Name
"TWFs"
"TRI Chemicals with
MP&My/n"
"TRI Master List"
"TRI Master Facility
List"
"Wood Preserving
Dioxin Distribution"
Created or
Imported?
Imported from
TRIReleases2005
Imported from
TRIReleases2005
Created using VBA
code
Created using VBA
code
Imported from
TRIReleases2005
Description
Lists TWF for chemicals based on the Office of Water references.
EPA originally created this table using TWFs as of December
2004 and updates it with TWFs created or revised after 2004 (see
Section 5).
Lists chemicals regulated by the MP&M rulemaking.
Lists calculated pounds and TWPE for every chemical released
by every facility reporting to TRI in 2007. EPA developed this
table using data from "All Water Releases" and "TWFs" tables.
Complete and unique list of all facilities reporting to TRI,
relevant facility information (address, contacts, etc.), and
corresponding primary NAICS codes. EPA developed this table
using data from TRIRawData2007.
Contains the dioxin distribution used for facilities in the Wood
Preserving Category that do not report a facility-specific
distribution. EPA developed this table using information obtained
from the wood preserving industry.
a Most of these changes are based on previous knowledge about the facility's operations from previous annual
reviews. EPA also added changes discovered as part of the 2009 annual review. See Section 2.4.1 for additional
information.
VBA - Visual Basic for Applications.
2.4.1 Modifications to TBI-ReportedData
Modifications to TRI-Reported data include the following facility-specific changes:
• Pollutant loading changes;
• Dioxin distribution changes; and
• NAICS code changes.
During the screening-level reviews of the 2000 through 2007 TRI data, EPA made
corrections to TRIReleases databases based on information received from stakeholders, including
industry trade associations, facilities, and pretreatment coordinators. The SIC code corrections
identified for past years of review were converted to NAICS code corrections using the U.S.
Economic Census linkages and similarly applied to the 2007 data, as appropriate. In addition,
EPA conducted a quality review of the TRIReleases2007 database (described in Section 6). As a
result of this review, EPA made 58 corrections1 to the 2007 releases. Table A-3 in Appendix A,
lists the corrections EPA made to the TRIReleases2007 database.
EPA assigned pollutant loadings to point source categories based on the primary NAICS
code that facilities reported (see Section 1). A facility reports up to six NAICS codes to TRI and
specifies one primary NAICS code. In cases where EPA was able to identify that chemical
1 In addition to the 58 changes that were made to individual releases reported to TRI, EPA made 795 NAICS code
changes to account for facilities that did not report a NAICS code and SIC code changes that EPA had made to
facilities in past years. EPA identified the appropriate NAICS code for facilities that did not report a NAICS code
using Envirofacts.
2-8
-------
Section 2 - Development of TRIReleases2007
releases to surface water or a POTW were related to activities covered by a different NAICS
code, EPA corrected the NAICS code assigned to the facility and/or chemical. For example, a
facility may report their primary NAICS code as 325110, Petrochemical Manufacturing. The
facility may also perform pesticide manufacturing, which is covered under NAICS code 325320,
Pesticide and Other Agricultural Chemical Manufacturing. If this facility reported a pesticide
release, EPA assigned the pesticide release to the Pesticide Chemicals Category, because these
pollutant discharges are regulated under the Pesticide Chemicals Category, not the Organic
Chemicals, Plastics, and Synthetic Fibers Category. Section 1 in this report provides a detailed
discussion of the development of the crosswalk between the NAICS code and point source
category.
2.4.2 POTW Removals
For facilities that reported transfers of chemicals to POTWs, EPA first adjusted the
reported pollutant loads to account for pollutant removal that occurs at the POTW prior to
discharge to the receiving stream. EPA estimated the pounds of facilities' waste released to the
surface water after POTW removal using the following equation2:
Release to Stream (Ibs/yr) = [Transfer to POTW (Ibs/yr)] x [1 - POTW Removal (%)]
EPA developed a POTW removal hierarchy for the TRIReleases2002 database, described
in the memorandum entitled "POTW Percent Removals Used for the TRIReleases Databases"
(Codding, 2005). The TRIReleases2007 database uses the same POTW removal hierarchy. In
short, EPA used removal efficiencies from the following data sources, listed in order of
preference:
1. Recent effluent guidelines rulemakings;
2. EPA/Office of Research and Development's National Risk Management and
Research Laboratories treatability database; and
3. EPA/Office of Prevention, Pesticides, and Toxic Substances' Risk Screening
Environmental Indicators model.
Table A-4 in Appendix A lists the POTW Removals and their data sources, in
alphabetical order.
2.4.3 TWFs
To identify potential impacts on human health and the environment, EPA estimated toxic
equivalent mass discharge through the use of TWFs. EPA used the "TWFs" table, which lists
TWFs by CAS number, in TRICalculations2007 to calculate TWPE for chemical discharges. If
the table did not list a TWF for a specific parameter, EPA did not include pollutant discharges
for this chemical in its TWPE estimates. Section 5 describes TWFs in more detail. See also Toxic
Weighting Factor Development in Support of the 304(m) Planning Process (ERG, 2005).
In some cases, EPA calculated industry-specific TWFs for certain chemical compound
categories reported in TRI. These TWFs were not used to calculate TWPE for chemical
For example, the POTW removal efficiency for lithium carbonate is 1.85 percent. That is if 10,000 Ibs of lithium
carbonate discharged to a POTW, only 9,815 Ibs of lithium carbonate will be discharged from the POTW to surface
waters as this amount is untreated by the POTW [9,815 Ibs = 10,000 Ibs x (1 - 0.0185)].
2-9
-------
Section 2 - Development of TRIReleases2007
discharges in PCS/ICIS-NPDES. EPA created specific TRI TWFs when it had additional
information about the composition of the compound category, as released from specific
industries. Table 2-3 lists the calculated TWFs.
Table 2-3. TWF Modifications
Chemical
Dioxins
Creosote
PACs
PACs
PACs
PACs
Point Source Category
All
All
All point source categories, except those in the
Petroleum Refining; Wood Preserving; and Pulp,
Paper, and Paperboard Categories
Petroleum Refining Category
Wood Preserving Category
Pulp, Paper, and Paperboard Category
TWF
Apply individual dioxin compound TWF
using the following dioxin congener
distribution: 1) facility-specific, 2) industry
specific, 3) NAICS-code-average, or 4)
median dioxin TWF for all dioxin congeners.
1.3577 a
100.66
26.28 b
8.36 b
34.21 c
a Calculations of TWF and TWPEfor Creosote from Wood Preserving Facilities (Bicknell, 2004).
b TRI 2002 PACs TWF for Petroleum Refining, Creosote, and Wood Preserves (Finseth, 2005).
c NCASI SARA Handbook - Table 5 PAC Concentrations in Pulp Mill Effluents (H.C. Lavallee, Inc., 2005).
The remainder of this subsection describes how EPA developed the TWFs, in the
following order:
• Dioxins for all categories;
• Creosote for all categories;
• Wood Preserving Category creosote;
• Polycyclic Aromatic Compounds (PACs) for all categories;
• Petroleum Refining Category PACs;
• Wood Preserving Category PACs; and
• Pulp, Paper, and Paperboard Category PACs.
Dioxins
The term 'dioxins' refers to poly chlorinated dibenzo-p-dioxins (CDDs) and
poly chlorinated dibenzofurans (CDFs), which constitute a group of PBT chemicals. There are 17
CDDs and CDFs congeners with chlorine substitution of hydrogen atoms at the 2, 3, 7, and 8
positions on the benzene rings, the most toxic of which is 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD). The 17 compounds (called congeners) are referred to as 'dioxin-like,' because they
have similar chemical structure, similar physical-chemical properties, and invoke a common
battery of toxic responses (U.S. EPA, 2000), though the toxicity of the congeners varies greatly.
In this report, EPA uses the term "dioxin and dioxin-like compounds" to refer to all 17 of the
2,3,7,8-substituted CDDs and CDFs.
EPA developed TWFs for each of the 17 dioxin congeners, ranging from 703,584,000 for
2,3,7,8-TCDD to 2,021 for octachlorodibenzofuran. Due to their toxicity and ability to
bioaccumulate, the various dioxin congeners have high TWFs relative to most chemicals.
2-10
-------
Section 2 - Development of TRIReleases2007
Consequently, even small mass amounts of dioxin and dioxin-like compound discharges translate
into high TWPEs. Table 2-4 presents the dioxin congener-specific TWFs used in the screening-
level analysis.
Table 2-4. Dioxins Congeners and Their Toxic Weighting Factors
CAS
Number
Chemical Name
Abbreviated Name
Toxic Weighting
Factor
CDDs
1746-01-6
40321-76-4
39227-28-6
57653-85-7
19408-74-3
35822-46-9
3268-87-9
2,3,7,8-tetrachlorodibenzo-p-dioxin
1,2,3,7,8-pentachlorodibenzo-p-dioxin
1,2,3,4,7,8-hexachlorodibenzo-p-dioxin
1,2,3,6,7,8-hexachlorodibenzo-p-dioxin
1,2,3,7,8,9-hexachlorodibenzo-p-dioxin
1,2,3,4,6,7,8-heptachlorodibenzo-p-dioxin
1,2,3,4,6,7,8,9-octachlorodibenzo-p-dioxin
2,3,7,8-TCDD
1,2,3,7,8-PeCDD
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8,9-OCDD
703,584,000
692,928,000
23,498,240
9,556,480
10,595,840
411,136
6,586
CDFs
51207-31-9
57117-41-6
57117-31-4
70648-26-9
57117-44-9
72918-21-9
60851-34-5
67562-39-4
55673-89-7
39001-02-0
2,3 ,7,8-tetrachlorodibenzofuran
1 ,2,3 ,7,8-pentachlorodibenzofuran
2,3,4,7,8-pentachlorodibenzofuran
1,2,3,4,7,8-hexachlorodibenzofuran
1,2,3,6,7,8-hexachlorodibenzofuran
1,2,3,7,8,9-hexachlorodibenzofuran
2,3,4,6,7,8-hexachlorodibenzofuran
1,2,3,4,6,7,8-heptachlorodibenzofuran
1,2,3,4,7,8,9-heptachlorodibenzofuran
1,2,3,4,6,7,8,9-octachlorodibenzofuran
2,3,7,8-TCDF
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
1,2,3,4,6,7,8,9-OCDF
43,819,554
7,632,640
557,312,000
5,760,000
14,109,440
47,308,800
51,204,160
85,760
3,033,984
2,021
Source: Toxic Weighting Factor Development in Support of the 304(m) Planning Process (ERG, 2005).
Beginning with reporting year 2000, facilities meeting certain reporting criteria were
required to report to TRI the total mass, in grams, of the 17 dioxin and dioxin-like compounds
released to the environment every year. This reporting method does not account for the relative
toxicities of the 17 compounds. However, reporting facilities are given the opportunity to report
a facility-specific congener distribution. Yet even if dioxins are released to more than one
medium, the facility can report only one distribution. EPA cannot know if the single dioxin
congener distribution reported by a facility accurately reflects the dioxin distribution in
wastewater3. Nevertheless, it is the best available information and EPA uses it to calculate the
reporting facility's dioxin TWPE.
To account for the relative toxi cities of the different dioxin congeners, EPA first
converted the reported dioxin releases from grams to pounds to be consistent with the units used
for other chemicals. EPA then calculated dioxin TWPE estimates using the facility-specific
congener distributions for all facilities that reported a distribution. Based on information
Beginning with reporting year 2008, facilities will be required to submit information on the amount of each
individual dioxin congener where that information is available. Facilities that cannot quantify dioxin releases by
congener may continue to report an aggregate number (72 FR 26544; May 10, 2007).
2-11
-------
Section 2 - Development of TRIReleases2007
provided by the facilities that were contacted as part of previous and current quality reviews,
EPA made corrections to the reported dioxin distributions for the facilities presented in Table
2-5. Table 2-5 also includes the reason for the correction.
Table 2-5. EPA Facility-Specific Dioxin Congeners Distribution Corrections
Facility Name
Cahaba Pressure Treated Forest
Products Inc.
Du Pont Memphis Plant
Louisiana Pigment Co LP
Colfax Treating Co LLC
Eastman Kodak Co Kodak Park
Facility
Location
Brierfield, AL
Memphis, TN
Westlake, LA
Pineville, LA
Rochester, NY
Reason for Dioxin Congener Distribution Change
The facility did not use the industry -provided dioxin
distribution (Woodruff, 2007).
The facility provided more detailed information about the
dioxin measurements (Zweig, 2000).
The facility provided more detailed information about the
dioxin measurements (Kashyap, 2009).
The facility did not use the industry-provided dioxin
distribution based on pentachlorophenol distribution
(Johnston, 2004).
The facility provided more detailed information about the
dioxin measurements (Moeller, 2009)
EPA calculated an average dioxin distribution for each NAICS code which had reported
dioxin releases. For facilities that did not report a dioxin distribution, EPA used the average
NAICS code distribution to calculate the facility's dioxin TWPE. EPA calculated industry-
specific dioxin distributions for the Petroleum Refining and the Pulp, Paper, and Paperboard
Categories based on information received from industry trade groups. For facilities that did not
report a congener distribution and did not have any facilities within its NAICS code that reported
a congener distribution, EPA used a TWF equal to 10,595,840 (the median of the 17 dioxin
congener TWFs).
Creosote
Creosote is a commonly used wood preservative, comprising many different chemicals.
EPA did not develop a TWF for creosote using creosote toxicity data. Instead, EPA used the
chemical composition of creosote, provided in IARC Monographs, Vol 35, "Coal Tar and
Derived Products," (WHO, 1998) and the TWFs for these individual chemicals to calculate a
TWF for creosote.
EPA made the following assumptions in developing the TWF for creosote:
1. Chemicals will be present in wastewater in the same proportion that they are
present in the creosote.
2. If no TWF was available for a specific chemical, its concentration in creosote was
assumed to be zero.
Using the data provided in IARC Monographs, Vol 35, EPA calculated the average percentage
that the chemical represents in creosote based on the high and low value (WHO, 1998). EPA
calculated an adjusted TWF for each chemical by multiplying its chemical-specific TWF by its
average percentage in creosote. EPA summed these values to calculate a new overall TWF for
creosote discharges. Table 2-6 lists the chemical composition of creosote, along with the
associated TWF of the various chemicals.
2-12
-------
Section 2 - Development of TRIReleases2007
Table 2-6. Chemical Composition of Creosote and TWF
Pollutant
Acenaphthene
Antracene
Benz(a)anthracene
Benzo(a)pyrene
Benzofluourenes
Biphenyl
Carbazole
Chrysene
Dibenz(a,h)anthtracene
Dibenzofuran
Dimethylnaphthalenes
Fluoranthene
Fluorene
Methylantracenes
Methylfluorenes
1 -Methy Inaphthalene
2-Methylnaphthalene
Methy Iphenanthrenes
Naphthalene
Phenanthrene
Pyrene
Chemical Percentage
(%)
11.85
4.50
0.21
0.05
1.50
1.20
1.60
2.80
0.03
5.75
2.15
5.25
8.65
3.95
2.65
6.45
6.60
3.00
9.65
18.50
4.75
TWF
0.0325697
2.5455945
30.695
100.66
0.1555556
0.0365558
0.709071
31.01
30.772
0.49215
1.2846944
0.70105
0.0486957
0.0062222
0.1930493
0.1037037
0.0158701
0.2947368
0.0932033
Total
Adjusted TWF
0.0038595
0.1145518
0.0644595
0.05033
0.0023333
0.0004387
0.0113451
0.86828
0.0092316
0.0282986
0
0.0674465
0.0606408
0
0.0012904
0.0004013
0.0127413
0.0031111
0.0015315
0.0545263
0.0044272
1.36
Creosote Releases from Wood Preserving Facilities
EPA received information from the Southern Pressure Treaters Association in 2005 that
indicates creosote discharges are estimated based on a surrogate analyte, such as oil and grease
or total phenols. The Southern Pressure Treaters Association also indicated that TRI-reported
PAC discharges are usually estimated based on the creosote estimates, but there is no standard
approach for making these estimates (H.M. Rolling Company, 2005). PACs and creosote contain
many of the same chemicals (compare Table 2-7 and Table 2-6, respectively). Consequently, if
EPA estimated the TWPE for both the PACs and the creosote in the same discharge, then the
discharges of some toxic chemicals would be double counted. For this reason, if a wood
preserving facility reports PACs and creosote in the same discharge (e.g., both are reported in
direct discharges to surface water), EPA included the TWPE for the PAC discharges, but not the
creosote discharges. If the wood preserving facility reports only creosote releases (and not
PACs), EPA used the calculated creosote TWF of 1.36 to calculate the TWPE.
2-13
-------
Section 2 - Development of TRIReleases2007
Polycyclic Aromatic Compounds (PACs)
PACs, sometimes known as polycyclic aromatic hydrocarbons (PAHs), are a class of
organic compounds consisting of two or more fused aromatic rings. Table 2-7 lists the 21
individual compounds in the PAC category for TRI reporting, CAS number, and TWF, if
available. EPA has TWFs for only eight of the 21 PACs chemicals.
Table 2-7. Definition of Polycyclic Aromatic Compounds
PAC Compound
Benzo(a)anthracene
Benzo(a)phenanthrene (chrysene)
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(j)fluoranthene
Benzo(k)fluoranthene
Benzo(j,k)fluorene (fluoranthene)
Benzo(r,s,t)pentaphene
Dibenz(a,h)acridine
Dibenz(aj )acridine
Dibenzo(a,h)anthracene
Dibenzo(a,e)fluoranthene
Dibenzo(a,e)pyrene
Dibenzo(a,h)pyrene
Dibenzo(a,l)pyrene
7H-Dibenzo(e,g)carbazole
7, 12-Dimethylbenz(a)anthracene
Indeno( 1 ,2,3 -cd)pyrene
3 -Methy Icholanthrene
5 -Methy Ichrysene
1-Nitropyrene
CAS Number
56-55-3
218-01-9
50-32-8
205-99-2
205-82-3
207-08-9
206-44-0
189-55-9
226-36-8
224-42-0
53-70-3
5385-75-1
192-65-4
189-64-0
191-30-0
194-59-2
57-97-6
193-39-5
56-49-5
3697-24-3
5522-43-0
Toxic Weighting Factor
36.2600
31.0100
100.6600
30.6600
NA
30.6600
0.8290
NA
NA
NA
30.6600
NA
NA
NA
NA
NA
NA
30.6600
NA
NA
NA
NA - Not available. EPA has not developed TWFs for these compounds.
PACs are classified as PBTs. They are likely present in petroleum products such as crude
oil, fuel oil, diesel fuel, gasoline, and paving asphalt (bituminous concrete) and refining by-
products such as heavy oils, crude tars, and other residues. PACs form as the result of incomplete
combustion of organic compounds. PACs and closely related compounds are major constituents
of creosote, a commonly used wood preservative.
For TRI, facilities that manufacture, process, or use more than 100 pounds of PACs per
year must report the combined mass of PACs released; they do not report releases of individual
compounds. In the development of TRIReleases2007 EPA assigned the TWF of benzo(a)pyrene
to PACs, with the exception of releases reported by facilities in the Petroleum Refining; Wood
Preserving; and Pulp, Paper, and Paperboard Categories (for which EPA has more detailed
information). Because the TWF for benzo(a)pyrene (100.66) is higher than any other PAC, this
2-14
-------
Section 2 - Development of TRIReleases2007
represents a worst-case scenario. For PAC discharges that are not completely benzo(a)pyrene,
this method overestimates the relative toxicity of the discharges.
Petroleum Refining PACs
EPA used available data for the Petroleum Refining Category to calculate TWPE for
PACs reported by petroleum refining facilities. Facilities report to TRI the combined mass of
PACs released, but for this industry EPA used information on the distribution of PACs in
refinery products from the American Petroleum Institute (API, 1994). EPA assumed that the
composition of PACs released by refineries is proportional to the composition of raw materials
(crude oil) and products throughput at U.S. refineries. EPA developed this methodology for the
Petroleum Refining Detailed Study supporting the 2004 Effluent Guidelines Program Plan (U.S.
EPA, 2004). After the methodology was developed, the calculated refinery PAC TWF changed
due to the changes in TWFs for individual PAC chemicals.
PACs can occur in a number of petroleum products and crude oils; this information is
available in literature (see Table 2-8 and Table 2-9). In addition, the Energy Information
Administration (EIA) publishes a yearly report of the amount of petroleum products produced in
all U.S. petroleum refineries as well as the amount of crude oil consumed (see Table 2-10).
EPA made the following assumptions in developing the TWF for Petroleum Refining
Category PACs:
1. PACs will be present in wastewater in the same proportion that they are present in
the crude oil and products throughput at U.S. refineries. Table 2-10 presents these
proportions.
2. If EPA did not have literature data available for a specific PAC compound, its
concentration in the crude oil or product was assumed to be zero. If a PAC
compound was reported as not detected, its concentration in the crude oil or
product was assumed to be zero.
3. Where PAC composition is not available, it can be estimated using the
composition from similar products. Table 2-11 lists the products for which PAC
composition is not available and the similar product used to estimate the
composition.
4. For crude oil, representative domestic and foreign oils can be used to calculate a
weighted average PAC composition for crude oil. According to EIA4, 39.1
percent (volumetric basis) of the total consumed crude oil in the United States in
the year 2000 was domestic while 60.9 percent (volumetric basis) was imported.
EPA selected South Louisiana Oil, for which PAC composition is available, as a
representative domestic oil and Alberta Oil as a representative foreign oil. EPA
assumed that a weighted average of the composition of these two crude oils is a
reasonable representation of crude oil composition for the purpose of this study.
EPA also used a specific weight of 0.92 for crude oil to convert PAC
concentrations reported as mg/kg to mg/L.
5. For refined products, EPA assumed a specific weight of 1.0 to simplify the
calculation (i.e., no need to convert between mg/kg and mg/L).
' EIA: Petroleum Supply Annual 2000, Vol 1, Page 6 (EIA, 2001).
2-15
-------
Section 2 - Development of TRIReleases2007
Based on the above assumptions, EPA calculated the overall TWF using Equation 2-1
where the concentration of each of the 21 TRI PACs in each crude or finished petroleum product
is multiplied by its respective TWF. The concentration of each PAC in petroleum crude oil or
products is represented by the variable [m;]j and listed in Table 2-8 for products or Table 2-9 for
crude oils. The products in each product produced by U.S. refineries are represented by the
variable Q and listed in Table 2-10 while the respective TWFs are listed in Table 2-12. This
calculation resulted in an equivalent TWF for each type of product supplied to U.S refineries.
EPA then multiplied the mass of each type of product by the respective equivalent TWF and
summed this quantity for all products received by U.S. refineries. Dividing this sum by the total
sum of all products received by U.S. refineries resulted in an equivalent TWF for the PACs
present in wastewaters from U.S. refineries.
Overall TWF =
(Eqn. 2-1)
where:
Quantity of Product V Supplied to U.S. Refineries
Estimated concentration of PAC compound 'j' in Product T
This calculation resulted in a TWF value of 25.417. The TWPE of the combined mass of PACs
reported to TRI by petroleum refineries can then be calculated by multiplying the reported PAC
releases by 25.417'.
Table 2-8. PAC Concentrations in Petroleum Products
PAC Chemical Name
Benzo(a)anthracene
Benzo(a)phenanthrene
(chrysene)
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(j)fluoranthene
Benzo(k)fluoranthene
Benzo(j,k)fluorene
(fluoranthene)
Benzo(r,s,t)pentaphene
Dibenz(a,h)acridine
Dibenz(aj )acridine
Dibenzo(a,h)anthracene
Dibenzo(a,e)fluoranthene
Dibenzo(a,e)pyrene
Dibenzo(a,h)pyrene
Dibenzo(a,l)pyrene
7H-Dibenzo(e,g)carbazole
Gasoline
mg/L
4.30
2.00
1.80
NP
NP
NP
6.50
NP
NP
NP
NP
NP
NP
NP
NP
NP
Kerosene
ppm (wt/vol)
0.75
2.00
0.50
0.75
NP
0.50
4.00
NP
0.20
NP
0.75
NP
0.45
1.00
NP
NP
No. 2
Diesel
Fuels
Bunker C
No. 6 Oil
Paving
Asphalt
mg/L or mg/kg
0.80
3.40
NP
NP
NP
NP
2.80
NP
NP
NP
NP
NP
NP
NP
NP
NP
90.00
196.00
44.00
NP
NP
NP
240.00
NP
NP
NP
NP
NP
NP
NP
NP
NP
90.00
80.00
1.30
NP
NP
1.80
NP
NP
NP
NP
4.60
NP
NP
NP
NP
NP
Lube
Oil6
mg/kg
0.68
3.20
0.23
0.627
NP
NP
2.00
NP
NP
NP
NP
NP
NP
NP
NP
NP
2-16
-------
Section 2 - Development of TRIReleases2007
Table 2-8. PAC Concentrations in Petroleum Products
PAC Chemical Name
7, 12-Dimethylbenz(a)anthracene
Indeno(a,2,3 -cd)pyrene
3 -Methy Icholanthrene
5-Methylchrysene
1-Nitropyrene
Gasoline
mg/L
NP
NP
NP
NP
NP
Kerosene
ppm (wt/vol)
NP
2.00
0.10
NP
NP
No. 2
Diesel
Fuels
Bunker C
No. 6 Oil
Paving
Asphalt
mg/L or mg/kg
NP
NP
NP
6.00
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
Lube
Oil6
mg/kg
NP
NP
NP
NP
NP
Source: Data compiled in the American Petroleum Institute's Transport and Fate of non-BTEXPetroleum
Chemicals in Soil and Groundwater (API, 1994).
NP - Not present.
Table 2-9. PAC Concentrations in Crude Oils (mg/kg)
PAC Chemical Name
Benzo(a)anthracene
Benzo(a)phenanthrene (chrysene)
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(j)fluoranthene
Benzo(k)fluoranthene
Benzo(j,k)fluorene (fluoranthene)
Benzo(r,s,t)pentaphene
Dibenz(a,h)acridine
Dibenz(aj )acridine
Dibenzo(a,h)anthracene
Dibenzo(a,e)fluoranthene
Dibenzo(a,e)pyrene
Dibenzo(a,h)pyrene
Dibenzo(a,l)pyrene
7H-Dibenzo(c,g)carbazole
7, 12-Dimethylbenz(a)anthracene
Indeno(a,2,3 -cd)pyrene
3 -Methy Icholanthrene
5 -Methy Ichrysene
1-Nitropyrene
South Louisiana Crude Oil
1.7000
17.5600
0.7500
0.5000
0.9000
1.3000
5.0000
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
Alberta Crude Oil
NP
30.0000
NP
4.0000
NP
NP
6.0000
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
3.0000
NP
NP
Weighted Average
0.6645
25.1372
0.2932
2.6319
0.3518
0.5082
5.6091
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
1.8273
NP
NP
Source: Data compiled in the American Petroleum Institute's Transport and Fate of non-BTEX Petroleum
Chemicals in Soil and Groundwater (API, 1994).
NP - Not present.
2-17
-------
Section 2 - Development of TRIReleases2007
Table 2-10. Supply and Disposition of Crude Oil and Petroleum Products
Crude and Finished Petroleum
Products
Finished Motor Gasoline
Reformulated
Oxygenated
Other
Finished Aviation Gasoline
Jet Fuel
Naphtha-Type
Kerosene-Type
Kerosene
Distillate Fuel Oil
0. 05% Sulfur and under
Greater than 0. 05% sulfur
Residual Fuel Oil
Naphtha For Petroleum Feed Use
Other Oils For Petroleum Feed Use
Special Naphthas
Lubricants
Waxes
Petroleum Coke
Asphalt and Road Oil
Still Gas
Miscellaneous Products
Total Products
Crude Oil
TOTAL VOLUME OF
PRODUCTS & CRUDE OIL
1,000 bbl/year
2,910,056
939,493
42,221
1,928,342
6,543
587,974
75
587,899
23,860
1,310,158
905,064
405,094
254,843
74,039
71,762
21,868
65,687
6,478
266,107
192,223
241,365
19,957
6,052,920
5,514,395
11,567,315
% (Products Only)
48.08
NP
NP
NP
0.11
9.71
NP
NP
0.39
21.65
NP
NP
4.21
1.22
1.19
0.36
1.09
0.11
4.40
3.18
3.99
0.33
100
—
—
Volume % (Total)
25.16
NP
NP
NP
0.06
5.08
NP
NP
0.21
11.33
NP
NP
2.20
0.64
0.62
0.19
0.57
0.06
2.30
1.66
2.09
0.17
52.33
47.67
100
Source: Petroleum Supply Annual 2000,
NP - Not present.
Vol. 1, Page 6(EIA, 2001).
2-18
-------
Section 2 - Development of TRIReleases2007
Table 2-11. Products for Which PAC Composition Is Not Available
Product
Finished Aviation Gasoline
Jet Fuel
Miscellaneous Products
Naphtha For Petroleum Feed Use
Other Oils For Petroleum Feed Use
Petroleum Coke
Special Naphtha
Still Gas
Waxes
PAC Composition Taken from:
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Paving Asphalt
Gasoline
Gasoline
Lube Oil
Table 2-12. Calculation of Toxic Weighting Factor for Petroleum PACs
Pollutant
Benzo(a)anthracene
Benzo(a)phenanthrene (Chrysene)
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(j)fluoranthene
B enzo (k)fluoranthene
Benzo(j,k)fluorene (Fluoranthene)
Benzo(r,s,t)pentaphene
Dibenz(a,h)acridine
Dibenz(aj )acridine
Dibenzo(a,h)anthracene
Dibenzo(a,e)fluoranthene
Dibenzo(a,e)pyrene
Dibenzo(a,h)pyrene
Dibenzo(a,l)pyrene
7H-Dibenzo(c,g)carbazole
7, 12-Dimethylbenz(a)anthracene
Indeno( 1,2,3 -cd)pyrene
3 -Methy Icholanthrene
5 -Methy Ichrysene
1-Nitropyrene
TWF
36.26
31.01
100.66
30.66
NA
30.66
1.2847
NA
NA
NA
30.66
NA
NA
NA
NA
NA
NA
30.66
NA
NA
NA
Chemical Percentage (%)
17.47
46.29
4.17
2.74
0.36
0.70
24.32
NP
NP
NP
0.43
NP
NP
NP
NP
NP
NP
0.01
NP
3.50
NP
Total
Adjusted TWF
5.36
14.35
4.20
0.84
0.21
0.31
0.13
0.00
25.417
NA - Not applicable (No TWF Available).
NP - Not present.
2-19
-------
Section 2 - Development of TRIReleases2007
Wood Preserving PACs
EPA used data available from wood preserving facilities to calculate TWPE for
discharges of PACs from wood preserving facilities (NAICS 321114, Wood Preservation). In
2005, 10 wood preserving facilities participated in a sampling program to determine the PACs
released with their stormwater runoff. Over the period of a few months, the facilities collected
grab samples of runoff during rainfall events. The 10 facilities collected a total of 74 samples. In
37 of these samples, at least one PAC was measured above the detection limit. EPA used the
concentrations in these 37 samples to calculate a TWF for the PACs discharged from wood
preserving facilities (H.M. Rollins, 2005).
For all PAC concentrations reported as not detected, EPA assumed the concentration to
be zero. Using the data provided, EPA calculated the average concentration of the six PAC
compounds measured. EPA calculated the percentage of each compound relative to the total
PACs. EPA calculated an adjusted TWF for each compound by multiplying its chemical-specific
TWF by its percentage relative to the total PACs. EPA summed these values to calculate a new
overall TWF value for PACs discharged by facilities in the wood preserving NAICS code. Table
2-13 lists TWFs for all PACs, the percent of total PACs, and the adjusted TWF for each PAC.
Table 2-13. Calculation of Toxic Weighting Factor for Wood Preserving PACs
Chemical Name
Benzo(a)anthracene
Benzo(a)phenanthrene(chrysene)
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(j)fluoranthene
B enzo (k)fluoranthene
Benzo(j,k)fluorene(fluoranthene)
Benzo(r,s,t)pentaphene
Dibenz(a,h)acridine
Dibenz(aj )acridine
Dibenzo(a,h)anthracene
Dibenzo(a,e)fluoranthene
Dibenzo(a,e)pyrene
Dibenzo(a,h)pyrene
Dibenzo(a,l)pyrene
7H-Dibenzo(e,g)carbazole
7, 12-Dimethylbez(a)anthracene
Indeno(a,2,3 -cd)pyrene
3 -Methy Icholanthrene
5 -Methy Ichrysene
1-Nitropyrene
Toxic Weighting Factor
36.2600
31.0100
100.6600
30.6600
NA
30.6600
1.2847
NA
NA
NA
30.6600
NA
NA
NA
NA
NA
NA
30.6600
NA
NA
NA
Chemical Percentage (%)
6.73
9.73
0.49
4.98
NP
0.78
77.29
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
Total PACs TWF
Adjusted TWF
2.44
3.02
0.49
1.53
NP
0.24
0.989
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
8.33
NA - Not available.
NP - Not present.
2-20
-------
Section 2 - Development of TRIReleases2007
Pulp, Paper, and Paperboard PACs
EPA used data available from pulp and paper mills to calculate TWPE for discharges of
PACs from facilities in the Pulp, Paper, and Paperboard Category. The National Council of the
Paper Industry for Air and Stream Improvement (NCASI) has provided guidance to the pulp,
paper, and paperboard industry for PAC discharges (NCASI, 1988). The NCASI guidance
includes a table listing the concentrations of PACs found in wastewaters for several pulping
types (kraft, bisulfite, chemithermomechanical (CTMP), and thermal (TMP)). EPA determined
that in the United States, there are few bisulfite, CTMP, and TMP mills compared to the number
of kraft mills. Therefore, EPA used the kraft mill concentrations to calculate the PAC TWF.
Since the NCASI guidance does not distinguish between effluents from mills with or without
bleaching, the calculated TWF was used for mills in all pulp, paper, and paperboard ELG phases.
NCASI calculated the emission factors for the industry based on six PACs:
• Benzo(a)anthracene;
• Benzo(a)pyrene;
• Benzo(b+k) fluoranthene;
• Dibenzo(a,h)anthracene;
• Fluoranthene; and
• Indeno(l,2,3-c,d)pyrene.
For the kraft mills, only fluoranthene was detected above the method detection limit
(MDL); however, four of the other five compounds were detected above the MDL for the other
pulping types. Because the calculated TWF will be used for all facilities in the Pulp, Paper, and
Paperboard Category, EPA used l/2 the detection limit for compounds that were not detected in
kraft mill wastewaters. NCASI also calculated the emission factor using !/2 the detection limit for
compounds that were not detected.
EPA used the concentrations of six PACs to calculate a Pulp, Paper, and Paperboard
Category PAC TWF. EPA summed the measured concentrations to calculate the total
concentration of PACs in the effluent. EPA then calculated the percentage of each chemical
relative to the total PACs in the effluent. EPA calculated an adjusted TWF for each compound
by multiplying its chemical-specific TWF by its percentage relative to the total PACs. EPA
summed these values to calculate a new overall TWF value for PACs discharged by facilities in
the Pulp, Paper, and Paperboard Category. Table 2-14 presents the TWFs for the six PACs, the
percentage of total PACs, and the adjusted TWF for each PAC.
2-21
-------
Section 2 - Development of TRIReleases2007
Table 2-14. Calculation of Toxic Weighting Factor for Pulp, Paper, and Paperboard PACs
Chemical Name
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b+k)fluoranthene
Benzo(j,k)fluorene(fluoranthene)
Dibenzo(a,h)anthracene
Indeno( 1,2,3 -cd)pyrene
Toxic Weighting Factor
36.2600
100.6600
30.6600
1.2847
30.6600
30.6600
Chemical Percentage (%)
11.74
11.74
11.74
17.84
23.47
23.47
Total PACs TWF
Adjusted TWF
4.25
11.81
3.60
0.229
7.20
7.20
33.66
2.4.4 Metal Compounds
For TRI reporting, facilities report metal compounds on a single reporting form for each
parent metal and do not specify the individual compound(s) released. In addition, if the facility is
required to report for a metal (e.g., zinc) and its compounds (e.g., zinc compounds), the facility
may report both the metal and metal compound on a single form (reported as the metal
compound). For metal compound reporting, the release quantities are based on the mass of the
parent metal, only. To calculate TWPEs for metal compounds, EPA used the TWF for the parent
metal. EPA then combined the TWPEs for the metal and metal compounds for ranking purposes
(i.e., TWPE reported for "zinc and zinc compounds," rather than one TWPE for "zinc" and one
TWPE for "zinc compounds"). This analysis does not double count metal discharges because all
discharges are separated until the rankings are created. For example, if a facility reported 5
pounds of zinc and 10 pounds of zinc compounds, the discharges would be kept separate in the
database. When the rankings are created however, the database would display that the facility has
one entry of 15 pounds of "zinc and zinc compounds."
2.4.5 Determination of "Basis of Estimate " of Reported TRI Releases
When reporting releases and transfers to TRI, facilities also indicate the basis for their
estimate using six reporting codes:
Ml: continuous monitoring data or measurements;
M2: periodic or random monitoring data or measurements;
C: mass balance calculations, such as calculation of the amount of the toxic
chemical in streams entering and leaving process equipment;
E: published emission factors;
E2: site-specific emission factors; and
O: other approaches, such as engineering calculations.
EPA developed a table in TRICalculations2007 that contains the basis of estimate for
direct discharges and indirect discharges (i.e., transfers to POTWs). This table is separate from
the "TRI Master List" table. EPA used this table in TRIReleases2007 to summarize how releases
are reported for certain NAICS codes and point source categories.
2-22
-------
Section 2 - Development of TRIReleases2007
2.5 TRIReleases2007
As the final step in developing TRIReleases2007, EPA grouped discharges from the "TRI
Master List" table to create the point source category rankings and to perform other analyses.
The remainder of this subsection describes the development of TRIReleases2007 and discusses
preliminary results in the following order:
• Section 2.5.1 discusses the NAICS/Point Source Category Crosswalk; and
• Section 2.5.2 describes the development of the 2007 TRI rankings, including
analysis of facilities with the highest TWPE, pollutants with the highest TWPE,
and category prioritization.
Table 2-15 lists the database tables that EPA created in TRIReleases2007.
Table 2-15. Tables Created in TRIReleases2007
Table Name
"Counts of Facilities by
NAICS"
"Counts of Facilities by PSC"
"Point Source Rankings"
"NAICS Code Rankings"
Description
Includes counts of direct dischargers, indirect dischargers, facilities that discharge
both directly and indirectly, total dischargers, and total facilities reporting to TRI
by NAICS code.
Similar to table "Counts of Facilities by NAICS"; however, it reports the counts by
point source category.
Presents rankings for all point source categories based on calculated TWPEs.
TWPEs were calculated using the total discharges to surface water by direct
dischargers and transfers to POTWs by indirect dischargers, taking into account
pollutant removal occurring at the POTWs.
Presents rankings for all NAICS codes based on calculated TWPEs. TWPEs were
calculated using the total discharges to surface water by direct dischargers and
transfers to POTWs by indirect dischargers, taking into account pollutant removal
occurring at the POTW.
EPA also imported or linked the following tables from TRICalculatiom2007:
• "Dioxin Distributions";
"NAICS_Codes";
• "Parent Metals and Compounds";
• "Pesticide Chemical List";
"PSC Codes";
"PSC/NAICS Crosswalk";
"TRI Master List";
• "TRI Master Facility List"; and
"TWFs"
2-23
-------
Section 2 - Development of TRIReleases2007
2.5.1 NAICS/Point Source Category Crosswalk
EPA has developed ELGs for 56 specific categories of industrial dischargers. The
categories, which may be divided into subcategories, are generally defined in terms of
combinations of products made and the processes used to make these products. Facilities with
data in TRI are identified by NAICS code. Thus, to use TRI data to estimate the pollutants
discharged by each point source category, EPA assigned each 6-digit NAICS code to an
appropriate point source category using the "NAICS/Point Source Category Crosswalk" table.
Section 1 of this report discusses the crosswalk in more detail.
2.5.2 Development of 2007 TRI Rankings
Figure 2-2 presents the TRIReleases2007 database structure, including fields used from
each data source. The NAICS codes in the "TRI Master List" table are specific to each facility
and each discharge. This allows EPA to make NAICS adjustments to differentiate between
various operations at one facility. The default NAICS code is the primary facility NAICS code
reported in TRI. For the development of the rankings, EPA associated the NAICS codes with the
appropriate point source categories using the "NAICS/Point Source Category Crosswalk" and
the "Point Source Category Codes" tables. The TWPE for each discharge was calculated
previously in TRICalculations2007 (see Section 2.4).
TRIReleases2007 groups releases by chemical, facility, and point source category to
allow EPA to perform the following analyses.
TRI Master List
NAICS Code
TRI Facility ID
CAS Number
Chemical Name
Total Pounds Released
TWPE
Direct/Indirect Discharge Indicator
k
NAICS/Point Source Category Crosswalk
NAICS Code
Point Source Category Code
Point Source Category Codes
Point Source Category Code
Category Name
Type of Group
Figure 2-2. Basic Structure of the TRIReleases2007 Database
Top Facilities Analysis. EPA created a table that ranks facilities according to the TWPE
discharged by the entire facility. This table also identifies the chemical that contributed the
greatest amount of TWPE to the total facility TWPE. EPA used the table to identify facilities
with unusually high reported discharges relative to other facilities in an industrial category. As
discussed in Section 6, EPA contacted these facilities to learn more about their reported releases.
Section 6 also presents EPA's findings about the top facilities' reported releases.
Top Pollutants Analysis. EPA created a table that ranks pollutants discharged according
to the TWPE discharged by all facilities reporting in TRIReleases2007. The table also includes
the number of facilities that report releasing the chemical. Using this analysis, EPA identified
pollutants or pollutant categories for further analysis (e.g., metals).
Category Prioritization. EPA uses point source category rankings to identify categories
that may warrant further review.
2-24
-------
Section 2 - Development of TRIReleases2007
2.6 Results of the Preliminary Analysis of the TRIReleases2007 Database
This section presents the results of the analysis of TRIReleases2007 database. Table 2-16
presents the point source category rankings by TWPE. Table A-l in Appendix A presents the
six-digit NAICS code rankings by TWPE. Table A-2 in Appendix A presents the total TWPE for
chemicals in TRI. See Section 5 of the 2009 TSD for EPA's 2009 annual review (U.S. EPA,
2009).
Table 2-16. Point Source Category Rankings
40CFR
Part
414.1a
414
423
430
419
420
433
415
440
421
432
458
455
429
417
NA
471
463
439
428
425
469
NA
464
468
NA
418
437
413
Point Source Category
Chlorine And Chlorinated Hydrocarbons
Organic Chemicals, Plastics And Synthetic
Fibers
Steam Electric Power Generating
Pulp, Paper And Paperboard
Petroleum Refining
Iron And Steel Manufacturing
Metal Finishing
Inorganic Chemicals Manufacturing
Ore Mining And Dressing
Nonferrous Metals Manufacturing
Meat and Poultry Products
Carbon Black Manufacturing
Pesticide Chemicals
Timber Products Processing
Soap And Detergent Manufacturing
National Security & International Affairs
Nonferrous Metals Forming And Metal
Powders
Plastics Molding And Forming
Pharmaceutical Manufacturing
Rubber Manufacturing
Leather Tanning And Finishing
Electrical And Electronic Components
Miscellaneous Foods And Beverages
Metal Molding And Casting (Foundries)
Copper forming
Tobacco Products
Fertilizer Manufacturing
Centralized Waste Treatment
Electroplating
Number of
Facilities
28
594
284
198
232
190
2047
142
28
107
144
7
67
107
58
43
105
121
96
182
19
87
133
184
116
21
29
34
352
Total
Discharge
before POTW
Removal
1,500,000
72,500,000
2,160,000
34,900,000
16,600,000
41,500,000
25,800,000
26,900,000
324,000
3,560,000
45,100,000
356
2,250,000
210,000
675,000
15,000,000
12,200,000
15,000,000
5,750,000
1,880,000
634,000
11,300,000
9,520,000
1,690,000
288,000
203,000
3,240,000
2,340,000
8,670,000
Total
Pounds
Released b
835,000
19,200,000
2,150,000
15,800,000
13,700,000
39,500,000
3,980,000
5,870,000
319,000
2,670,000
41,400,000
356
1,450,000
32,500
69,300
14,900,000
1,330,000
2,140,000
1,510,000
865,000
318,000
3,210,000
5,810,000
204,000
35,500
189,000
3,190,000
448,000
886,000
TWPE
7,270,000
575,000
542,000
460,000
172,000
104,000
62,000
54,700
44,400
38,900
35,900
32,400
24,700
16,300
14,600
14,500
8,830
8,780
8,000
7,860
7,800
7,550
6,580
6,110
4,950
4,760
4,460
3,790
3,210
2-25
-------
Section 2 - Development of TRIReleases2007
Table 2-16. Point Source Category Rankings
40CFR
Part
407
467
436
405
410
406
461
438
426
434
411
424
422
443
465
408
466
446
NA
445
NA
454
444
NA
409
447
457
NA
NA
NA
NA
NA
NA
NA
NA
NA
Point Source Category
Canned And Preserved Fruits And Vegetables
Processing
Aluminum forming
Mineral Mining And Processing
Dairy products processing
Textile Mills
Grain mills
Battery Manufacturing
Metal Products And Machinery
Glass Manufacturing
Coal Mining
Cement Manufacturing
Ferroalloy Manufacturing
Phosphate Manufacturing
Paving And Roofing Materials (Tars And
Asphalt)
Coil Coating
Canned And Preserved Seafood Processing
Porcelain Enameling
Paint Formulating
Printing & Publishing
Landfills
Justice, Public Order, & Safety
Gum And Wood Chemicals Manufacturing
Waste Combustors
Independent And Stand Alone Labs
Sugar Processing
Ink Formulating
Explosives Manufacturing
Apparel & Other Textile Products
Miscellaneous Retail
Wholesale Trade- Nondurable Goods
Wholesale Trade- Durable Goods
Coal Mining
Engineering & Management Services
Business Services
Drinking Water Treatment
Trucking & Warehousing
Number of
Facilities
20
115
60
243
63
23
62
32
64
14
36
4
11
19
50
8
5
49
65
13
1
10
8
7
o
J
8
9
2
1
1
5
1
1
2
2
1
Total
Discharge
before POTW
Removal
4,370,000
2,000,000
2,410,000
20,700,000
2,830,000
10,700,000
1,180,000
116,000
1,510,000
245,000
27,900
2,350
16,200
1,330
67,300
312,000
3,430
1,130,000
370,000
69,500
31.2
3,020
18,300
9,660
72,900
4,500
17,300
6,710
7
44,600
2,990
16.6
720
95
681
66
Total
Pounds
Released b
3,760,000
304,000
1,800,000
3,170,000
1,170,000
1,800,000
120,000
15,700
185,000
245,000
3,410
2,300
16,100
227
21,600
312,000
2,180
91,500
31,800
22,400
31.2
507
18,300
2,930
23,700
573
16,200
4,090
1.58
4,460
307
16.6
371
9.46
171
40.3
TWPE
2,960
2,710
2,420
2,400
2,390
2,080
1,640
917
546
493
452
340
250
249
241
234
164
140
110
82.7
69.9
54.8
39.6
30.0
25.5
20.0
13.6
4.61
3.54
3.33
2.51
0.458
0.441
0.294
0.29
0.0447
2-26
-------
Section 2 - Development of TRIReleases2007
Table 2-16. Point Source Category Rankings
40CFR
Part
NA
NA
Point Source Category
Misc. Manuf. Industries
Food & Kindred Products
Number of
Facilities
1
1
Total
Discharge
before POTW
Removal
5
0.004
Total
Pounds
Released b
5
0.000784
TWPE
0.0281
0.00013
Source: TRIReleases2007_v2.
a 414.1 refers to the chlorinated hydrocarbon segment of the Organic Chemicals, Plastics, and Synthetic Fibers
Category (40 CFR Part 414) and the Chlor-Alkali Subcategory of the Inorganic Chemicals Manufacturing Category
(40 CFR Part 415).
b Discharges include transfers to POTWs and account for POTW removals.
NA - Not applicable; no existing ELGs apply to discharges.
2.6.1 Metals Analysis
For the 2009 screening-level analysis, EPA gave special consideration to reported
discharges of metals. Releases of metals from industrial facilities may be associated with current
operations or may be from cleanup actions for past practices. If releases are not related to current
operations, they are not useful in reviewing the ELGs intended to control discharges from current
operations. EPA identified the following metals for further analysis as part of the 2009 annual
review, based on total TWPE calculated by TRIReleases2007:
• Manganese;
• Arsenic; and
• Copper.
Table 2-17 presents all the metals reported in TRI 2007 ranked by TWPE, including the number
of facilities reporting discharges and the pounds discharged. In 2007, 4,428 facilities reported
discharging 17 metals5. The total metals discharges after accounting for POTW removals, as
appropriate, was 1,240,000 TWPE, which represented 12.9 percent of total TRI TWPE for 2007.
Manganese discharges were the largest metals discharges, as measured by TWPE, accounting for
almost 25 percent of the total metals TWPE. Arsenic and copper were also significant
contributors, with discharges of each accounting for greater than 18 percent of the total metals
TWPE.
5 TRI 2007 includes release information for the following metals, including their metal compounds: antimony,
arsenic, barium, beryllium, cadmium, chromium, cobalt, copper, lead, manganese, mercury, nickel, selenium, silver,
thallium, vanadium, and zinc.
2-27
-------
Section 2 - Development of TRIReleases2007
Table 2-17. Metals Discharged by TWPE in TRI 2007
Chemical Name
Manganese and Manganese Compounds
Arsenic and Arsenic Compounds
Copper and Copper Compounds
Lead and Lead Compounds
Mercury and Mercury Compounds
Zinc and Zinc Compounds
Selenium and Selenium Compounds
Silver and Silver Compounds
Cadmium and Cadmium Compounds
Nickel and Nickel Compounds
Cobalt and Cobalt Compounds
Vanadium and Vanadium Compounds
Chromium and Chromium Compounds
Barium and Barium Compounds
Thallium and Thallium Compounds
Antimony and Antimony Compounds
Beryllium and Beryllium Compounds
Total
Number
of
Facilities
965
145
1,688
2,401
331
1,297
49
42
46
1,368
204
148
1,144
353
10
163
10
4,428
Total
Pounds
Released a
4,330,000
55,800
354,000
77,000
790
865,000
32,000
2,170
1,420
254,000
194,000
376,000
74,600
860,000
1,630
19,600
89.9
7,490,000
TWPE after
POTW
Removals
(Ib-eq/yr)
305,000
226,000
225,000
173,000
92,500
40,600
35,800
35,700
32,900
27,600
22,100
13,200
5,650
1,710
1,680
240
95.0
1,240,000
TWPE
percent of
Total Metals
TWPE
24.7
18.2
18.1
13.9
7.48
3.28
2.89
2.88
2.66
2.23
1.79
1.06
0.46
0.14
0.14
0.02
0.008
100
TWPE
percent of
Nationwide
TWPE
3.19
2.36
2.35
1.80
0.97
0.42
0.37
0.37
0.34
0.29
0.23
0.14
0.06
0.02
0.02
0.003
0.001
12.9
Source: TRIReleases2007_v2.
a Discharges include transfers to POTWs and account for POTW removals.
Table 2-18 lists the facilities reporting discharges of greater than 6,000 TWPE of
manganese, arsenic, or copper in TRI 2007.
Manganese is commonly found in discharges from pulp and paper facilities. All but one
of the top facilities (Tronox, LLC) discharging manganese are pulp and paper mills. In the Final
Report on the Pulp, Paper, and Paperboard Detailed Study (U.S. EPA, 2006), EPA identified
manganese and aluminum as the top metals of concern from pulp and paper mills. EPA reviewed
manganese discharges from Tronox LLC as part of the 2009 annual review (Freeze, 2009a;
Freeze, 2009b) and determined that manganese in the discharges from Tronox LLC originated in
the titanium dioxide manufacturing process onsite known as the chloride process (see Section 7
of the Preliminary 2010 Effluent Guidelines Program Plan Technical Support Document (U.S.
EPA, 2009)). Of the 17 metals reported to TRI, manganese is ranked 14th in terms of relative
toxicity. Manganese TWPE discharges contributed 24.7 percent to the total metals TWPE to TRI
in 2007.
2-28
-------
Section 2 - Development of TRIReleases2007
Table 2-18. Facilities Reporting Discharges of Metals with the Highest TWPE in TRI 2007
Chemical
Manganese
and
Manganese
Compounds
Arsenic and
Arsenic
Compounds
Facility
Domtar Industries Inc. Ashdown Mill
Rayonier Performance Fibers Jesup Mill
Alabama River Pulp Co In C.
Tronox LLC
Brunswick Cellulose Inc
Georgia-Pacific Crossett Operations
Meadwestvaco Texas L.P.
Georgia-Pacific Consumer Products LP
Bowater Inc -Catawba Operations
Georgia-Pacific Consumer Products LLC
Georgia-Pacific Brewton LLC
Chesterfield Power Station
U.S. TVA Johnsonville Fossil Plant
U.S. TVA Widows Creek Fossil Plant
Duke Energy Corp Wabash River Generating Station
U.S. TVA Kingston Fossil Plant
Detroit Edison Monroe Power Plant
Gaston Steam Plant
Kentucky Utilities Co. - E. W. Brown Station
Kentucky Utilities Co Ghent Station
Eastman Chemical Co Tennessee Operations
American Electric Power Cardinal Plant
Barry Steam Plant
Cliffside Steam Station
Point Source Category
Pulp, Paper and Paperboard
Pulp, Paper and Paperboard
Pulp, Paper and Paperboard
Inorganic Chemicals Manufacturing
Pulp, Paper and Paperboard
Pulp, Paper and Paperboard
Pulp, Paper and Paperboard
Pulp, Paper and Paperboard
Pulp, Paper and Paperboard
Pulp, Paper and Paperboard
Pulp, Paper and Paperboard
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Organic Chemicals, Plastics and Synthetic
Fibers
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
CFR Citation
40 CFR Part 430
40 CFR Part 430
40 CFR Part 430
40 CFR Part 415
40 CFR Part 430
40 CFR Part 430
40 CFR Part 430
40 CFR Part 430
40 CFR Part 430
40 CFR Part 430
40 CFR Part 430
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 4 14
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
Total TWPE (Ib-
eq/yr)
15,700
8,450
8,450
8,110
7,750
7,350
7,160
6,830
6,740
6,470
6,000
18,200
17,000
15,800
13,700
10,900
9,700
9,300
9,110
8,850
8,810
8,490
8,080
6,540
2-29
-------
Section 2 - Development of TRIReleases2007
Table 2-18. Facilities Reporting Discharges of Metals with the Highest TWPE in TRI 2007
Chemical
Copper and
Copper
Compounds
Facility
Great River Energy Stanton Station
Georgia Power Scherer Steam Electric Generating
Plant
Chesterfield Power Station
Bowen Steam Electric Generating Plant
U.S. TVA Paradise Fossil Plant
American Electric Power Kammer / Mitchell Plants
Point Source Category
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
Steam Electric Power Generating
CFR Citation
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
40 CFR Part 423
Total TWPE (Ib-
eq/yr)
12,700
10,800
10,200
7,620
6,980
6,380
Source: TRIReleases2007 v2.
2-30
-------
Section 2 - Development of TRIReleases2007
Arsenic and copper are commonly found in discharges from steam electric power
generating facilities. In EPA's analysis of pollutants found in discharges from the steam electric
industry, EPA found that arsenic and copper are two of the top pollutants reported to PCS/ICIS-
NPDES and TRI. All but one of the top facilities (Eastman Chemical Co Tennessee Operations)
discharging arsenic are steam electric power generating facilities. Of the 17 metals reported to
TRI, arsenic is ranked 4th in terms of relative toxicity. Arsenic TWPE discharges contributed
18.2 percent to the total metals TWPE. All of the top facilities discharging copper are steam
electric power generating facilities. Of the 17 metals reported to TRI, copper is ranked 9th in
terms of relative toxicity. Copper TWPE discharges contributed 18.1 percent to the total metals
TWPE.
Conclusions
• A total of 4,428 facilities reported discharging 7,490,000 pounds, including
transfers to POTWs and accounting for POTW removals, and 1,240,000 TWPE of
metals, accounting for 12.9 percent of total nationwide TRI 2007 TWPE.
• Manganese, arsenic, and copper are the metals with the highest TWPE
contributions to the total nationwide metals TRI 2007 TWPE.
• All but one of the top 11 facilities discharging manganese (ranked by TWPE) are
pulp and paper mills.
• All but one of the top 13 facilities discharging arsenic (ranked by TWPE) are
steam electric power generating facilities.
• All of the top 6 facilities discharging copper (ranked by TWPE) are steam electric
power generating facilities.
2.7 Development of TRIReleases2007 References
1. American Petroleum Institute (API). 1994. Transport and Fate of non-BTEX Petroleum
Chemicals in Soil and Groundwater - Appendix A. API Publication Number 4593.
(September). EPA-HQ-OW-2003-0074-0312.
2. Bicknell, Betsy. 2004. Calculation of TWF & TWPE for Creosote from Wood Preserving
Plants. (August). EPA-HQ-OW-2003-0074-123 8.
3. Codding, Ellie and Deb Bartram. ERG. 2005. Memorandum to 2006 Effluent Guidelines
Program Plan Docket, EPA Docket Number OW-2004-0032. RE: Publicly Owned
Treatment Works (POTW) Percent Removals Used for the TRIReleases2002 Database.
(August 12). EPA-HQ-OW-2004-0032-0967.
4. Eastern Research Group, Inc. (ERG). 2005. Draft Toxic Weighting Factor Development
in Support of the 304(m) Planning Process. (July 29). EPA-HQ-OW-2004-0032-0857.
5. Energy Information Administration (EIA). Office of Oil and Gas. 2001. Petroleum
Supply Annual 2000, Vol. 1, Page 6. (June). EPA-HQ-OW-2004-0032-0242.
6. Finseth, Thomas. 2005. TRI 2002 PACs TWF for Petroleum Refining, Creosote, and
Wood Preservers. (January). DCN 01552.
2-31
-------
Section 2 - Development of TRIReleases2007
1. Freeze, Terry. 2009a. Notes from Email Correspondence between Chris Krejci, Eastern
Research Group, Inc., and Terry Freeze, Tronox Hamilton. (March 23). EPA-HQ-OW-
2008-0517 DCN 06406.
8. Freeze, Terry. 2009b. Notes from Telephone Communication between Chris Krejci,
Eastern Research Group, Inc., and Terry Freeze, Tronox Hamilton. RE: Verification of
Magnitude and Basis of Estimate for Manganese Discharges Reported to TRI. (March
12). EPA-HQ-OW-2008-0517 DCN 6405.
9. H.C. Lavallee, Inc., 2005. Table 5 PAC Concentrations in Pulp Mill Effluents, Excerpt
from NCASI's SARA Handbook, Enclosure 10. (April). EPA-HQ-OW-2004-0032-0498.
10. H.M Rollins. 2005. Letter to Carey Johnston, U.S. EPA. RE: Creosote Wood Treating
Industry Storm Water Runoff Study Conducted on Behalf of the Southern Pressure
Treaters Association and Creosote Council III. (May 26). EPA-HQ-OW-2004-0032-
0301.
11. H.M. Rollins Company, Inc. 2005. Study ofPolycyclic Aromatic Compounds in Storm
Water Runoff. Prepared for: Southern Pressure Treaters Association and Creosote
Council III. Gulfport, MS. (April 29). EPA-HQ-OW-2004-0032-0294.
12. Johnston, Carey. USEPA. 2004. Memorandum to Public Record for the Effluent
Guidelines Program Plan for 2006/2007. RE: November 18, 2004, Meeting Between EPA
and Wood Preserving Industry Representatives. Washington, D.C. (November 18). EPA-
HQ-OW-2004-0032-0306.
13. Kashyap, Vikram. 2009. Notes from Telephone Communication and E-mail
Correspondence between Vikram Kashyap, Louisiana Pigments, and Eleanor Codding,
Eastern Research Group, Inc. RE: 2007 TRI Dioxin Water Releases. (May 8). EPA-HQ-
OW-2008-0517 DCN 06849.
14. Moeller, Greg. 2009. Notes from Telephone Communication and E-mail Correspondence
between Greg Moeller, Eastman Kodak, and Christopher Krejci, Eastern Research Group,
Inc. RE: Basis of Estimate for TRI Discharges Reported to EPA. (March 16). EPA-HQ-
OW-2008-0517 DCN 06409.
15. National Council of the Paper Industry for Air and Stream Improvement (NCASI). 1988.
Handbook of Chemical-Specific Information for SARA Section 313 Form R Reporting.
Table 5 PAC Concentrations in Pulp Mill Effluents. EPA-HQ-OW-2004-0032-0498.
16. U.S. EPA. 2000. Guidance for Reporting Toxic Chemicals Within the Dioxins and
Dioxin-Like Compounds Category. Washington, D.C. (December). EPA-HQ-OW-2003-
0074.
17. U.S. EPA. 2001. The Emergency Planning and Community Right-to-Know Act: Section
313 Release and Other Waste Management Reporting Requirements. EPA 260-K-01-001.
Washington, D.C. (February). EPA-HQ-OW-2008-0517 DCN 01996.
2-32
-------
Section 2 - Development of TRIReleases2007
18. U. S. EPA. 2004. Technical Support Document for the 2004 Effluent Guidelines Program
Plan. EPA-821-R-04-014. Washington, DC. (August). EPA-HQ-OW-2003-0074-1346
through 1352.
19. U.S. EPA. 2006. Final Report: Pulp, Paper, and Paperboard Detailed Study.
Washington, D.C. (November). EPA-821-R-06-016. EPA-HQ-OW-2004-0032-2249.
20. U. S. EPA. 2009. Preliminary 2010 Effluent Guidelines Program Plan Technical Support
Document. Washington, D.C. EPA-821-R-09-006. (October). EPA-HQ-OW-2008-0517
DCN 06703.
21. Woodruff, Al. 2007. Notes from Telephone Communication and E-mail Correspondence
between Al Woodruff, Cahaba Pressure Treated Forest Products, Inc., and Christopher
Krejci, Eastern Research Group, Inc. RE: Dioxin Distribution Reported at Cahaba
Pressure Treated Forest Products. (December 12). EPA-HQ-OW-2006-0771-1165.
22. World Health Organization (WHO). International Agency for Research on Cancer
(IARC). 1998. Volume 35 Polynuclear Aromatic Compounds, Part 4, Bitumens, Coal-
tars and Derived Products, Shale-oils and Soots. "Summary of Data Reported and
Evaluation." (April 20). EPA-HQ-OW-2004-0032-2394.
23. Zweig, Greg. 2000. DuPont Memphis Dioxin Analytical Data. (February). EPA-HQ-OW-
2004-0032-1151.
2-33
-------
Section 3 - DMRLoads2007: Development and Category Rankings
3. DMRLOADS2007: DEVELOPMENT AND CATEGORY RANKINGS
As discussed in Section 1, EPA annually reviews promulgated effluent limitations
guidelines and pretreatment standards (ELGs) and investigates available information on
industrial pollutant discharges. EPA identified that the discharge monitoring reports (DMRs) in
the Permit Compliance System (PCS) and the Integrated Compliance Information System for the
National Pollutant Discharge Elimination System (ICIS-NPDES) contain readily available and
relevant data on industrial pollutant discharges to surface waters ("direct discharges"). Neither
PCS nor ICIS-NPDES has information on pollutant discharges to Publicly Owned Treatment
Works (POTWs) ("indirect discharges"). Consequently, EPA was able to use DMR data in PCS
and ICIS-NPDES for its review of: (1) promulgated effluent guidelines ("direct discharges"); and
(2) direct industrial pollutant discharges not currently subject to effluent guidelines. Due to the
limitations of PCS and ICIS-NPDES, EPA was unable to use these two data systems to review
promulgated pretreatment standards or indirect industrial pollutant discharges not currently
subject to pretreatment standards. As discussed in Section 7, EPA combined the toxic-weighted
pound equivalent (TWPE) calculated from the DMR data contained in PCS and ICIS-NPDES
and Toxics Release Inventory (see Section 2 for information about TRI). EPA used this
combined TWPE to prioritize its review of industry sectors to offer the greatest potential for
reducing hazard to human health or the environment.
This section describes how EPA compiled DMR data from PCS and ICIS-NPDES into
the database DMRLoads2007 to estimate the mass and relative toxicity of pollutants discharged
by industry categories. DMRLoads2007 compiles information for all facilities classified as major
dischargers in PCS and ICIS-NPDES for reporting year 2007 and for the point source categories
that these facilities represent. Tables B-l and B-2 in Appendix B list annual loads and TWPE
calculated by DMRLoads2007 presented by 4-digit Standard Industrial Classification (SIC) code
and pollutant, respectively. The remainder of Section 3 is organized in the following subsections:
• Section 3.1- Overview ofDMRLoads2007;
• Section 3.2 - DMRLoads2007: Database Development and Methodology;
• Section 3.3 - Results of the Preliminary Analysis;
• Section 3.4 - Data Quality Review; and
• Section 3.5- DMRLoads2007 References.
3.1 Overview of DMRLoads2007
EPA's Office of Enforcement and Compliance Assistance (OECA) stores DMR data in
national databases. EPA has used these DMR data as a part of its screening level review of
existing effluent guidelines since the 2003 annual review (68 FR 75515, December 31, 2003).
Historically, OECA stored DMR data in PCS, but in 2006 began storing certain states' data in
ICIS-NPDES. Therefore the 2009 annual review of nationwide discharges required two sets of
data, which EPA merged to create DMRLoads2007'.
3.1.1 NPDES Permitting and Reporting Requirements
As authorized by the CWA, the NPDES program controls water pollution by regulating
point sources that discharge pollutants directly into waters of the United States. Specifically,
Federal Water Pollution Control Act Title IV, Permits and Licenses, of the Federal Water
Pollution Control Act created the NPDES system for permitting wastewater discharges (CWA
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Section 402). The Water Permits Division within EPA's Office of Wastewater Management
leads and manages the NPDES permit program in partnership with EPA Regional Offices, states,
tribes, and other stakeholders. Industrial, municipal, and other facilities must obtain NPDES
permits if they discharge directly to surface waters. In most cases, authorized states administer
the NPDES permit program.
More than 65,000 industrial facilities and municipal wastewater treatment plants have
obtained permits for discharges of regulated pollutants. Permitted facilities are required to file
DMRs that include permit monitoring data (e.g., pollutant concentration/quantity, flow) to the
appropriate regulating authority. The majority of NPDES permits are issued to direct point
source dischargers (i.e., those entities that discharge directly into the receiving water body). PCS
and ICIS-NPDES contain only limited data for indirect dischargers (i.e., those entities that
discharge to POTWs).
To provide an initial framework for setting permit issuance priorities, EPA developed a
major/minor classification system for industrial and municipal wastewater dischargers. Facilities
are classified as major based on an assessment of six characteristics:
1. Toxic pollutant potential;
2. Flow/stream flow volume;
3. Conventional pollutant loading;
4. Public health impact;
5. Water quality factors; and
6. Proximity to coastal waters.
Each permitting authority establishes its own specific definitions based on the above
characteristics, but major dischargers have the capability to impact receiving waters if not
controlled, and, therefore, receive more regulatory attention than minor dischargers. DMR data
for approximately 6,600 major facilities are in PCS and ICIS-NPDES for 2007.
Facilities with major discharges must demonstrate compliance with NPDES permit limits
by submitting monthly DMRs to the permitting authority. The permitting authority enters the
reported DMR data into PCS or ICIS-NPDES, including the type of violation (if any), measured
concentration and quantity values, and Quarterly Non-Compliance Report indicators. EPA does
not require permitting authorities to enter DMR data for minor dischargers into PCS and ICIS-
NPDES. Therefore, these databases contain only very limited data for minor dischargers.
Table 3-1 identifies states and territories with data in PCS versus ICIS-NPDES at the
time DMRLoads2007 was created. Note that three states were in the process of migrating from
PCS to ICIS-NPDES in 2007 and had data in both systems. EPA created the database
DMRLoads2007 to combine the two systems (PCS and ICIS-NPDES) and generate industrial
category rankings for all U.S. states and territories.
3-2
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-1. States and Territories Included in DMRLoads2007
State/Province a
Alabama
Alaska
American Samoa
Arizona
Arkansas °
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Guam
Hawaii
Idaho
Illinois b
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Missouri
Northern Mariana Islands
Database
PCS
ICIS-NPDES
ICIS-NPDES
PCS
PCS
PCS
PCS
ICIS-NPDES
PCS
ICIS-NPDES
PCS
ICIS-NPDES
ICIS-NPDES
ICIS-NPDES
ICIS-NPDES
ICIS-NPDES /PCS
ICIS-NPDES
PCS
PCS
PCS
PCS
PCS
ICIS-NPDES
ICIS-NPDES
PCS
PCS
PCS
ICIS-NPDES
State/Territory a
Montana
Mississippi
North Dakota
Nebraska13
North Carolina
New Hampshire
New Jersey
New Mexico
New York
Nevada
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Puerto Rico
Tennessee
Texas
Utah
Vermont
Virgin Islands of the U.S.
Virginia
Washington
Wisconsin
West Virginia
Wyoming
Database
ICIS-NPDES
PCS
PCS
ICIS-NPDES / PCS
PCS
ICIS-NPDES
PCS
ICIS-NPDES
ICIS-NPDES
ICIS-NPDES
PCS
PCS
PCS
ICIS-NPDES
ICIS-NPDES
PCS
ICIS-NPDES
ICIS-NPDES
PCS
PCS
ICIS-NPDES
PCS
ICIS-NPDES
PCS
PCS
PCS
PCS
PCS
a 2007 DMR data were not available for the following territories/tribes and were not included in DMRLoads2007:
Atlantic Offshore, Canal Zone, Federal Micronesia, George's bank, Gulf of Mexico East, Johnson Atoll, Marshall
Islands, Midway Islands, Palau, Saint Regis Tribe, Trust Territory, and U.S. Minor Islands.
b Indicates states that were in the process of migrating from PCS to ICIS-NPDES in the year 2007. Some facilities in
these states have DMR data in either PCS or ICIS-NPDES, while some facilities have DMR data in both PCS and
ICIS-NPDES. For facilities with data in both databases for 2007, EPA used the DMR data from ICIS-NPDES (see
Section 3.2.4.1).
c Indicates states that had DMR data in PCS for 2007 and have since migrated their DMR data to ICIS-NPDES.
3.1.2 Overview of PCS and ICIS-NPDES
Both PCS and ICIS-NPDES automate entering, updating, and retrieving NPDES data and
tracking permit issuance, permit limits, monitoring data, and other data pertaining to facilities
regulated by the NPDES program. Major dischargers are required to submit effluent monitoring
data to the permitting authority on DMR. The permitting authority then enters these data into
PCS or ICIS-NPDES and evaluates them for compliance with the NPDES permit requirements.
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Permit limits include water quality parameters (e.g., dissolved oxygen and temperature), specific
chemicals (e.g., phenol), bulk parameters (e.g., biochemical oxygen demand), and flow.
Facilities report pollutant discharges in their DMR as mass-based quantities and/or
concentrations using a wide variety of units. PCS and ICIS-NPDES also include information on
the facility's permit requirements, such as monitoring frequency.
3.1.2.1 Utility of PCS and ICIS-NPDES
The data stored in PCS and ICIS-NPDES are particularly useful for the annual review
process for the following reasons:
• PCS and ICIS-NPDES are national in scope, including data from all 50 states and
19 U.S. territories/tribes6;
• Discharge reports included in PCS and ICIS-NPDES are based on effluent
chemical analysis and metered flow;
• PCS and ICIS-NPDES collectively include direct discharging facilities in all point
source categories; and
• PCS and ICIS-NPDES include data on conventional pollutants for most facilities
and for the nutrients nitrogen and phosphorous for many facilities.
3.1.2.2 Limitations of PCS and ICIS-NPDES
Limitations of the data collected in the PCS and ICIS-NPDES databases include the
following:
• The databases contain data only for pollutants a facility is required by permit to
monitor; the facility is not required to monitor or report all pollutants actually
discharged.
• The databases include very limited discharge monitoring data from minor
dischargers.
• The databases include very limited data characterizing indirect discharges from
industrial facilities to POTWs.
• Many of the pollutant parameters included in PCS and ICIS-NPDES are not
chemical compounds (e.g., "total Kjeldahl Nitrogen," "oil and grease") and
cannot have toxic weighting factors (TWFs).
• In some cases, the databases identify the type of wastewater being discharged;
however, most reported flow rates do not indicate the type of wastewater and
therefore, total flow rates reported to PCS and ICIS-NPDES may include
stormwater and noncontact cooling water, as well as process wastewater.
• Facilities are identified by SIC code, not point source category. For some SIC
codes, it may be difficult or impossible to identify the point source category that
is the source of the reported wastewater discharges7.
6 The following territories' data/regions are not available in PCS or ICIS-NPDES for 2007: Atlantic Offshore, Canal
Zone, Federal Micronesia, George's Bank, Gulf of Mexico East, Johnson Atoll, Marshall Islands, Midway Islands,
Palau, Saint Regis Tribe, Trust Territory, and U.S. Minor Islands.
7 ICIS-NPDES includes a data field for entering the applicable ELG (i.e. 40 CFR Part 423 for the Steam Electric
Power Generating Category). However, entering data into this field is not required and therefore this field is
typically not populated.
3-4
-------
Section 3 - DMRLoads2007: Development and Category Rankings
• Some facilities in PCS and ICIS-NPDES do not provide information on
applicable SIC codes. Additionally, facilities in PCS do not provide information
on applicable North American Industrial Classification System (NAICS) codes,
while only a few facilities in ICIS-NPDES provide information on applicable
NAICS codes.
• Although facilities may provide more than one SIC code to describe their
operations, EPA uses only the primary SIC code to classify facilities.
• PCS and ICIS-NPDES were designed as permit compliance tracking systems and
do not contain production information.
• DMR data may be entered into the PCS or ICIS-NPDES database manually,
which leads to data-entry errors.
Despite the limitations and constraints of the PCS and ICIS-NPDES databases, EPA has
determined that they are appropriate for an initial screening-level review and prioritization of the
pollutant loads discharged by industrial categories. EPA will further evaluate the prioritized
categories in a second level of review, which may include additional data collection and
additional verification of data reported in PCS and ICIS-NPDES.
3.1.3 PCS and ICIS-NPDES Data Structure
The PCS and ICIS-NPDES databases contain more than 5 million records organized by
individual permit files. Each permit file contains information about the following elements:
• The permit and the permitted facility, including permit number, dates of issue and
expiration, facility name, location, and type of facility;
• Permit events, including date application was received, scheduled, and achieved
dates for completion of compliance schedule;
• Identity of outfalls within the facility and a description of the associated
monitoring requirements;
• Parameters to be measured at each outfall and the corresponding limitations; and
• Inspections performed at the facility, such as type of inspection, inspector
identity, and inspector comments.
To develop DMRLoads2007, EPA developed two pollutant loading tools: the database
PCSLoadCalculator2007 and the ICIS-NPDES Pollutant Loading Tool. These loading tools start
with DMR data stored in their respective databases and use similar methodologies to calculate
annual mass discharges from DMR data. PCS and ICIS-NPDES store data in a series of tables.
Table 3-2 lists the PCS and ICIS-NPDES data types that EPA used to create DMRLoads2007.
EPA uses data in the Permit Facility, Pipe Schedule, Measurement Violation, and Permit Event
data types to develop PCSLoadCalculator2007'. In ICIS-NPDES, these types of data are stored in
several relational database tables, also shown in Table 3-2. In addition to the four data types used
to develop PCSLoadCalculator2007, EPA also used data in the Parameter Limits data type to
develop the ICIS-NPDES Pollutant Loading Tool. EPA did not use the following nine data types
in PCS and ICIS-NPDES in developing its load calculators:
• Compliance schedule data;
• Enforcement action data;
• Evidentiary hearing data;
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-2. Data Types in PCS and ICIS-NPDES Used for DMRLoads2007 Development
PCS Data Type
Permit-Facility
Data
Pipe-Schedule
Data
Parameter-Limits
Data
Measurement-
Violation Data
Permit Events
Data
Description
General descriptive information on each permitted facility (such as its name,
address, classification and design flow rate). It contains the basic information
regarding a permit, permit-facility data is the one data type that belongs to all
of the families of logically related data types.
Detailed information describing each outfall within a permitted facility and
the discharge monitoring requirements associated with each (such as effluent
waste types, treatment types, and limit start and end dates-initial, interim, or
final).
Detailed information specifying the monitoring requirements associated with
each outfall within a permitted facility (such as monitoring location, the
parameter to be monitored, the required frequency of analysis, the units in
which the measurements are expressed, and the quantity and concentration
limits for each parameter).
Detailed information on the reported measurement values for effluent
parameters including those that are in violation of established limits for the
permit, the type of violation, the reported number of excursions, the actual
measurement values, and the percentage by which a measurement exceeds
quantity and/or concentration limits.
Information tracking the events relating to the issuance of a permit, from
initial receipt of the application for a permit through actual permit issuance.
Included in
PCSLoadCalculator200 7
Yes
Yes
No
Yes
Yes
Relational Tables used in
ICIS NPDES Pollutant
Loading Tool
Facility interest
Permit
Facility interest SIC
Facility interest NAICs
Permitted Feature
Permitted Feature Coordinates
Limit
Limit Set
Limit Set Schedule
Limit Value
DMR
DMR Event
DMR Form
DMR Form Parameter
DMR Form Value
DMR Parameter
DMR Value
REF_Parameter
Permit
Source: Permit Compliance System Generalized Retrieval Training Manual, Table 1-1, pg 1-4 (U.S. EPA, 2001); Results of ICIS-NPDES Pollutant Loading Tool
Convert Module Development and DMR Data Review - Update 1 (Camp, 2009).
3-6
-------
Section 3 - DMRLoads2007: Development and Category Rankings
• Grant data;
• Inspection data;
• Inspection scheduling data;
• Pretreatment audit/PCI data;
• Pretreatment performance data; and
• Schedule violation data.
3.2 DMRLoads2007: Database Development and Methodology
To develop DMRLoads2007, EPA developed two pollutant loading tools: the database
PCSLoadCalculator2007 and the ICIS-NPDES Pollutant Loading Tool. These loading tools start
with DMR data stored in their respective databases and use similar methodologies to calculate
annual mass discharges from DMR data. Due to differences in the PCS and ICIS-NPDES data
structures, EPA's analysis required two separate loading tools - one for PCS and one for ICIS-
NPDES8. EPA created DMRLoadsAnalysis2007 to merge data from the two systems, evaluate
the impacts of calculation assumptions, and track database corrections. EPA also created
DMRNutrients2007 to evaluate point source discharges of nitrogen and phosphorus.
The remainder of this subsection describes the methodology and assumptions used in
creating the DMRLoads2007 database to generate point source category rankings:
• Section 3.2.1 describes the data sources used to create DMRLoads2007';
• Section 3.2.2 describes PCSLoadCalculator2007;
• Section 3.2.3 describes the ICIS-NPDES Pollutant Loading Tool;
• Section 3.2.4 describes DMRLoadsAnalysis2007;
• Section 3.2.5 describes DMRNutrients2007; and
• Section 3.2.6 describes DMRLoads2007'.
3.2.1 Data Sources used in the Development of DMRLoads200 7
Figure 3-1 shows the relationship between PCS, ICIS-NPDES, and
DMRLoadsAnlaysis2007, DMRNutrients2007, andDMRLoads2007.
In future years of DMR data analysis, the ICIS NPDES Pollutant Loading Tool functions will include PCS load
calculations, and only one loading tool will be required.
3-7
-------
Section 3 - DMRLoads2007: Development and Category Rankings
ICIS-NPDES
ICIS-NPDES Convert Module
PCS Facility Data
NPDES Permit ID (NPID)
Standard Industrial Classification (SIC)
NAME, CITY, STATE
Major Facility Indicator (MADI)
Facility Latitude/Longitude Data
Permit Issuance Date (PERD)
Create Tables
Facility
Permit Feature
Convert DMR
PRAM/CAS Crosswalk
ICIS-NPDES Load Calculator Module
Imported Tables
Convert DMR
Create Tables
ICIS-NPDES Annual Loads
PCS CNVRT Module
PCSLoadCalculator2007
Imported Tables
Manual Database Corrections
Create Tables
PCS Annual Loads
DMRLoadsAnalysis2007
Imported Tables
Annual Loads (ICIS & PCS)
PCSFAC (PCS)
ICISFAC (ICIS)
Parameter Groupings (ICIS)
Create Tables
DMR2007
DMRFAC2007
Sensitivity Analysis (PCS only)
•DL (PCS & ICIS)
•EST (PCS)
•NOMAX (PCS)
•NOCNC (PCS)
Other Linked Tables
SIC/PSC Crosswalk
SUPERCAS
TWFs
DMRLoads2007
Linked Tables
DMR2007
DMRFAC2007
Create Tables
Category Rankings -
Toxic Weight
Other Linked Tables
SIC/Point Source
Category Crosswalk
SIC Codes
Point Source Category
Codes
DMRNutrients2007
Linked Tables
DMR2007
DMRFAC2007
Create Tables
Nutrient Loads by
Discharge
Category Rankings -
Total Nitrogen
Category Rankings-
Total Phosphorus
Figure 3-1. Relationship Between Data Sources and Database Development Tools for the
Development of DMRLoads2007
EPA used the following data sources and database development tools to create
DMRLoads2007:
• PCS: This mainframe database is the source of the pollutant discharge data and
facility information used in the development ofPCSLoadCalculator2007. EPA
used year 2007 data from PCS to develop PCSLoadCalculator2007'.
• ICIS-NPDES: This web-based Oracle™ database is the source of the pollutant
discharge data and facility information used in the development of the ICIS-
NPDES Pollutant Loading Tool. EPA used year 2007 data from ICIS-NPDES to
develop the ICIS-NPDES Pollutant Loading Tool. EPA obtained ICIS-NPDES
data directly from OECA, as it is not yet available through a public download
system.
• EPA's Convert Programs: Two EPA programs convert pollutant concentrations
and loads in DMR data into standard units and match them with flows and permit
limits. The PCS convert program (CNVRT) is a mainframe computer program
developed and maintained by OECA. The ICIS-NPDES Convert Module is a
ColdFusion™ based program developed by EPA for the 2009 and future annual
reviews that extracts DMR data from ICIS-NPDES and stores the converted data
to an Oracle™ database.
3-8
-------
Section 3 - DMRLoads2007: Development and Category Rankings
• EPA's Load Calculator Routines: EPA developed its PCS and ICIS-NPDES
Load Calculator Routines based on OECA's Effluent Data Statistics (EDS)
System for PCS:
— The PCS Load Calculator Routine uses a series of Microsoft Access™
database queries in PCSLoadCalculator2007 to compute annual pollutant
loads. In addition, PCSLoadCalculator2007 tracks database corrections
for monthly flow, quantity, concentration, reporting frequency, and
internal monitoring locations.
— The ICIS-NPDES Load Calculator Routine uses the ICIS-NPDES Load
Calculator Module, a ColdFusion™-based program that extracts converted
DMR data from the ICIS-NPDES Pollutant Loading Tool Oracle™
database and calculates annual pollutant loads. The ICIS-NPDES Load
Calculator Module differs from PCSLoadCalculator2007 in that it is part
of a dynamic web application that allows users to selectively query loads
and specify calculation assumptions.
Both load calculator routines produce five alternative loads by applying variations
in calculations assumptions (see Sections 3.2.2.2 and 3.2.3.2).
• DMRLoadsAnalysis2007: This PC-based Microsoft Access™ database
standardizes and then combines the annual loads data from
PCSLoadCalculator2007 and the ICIS-NPDES Loading Tool. The database also
examines the impact of the alternative load calculations (see Section 3.2.4.2 for
additional details). The database uses the calculation assumptions that EPA
selected based on the results of the data sensitivity analyses conducted for the
2007 annual review, and creates the DMR2007 table, which provides one annual
load per pollutant discharge. Additionally, this database calculates the toxic-
weighted pound equivalent (TWPE) for each pollutant discharge. This database
applies several database corrections, based on findings during previous annual
reviews and the 2009 annual review quality review (see Section 3.4), to correct
errors related to facility categorization, pollutant discharge categorization,
parameter groupings, intermittent discharges, and internal monitoring locations.
See Section 3.2.4 for additional details onDMRLoadsAnalysis2007.
• DMRNutrients2007: This PC-based Microsoft Access™ database uses the annual
loads for nitrogen and phosphorus compounds from the DMR2007 table to
calculate aggregate "nitrogen as N" and "phosphorus as P" loads for each facility
outfall. The database sums the aggregate nitrogen and phosphorus loads by
facility and by point source category and exports the aggregated loads to
DMRLoads2007.
3.2.2 PCSLoadCalculator2007
EPA developed PCSLoadCalculator2007 to process CNVRT output into a structure
usable to calculate annual loads. PCSLoadCalculator2007 is a Microsoft Access™ database that
implements EPA's PCS Load Calculator routine. As depicted in Figure 3-1,
PCSLoadCalculator2007 uses CNVRT output and calculates annual loads for each pollutant and
discharge point using PCS Load Calculator. The output from PCSLoadCalculator2007 is the
-------
Section 3 - DMRLoads2007: Development and Category Rankings
"PCS Annual Loads" table, which is exported to DMRLoadsAnalysis2007 for combination with
ICIS-NPDES Pollutant Loading Tool annual loads for further calculations and analyses.
The PCS Load Calculator routine is based on OECA's mainframe computer program,
called the EDS system. This system establishes how to calculate annual loads from the CNVRT
output and was used in the 2003 and 2005 annual reviews (U.S. EPA, 1997). In 2005, EPA
developed the PCS Load Calculator to duplicate the EDS system and to address data processing
difficulties when using EDS (U.S. EPA, 2005). EPA continues to use the PCS Load Calculator
routine instead of the EDS system because it allows EPA flexibility and control over the annual
load calculations and provides transparent documentation of the calculations.
3.2.2.1 CNVRT Module Input for PCSLoadCalculator2007
EPA used CNVRT module output to create PCSLoadCalculator2007 (see Figure 3-1).
From the PCS mainframe, the CNVRT module performs units conversions, matches flow rates
with pollutant measurements, assigns a statistical basis of measurement, and performs formatting
changes to convert the PCS data into a format that is usable for annual load calculations. Table
3-3 presents the CNVRT module output that EPA used as a starting point for its annual load
calculations for PCSLoadCalculator2007'.
Table 3-3. PCS CNVRT Module Output
PCS Field
NPID
SIC2
DSCH
DRID
NRPU
PRAM
MLOC
SEAN
MODN
LIPQ
STAT
MVDT
MVIO
NODI
LMQAV
LMQMX
LMCMN
LMCAV
LMCMX
MQAV
MQMX
MCMN
MCAV
Description
NPDES Number
Standard Industrial Classification Code
Discharge Pipe
Report Designator
Number of Units in Reporting Period
Parameter Code
Monitoring Location
Season Number
Modification Number
Limit Pipe Set Qualifier
Statistical Base Code
Measurement/Violation Monitoring Period End Date
Measurement/Violation Code
No Data Indicator
Measurement/Violation Quantity Average BDL Indicator
Measurement/Violation Quantity Maximum BDL Indicator
Measurement/Violation Concentration Minimum BDL Indicator
Measurement/Violation Concentration Average BDL Indicator
Measurement/Violation Concentration Maximum BDL Indicator
Measurement/Violation Quantity Average
Measurement/Violation Quantity Maximum
Measurement/Violation Concentration Minimum
Measurement/Violation Concentration Average
3-10
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-3. PCS CNVRT Module Output
PCS Field
MCMX
FMQAV
FMQMX
FMCMN
FMCAV
FMCMX
Description
Measurement/Violation Concentration Maximum
Measurement/Violation Quantity Average Flow
Measurement/Violation Quantity Maximum Flow
Measurement/Violation Concentration Minimum Flow
Measurement/Violation Concentration Average Flow
Measurement/Violation Concentration Maximum Flow
The following describes the functions of the CNVRT module:
• Unit Conversions: The CNVRT module converts the PCS measurement data into
standard units of kg/day for mass quantities, mg/L for concentrations, and million
gallons per day (MOD) for flow rates.
• Matching Flows with Pollutant Discharges: Quantities and concentrations are
reported to PCS using five pollutant parameter measurement fields (MQAV,
MQMX, MCMN, MCAV, MCMX (see Table 3-3 for measurement field
descriptions). Wastewater flow rates are reported to PCS as a pollutant parameter
using the same five measurement value fields. CNVRT matches wastewater flow
rates with pollutant measurements using identifying fields in PCS, such as
monitoring period end date, monitoring location, discharge pipe number, report
designator, and season number. CNVRT creates five new columns for each
pollutant discharge record and stores the matching flow information in these
fields (FMQAV, FMQMX, FMCMN, FMCAV, and FMCMX).
• Assigning Statistical Basis: The statistical basis of measurements in PCS is
identified by the statistical base code. CNVRT categorizes the 150 statistical base
codes in PCS as representing average, maximum, minimum, or total measured
values. CNVRT then simplifies the statistical base code by assigning a number
from 0 to 4 to each measurement value field. The assigned numbers are as
follows:
— 0 - No Value Reported,
— 1 - Average,
— 2 - Total Monitoring Period Value,
— 3 - Maximum, and
— 4 - Minimum.
CNVRT combines the numbers assigned to each of the five measurement values
into one five-digit code called STATS. Each of the five digits in STATS
corresponds to one of the five measurement fields for pollutant loads or
concentrations. Figure 3-2 shows an example of a possible STATS code. In this
figure, the measurements reported for MQAV, and MCAV are based on average
values, MQMX and MCMX are based on maximum values, and MCMN is based
on the minimum value.
3-11
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Formatting Changes: For pollutants measured at concentrations below their
detection limit (BDL), facilities report the detection limit concentration to PCS
and indicate that the measurement is BDL using a less-than sign (<). CNVRT
pulls the less-than signs from the measurement value fields and places them in a
separate field.
MQAV MQMX MCMN MCAV MCMX
(Average quantity (Maximum quantity (Minimum concentration (Average concentration (Maximum concentration
measurement) measurement) measurement) measurement) measurement)
Figure 3-2. Example PCS STATS Code
3.2.2.2 PCS Annual Load Calculation Routine
This section describes the calculations used to produce annual loads from CNVRT output
files in PCSLoadCalculator2007'. Figure 3-3 presents a flow diagram for the
PCSLoadCalculator2007 routine. Files obtained from the CNVRT module are the starting point
for the PCS Load Calculator routine.
PCS Data Selection
Some monitoring data in the CNVRT output are not relevant to calculating effluent loads,
and the PCS Load Calculator selects relevant CNVRT output. Irrelevant information includes
pollutant discharges for internal monitoring locations, pollutant discharges reported for certain
measurement fields, and flows reported for certain measurement fields. For example, for a
certain monitoring location, pollutant discharges may be reported as both a mass quantity and a
concentration. However, EPA does not use concentration data if the quantity is also reported.
The PCS Load Calculator routine selects relevant PCS data for the following parameters: 1)
monitoring location, 2) measurement value, and 3) flow value, as described below.
Monitoring Location Selection. Permits often require a facility to monitor at multiple
locations. The monitoring location is indicated in PCS in the MLOC field. Two of the many PCS
MLOC codes designate effluent discharges:
• MLOC 1 - Effluent gross discharge; and
• MLOC 2 - Effluent net discharge.
For its screening level review, EPA estimates annual loads that represent effluent
discharges. Therefore, the PCS Load Calculator searches the monitoring field location (MLOC)
in PCS to find effluent data only (MLOC 1 or MLOC 2). When both types of effluent data are
present for an outfall, MLOC 2 is used in preference to MLOC 1.
3-12
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Includes only Effluent Monitoring Locations
(MLOC=1 or MLOC=2)
NOMAX: Sets all maxima to zero
(MQMX and MCMX=01
Calculate Monitoring Period Load:
Quantity * 30 days/month * NRPU; or
Concentration * Flow * 30 days/month * NRPU
Calculate Annual Load using EST=YES:
Sum(Monitoring Period Load) * 12/(SUM(NRPU))
NOMAX KGY01,
JNOMAX KGYE1
Apply Hybrid Method
T
NOMAX_KGYH1
For use in NOMAX Analysis
CNVRT Output
Monitoring Location Selection
• Includes Internal and Effluent Monitoring Locations
FMQAV
FMQMX
>• Flow Values: FMCMN
FMCAV
FMCMX
Flow Value Selection
Flow (MGD)
Measurement Values:
MQAV MQMX MCMN MCAV MCMX
Measurement Value Selection
NOCONC: Sets all concentrations to zero
(MCMN, MCAV, and MCMX=0)
Normal Hierarchy
Calculate Monitoring Period Load:
Quantity * 30 days/month * NRPU; or
Concentration * Flow * 30 days/month * NRPU
Calculate Monitoring Period Load:
Quantity * 30 days/month * NRPU
Apply DL Options
Apply DL Options
BDL =
Calculate Annual Load using EST=YES:
Sum(Monitoring Period Load)* 12/(SUM(NRPU))
[
KGY01
KGYH01
For use in DL Analysis
I
Calculate Annual Load using EST=YES:
Sum(Monitoring Period Load) * 12/(SUM(NRPU))
NOCONC KGY01
Calculate Annual Load using EST=NO:
Sum(Monitoring Period Loads)
KGYE-I
KGYOO
Apply Hybrid Method
Apply Hybrid Method
KGYEO
Apply Hybrid Method
NOCONC_KGYH1
For use in NOCONC Analysis
KGYH1
For generating Category Rankings
KGYHO
For use in EST Analysis
Figure 3-3. Flow Diagram for PCS Load Calculator Routine
3-13
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Measurement Value Selection. PCS contains five measurement value fields for
measured pollutant data (MQAV, MQMX, MCMN, MCAV, and MCMX). The PCS Load
Calculator uses a two-step process to select which of these measurement values to use to
calculate the annual loads. In the first step, the PCS Load Calculator attempts to identify an
average value using STATS and a measurement field hierarchy. (See Section 3.2.2.1 for how
CNVRT develops STATS number using statistical base codes in PCS data.) The PCS Load
Calculator first searches each STAT digit corresponding to the PCS measurement fields in the
following sequence, or hierarchy:
• Average Load (MQAV);
• Maximum Load (MQMX);
• Average Concentration (MCAV);
• Maximum Concentration (MCMX); or
• Minimum Concentration (MCMN).
A measurement must meet two criteria to be selected for loads calculation: 1) the mass
quantity or concentration must be nonzero, and 2) the corresponding STAT digit for the
measurement value field must equal 1.
If the PCS Load Calculator cannot identify a measurement that meets these two criteria,
then the PCS Load Calculator selects measurement values based on which field they populate
without considering the STATS digit. In this step, the following hierarchy is used:
• The average load (MQAV) field is used if it contains a non-zero value;
• If MQAV cannot be used, and a flow rate is reported, the concentration fields are
searched in the following order and the first nonzero concentration is multiplied
by the flow to calculate the load:
— Average Concentration (MCAV);
— Maximum Concentration (MCMX);
— Minimum Concentration (MCMN); and
• If flow and concentration cannot be used to calculate the load, the maximum load
(MQMX) is used.
For sensitivity analyses, EPA calculated two sets of alternative loads ("NOMAX" and
"NOCONC") using variations on the measurement value selection hierarchy. Figure 3-3 shows
how these alternative loads relate to the loads calculated using the normal hierarchy. Section
3.2.4.2 describes the purpose of the alternative calculations and EPA's analysis of NOMAX and
NOCONC annual loads.
Flow Value Selection. To select the appropriate flow data to use to calculate annual
loads, the PCS Load Calculator uses a hierarchy that is similar to the measurement value
selection hierarchy. The PCS Load Calculator searches the flow measurement fields in the
following sequence and selects the first non-zero value it finds:
3-14
-------
Section 3 - DMRLoads2007: Development and Category Rankings
• Average Quantity Flow (FMQAV);
• Average Concentration Flow (FMCAV)9;
• Maximum Concentration Flow (FMCMX);
• Minimum Concentration Flow (FMCMN); and
• Maximum Quantity Flow (FMQMX).
While conducting the flow selection process, the PCS Load Calculator attempts to
identify and correct flows that have misreported units, which is a common problem for flows in
PCS. The PCS Load Calculator attempts to correct this problem by assuming that any reported
flow rate greater than 5,000 MGD is actually gallons per day (GPD), and divides the reported
flow by one million. For flows ranging from 1,300 to 5,000 MGD, EPA compares units for flow
permit limits to verify the units reported in PCS and makes corrections on a case-by-case basis10.
This is a change from the EDS methodology, which divides all flows that are greater than 1,300
MGD by one million. Section 3.2.2.3 discusses EPA's basis for this change in methodology.
Calculate Monitoring Period Load
After completing the monitoring location, measurement value, and flow selection
hierarchies, the PCS Load Calculator has identified one mass quantity or one concentration and
flow to calculate a load for each pollutant discharge for each monitoring period. The duration of
discharge that each monitoring period represents depends on the reporting frequency required by
a facility's NPDES permit. For example, if a facility is required to report monthly, then the
reported discharge for the monitoring period will represent one month of discharges (assuming
continuous discharges). If a facility is required to report quarterly, then the reported discharge for
the monitoring period will represent three months of discharges. EPA assumes that an outfall
discharges continuously for 30 days per month, and the PCS Load Calculator calculates the
monthly load using one of the following equations:
• Calculation of monthly load from daily load (MQAV or MQMX):
Monthly Load (kg/mo) = Daily Load (kg/day) x 30 (days/mo)
• Calculation of monthly load from concentration and flow (MCAV, MCMX, or
MCMN):
Monthly Load (kg/mo) = Cone. (mg/L) x Flow (MGD) x 3.785 (L/gal) x 30 (days/mo)
As Figure 3-3 shows, the PCS Load Calculator then adjusts the monthly load to represent
quarterly, semiannual, or annual loads where appropriate by multiplying each monthly load by
the number of reporting units (NRPU). The NRPU data element is a numeric code that indicates
whether a pollutant is monitored monthly (NRPU = 1), quarterly (NRPU = 3), semiannually
(NRPU = 6), or annually (NRPU = 12). For example, if a facility reported a 30-day average load
9 A "concentration flow" is a flow measurement that was reported to a concentration measurement field. Facilities
may report flows in any of the five measurement value fields. However, all flows are reported in units of MGD
whether they are reported in a quantity field or a concentration field.
10 EPA determined that all flows between 1,300 and 5,000 MGD reported by facilities in Ohio were flows in GPD.
EPA automatically divided these flows by 1,000,000. However, because power plants are known to have high flows,
EPA made flow corrections to Ohio facilities reporting SIC code 4911 (Electrical Services) on a case-by-case basis.
3-15
-------
Section 3 - DMRLoads2007: Development and Category Rankings
of 25 kg/day for its required quarterly report (NRPU=3), the PCS Load Calculator calculates the
load for the quarter as 25 kg/day x 30 days/month x 3 month/quarter = 2,250 kg/quarter.
Apply DL Options
As shown in Figure 3-3, the PCS Load Calculator produces two monitoring period loads
by using different calculation assumptions for pollutants that were measured BDL. Using the
BDL indicator field from the CNVRT output, the PCS Load Calculator identifies pollutants that
were measured BDL. If the BDL indicator field contains a less-than sign (<), the PCS Load
Calculator calculates two period loads: one by setting the monitoring period load to zero (BDL =
0) and a second by dividing the monitoring period load in half (BDL = 1A DL). If the BDL
indicator field is blank, then the PCS Load Calculator uses the calculated period load for both
options. Table 3-4 shows an example calculation of loads for the two DL options.
Table 3-4. Example Calculation for DL Option Loads
Calculated Monitoring
Period Load (kg/period)
100
100
BDL Indicator Field
Blank
<
Option BDL = 0 Load
(kg/period)
100
0
Option BDL = Vz DL
Load (kg/period)
100
50
Calculate Annual Load Scenarios
The output from the monitoring period load calculation step should include the following
data for each pollutant discharge:
• Twelve loads for monthly reports;
• Four loads for quarterly reports;
• Two loads for semiannual reports; and
• One load for annual reports.
However, in some cases, PCS does not contain a complete set of discharges for the year.
If a facility does not report a pollutant concentration or mass quantity on its DMR, then the
facility uses the no data indicator (NODI) field to explain why no discharge is reported. NODI is
a single character code in PCS, which corresponds to a no data indicator description. Table 3-5
presents descriptions of all the NODI codes.
Table 3-5. NODI Code Descriptions
NODI Code
1
2
4
5
7
8
9
NODI Description
Wrong flow
Operations shutdown
Discharge to lagoon/groundwater
Frozen conditions
No influent
Other
Conditional monitoring
3-16
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-5. NODI Code Descriptions
NODI Code
A
B
C
D
E
F
G
H
I
J
K
L
Q
R
S
V
w
X
NODI Description
General permit exemption
Below detection limit/no detection
No discharge occurred for the monitoring period
Lost sample
Analysis not conducted
Insufficient flow for sampling
Sampling equipment failure
Invalid test
Land applied
Recycled - water-closed system
Flood disaster
DMR received but not entered
Not quantifiable
Administratively resolved
Fire conditions
Weather related
Dry lysimeter/well
Parameter/value not reported
The PCS Load Calculator includes two options for calculating the annual load when PCS
does not contain a complete set of monitoring period loads for the year: 1) sum the existing
monitoring period loads to calculate the annual load (EST=NO); or 2) estimate loads for the
missing monitoring periods (EST=YES). The following sections describes the calculation of
EST=YES and EST=NO loads. For the 2009 annual review, EPA used only the EST=YES loads.
Calculate EST=YES Annual Loads. The PCS Load Calculator uses the sum of NRPU
values to identify annual loads that do not include a complete set of monitoring period loads.
First, the PCS Load Calculator sums the NRPU values for the monitoring periods that have
calculated pollutant loads. In addition, the PCS Load Calculator sums the NRPU values for blank
records with NODI codes that indicate no discharge occurred for the monitoring period. As part
of the 2009 annual review, EPA reviewed all NODI codes and determined that the following
NODI codes represent "no discharge" events:
2: Operations Shutdown;
4: Discharge to Lagoon/Groundwater;
7: No Influent;
9: Conditional Monitoring;
C: No Discharge;
I: Land Applied;
J: Recycled - Water-Closed System; and
W: Dry Lysimeter/Well.
3-17
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Note that EPA updated the list of NODI codes for the 2009 annual review. For previous annual
reviews, EPA used a different list of "no discharge" NODI codes. See Section 3.2.2.3 for
additional information.
The PCS Load Calculator then combines the sum of NRPU values for monitoring period
loads and monitoring periods with no discharge. If all monitoring periods for the annual data set
either have discharge data or indicate no discharge, then the sum of NRPU will equal 12. For
example, if a facility is required to monitor quarterly, the NRPU assigned to each quarterly
report is 3. If four quarterly reports are present, the total NRPU is 12 (3+3+3+3), indicating all
required reports are present. However, if the annual data set includes blanks for any of the
monitoring periods and does not indicate that no discharge occurred for the monitoring period,
then the sum of NRPU will be less than 12.
As shown in Figure 3-3, the input to the Calculate EST=YES Annual Loads step includes
two sets of monitoring period loads from the Calculate DL Options step: BDL = 0 and BDL = 1A
DL. To calculate the EST=YES load, the PCS Load Calculator sums monitoring period loads for
the DL = 0 option and separately sums the monitoring period loads for the DL = 1A DL option.
For each sum, the PCS Load Calculator then extrapolates the calculated annual load to account
for blank records using the following equation:
(EST=YES) Annual Load (kg/yr) = Sum(Monitoring Period Load x NRPU) x (12/Sum(NRPU))
Calculate EST=NO annual loads. During the EST=YES calculation step, the PCS Load
Calculator also calculates an alternative annual load using the EST=NO option. The calculation
for EST=NO is the same as the EST=YES calculation except EST=NO does not multiply the
sum of the period loads by the ratio of 12 and the sum of NRPU values. The EST=NO annual
load is shown in the following equation:
(EST = NO) Annual Load (kg/yr) = ^(MonitoringPeriodLoadxNRPU)
Apply Hybrid Method
As shown in Figure 3-3, the output from the Calculate EST=YES Annual Loads step
includes two annual loads for the DL options: BDL = 0 and BDL = 1A DL. During this
calculation step, the PCS Load Calculator applies the following logic to select which calculated
load to use to represent the final annual load:
• If the BDL = 0 load equals zero, use the BDL = 0 load (all monitoring period
loads for 2007 are zero); and
• If the BDL = 0 load is greater than zero, use the BDL = !/2 DL load (at least one
monitoring period was not zero, i.e., the pollutant was detected at least once
during 2007).
As shown in Figure 3-3, the PCS Load Calculator calculates alternative annual loads starting at
the Measurement Value Selection step. During this step, the PCS Load Calculator calculated two
sets of alternative monitoring period loads using variations on the measurement value selection
hierarchy: 1) set all maximum concentrations and loads to zero (NOMAX); and 2) set all
average, maximum, and minimum concentrations to zero (NOCONC). The PCS Load Calculator
-------
Section 3 - DMRLoads2007: Development and Category Rankings
then applied the DL options to these alternative loads and calculated EST=YES and EST=NO
annual loads for the NOMAX and NOCONC alternatives. As a final step the PCS Load
Calculator applies the Hybrid Method to the calculated alternative loads. See Section 3.2.4.2 for
discussion of the alternative annual loads.
PCSLoadCalculator2007 Output
The PCS Load Calculator produces 12 calculated annual loads for each pollutant
discharge. Table 3-6 lists the 12 calculated annual loads and describes the purpose of each load.
Seven of the loads use various assumptions for pollutant measurements reported as BDL, which
are used to calculate final loads using the Hybrid Method. Five of the loads are final loads, which
are used for category rankings and sensitivity analyses (see Section 3.2.4.2). The five final
annual loads are included in the PCSLoadCalculator2007 output to DMRLoadsAnalysis2007'.
Table 3-6. PCSLoadCalculator2007 Output
Annual Load
EST
Option
DL Option
Measurement
Selection Hierarchy
Purpose
Interim Loads
KGYE1
KGYOO
KGYEO
NOMAX JCGY01
NOMAX JCGYE1
NOCONC JCGY01
NOCONC JCGYEl
Yes
No
No
Yes
Yes
Yes
Yes
BDL = !/2DL
BDL = 0
BDL=!/2DL
BDL = 0
BDL = !/2DL
BDL = 0
BDL = !/2 DL
Normal
Normal
Normal
All maxima set to
zero
All maxima set to
zero
All concentrations set
to zero
All concentrations set
to zero
Used with KGY01 to calculate Hybrid
(KGYH1)
Used with KGYEO to calculate Hybrid
(KGYHO)
Used with KGYOO to calculate Hybrid
(KGYHO)
Used with NOMAX KGYE1 to
calculate Hybrid (NOMAXJCGYHl)
Used with NOMAX KGY01 to
calculate Hybrid (NOMAXJCGYHl)
Used with NOCONC KGYE1 to
calculate Hybrid (NOCONCJCGYHl)
Used with NOCONC KGY01 to
calculate Hybrid (NOCONCJCGYHl)
Final Loads
KGYH1
KGYHO
KGY01
NOMAXJCGYHl
NOCONCJCGYHl
Yes
No
Yes
Yes
Yes
Hybrid
Hybrid
BDL = 0
Hybrid
Hybrid
Normal
Normal
Normal
All maxima set to
zero
All concentrations set
to zero
Category Rankings
EST Analysis
DL Analysis
No Max Analysis
No Cone Analysis
3.2.2.3 Changes to EDS Methodology
As stated previously, EPA followed the EDS methodology to develop the annual load
calculation methodology for PCSLoadCalculator2007. This section discusses changes that EPA
made to the methodology including the reason for the change.
3-19
-------
Section 3 - DMRLoads2007: Development and Category Rankings
NRPU Correction. Monitoring frequencies may vary for certain pollutants or outfalls
depending on a facility's permit requirements. Discharges may be reported monthly, quarterly,
semiannually, or annually. As discussed previously, the NRPU data element is a numeric code
that indicates whether a pollutant is monitored monthly (NRPU = 1), quarterly (NRPU = 3),
semiannually (NRPU = 6), or annually (NRPU = 12). As described in Section 3.2.2.2, the PCS
Load Calculator uses the NRPU value for two steps in the annual load calculation.
• The first step that uses the NRPU value is the monitoring period load calculation.
During this step, the PCS Load Calculator calculates a monthly load by
multiplying a mass quantity by 30 days per month, and then multiplies the
monthly load by the NRPU value to calculate a quarterly, semi-annual, or annual
load.
• The second step that uses the NRPU value is the calculation of annual loads using
the EST=YES option. During this step, the PCS Load Calculator uses the sum of
the NRPU values associated with the reported discharges to determine if all DMR
data for the pollutant are present in PCS. If the sum of the NRPU values equals
12, then all required discharge data are present for that reporting year.
During the development ofPCSLoadCalculator2007, EPA observed that the sum of
NRPU values for several annual loads was greater than 12, indicating that discharge data for
more than the required number of DMRs were present in PCS. Following are two scenarios that
resulted in the sum of NRPU exceeding 12.
• Scenario 1: Incorrect NRPU reported. The first scenario is a data-entry error
where the NRPU in PCS was incorrect for the frequency of the reported
discharges. For example, a quarterly discharge report should have an NRPU value
of 3, but the NRPU value in PCS was 6. As a result, the monthly load for each
quarter was multiplied by 6 instead of 3 during the quarterly load calculation,
which double-counted the quarterly loads. The EST=YES calculation
automatically corrects this error by multiplying the annual load by the ratio of 12
to the sum of the NRPU values. For this example, the sum of NRPU values for the
four quarterly reports would be 24 instead of 12. Therefore, using EST=YES, the
annual load would be multiplied by 12/24 (0.5), which eliminates the double-
counting. For EST=NO, however, this error results in double-counting the annual
load since the EST=NO calculation does not multiply the annual load by the ratio
of 12 to the sum of NRPU values. EPA corrected the NRPU values for the
Scenario 1 cases by changing the NRPU values in the monthly data to correctly
reflect the monitoring frequency.
• Scenario 2: Multiple monthly measurements. The second scenario occurred if a
facility reported discharges twice in one month. For example, a facility reports a
discharge monthly to PCS (NRPU = 1), but reported two discharges for
September (one on September 15 and one on September 30). The NRPU values
for both September reports are 1. Similar to Scenario 1, the double-counting that
results from this error is corrected during the EST=YES calculation but not during
the EST=NO calculation. In addition to double-counting, this error also causes the
discharges reported for September to account for a disproportionate amount of the
annual load. For example, the monthly load calculation multiplies both the
3^20
-------
Section 3 - DMRLoads2007: Development and Category Rankings
September 15th and September 30th loads by 30. As a result, September
discharges account for 2 out of 13 months instead of 1 out of 12 months. EPA
corrected the NRPU values for the Scenario 2 cases by dividing the NRPU values
for months with multiple discharges by the number of discharges reported for the
month. For this example, the September NRPU value of 1 was divided by 2
because there were two discharge reports for September (corrected NRPU = 0.5).
As a result, the monthly load calculation multiplies each September discharge by
30 days per month and 0.5, making each discharge account for one half of a
month (15 days).
NODI Codes Excluded from EST=YES Assumption. As stated in Section 3222, EPA
updated the list of NODI codes that indicate that no discharge occurred. Prior to the 2009 Annual
Review, EPA used the NODI codes shown below:
• C: No discharge;
• D: Lost sample;
• E: Analysis not conducted;
• F: Insufficient flow for sampling;
• G: Sampling equipment failure;
• H: Invalid test;
• K: Flood disaster;
• 5: Frozen conditions; and
• 8: Other.
EPA revised this list to the following using information obtained from OECA during the
2009 annual review:
• 2: Operations shutdown;
• 4: Discharge to Lagoon/Groundwater;
• 7: No Influent;
• 9: Conditional Monitoring;
• C: No discharge;
• I: Land Applied;
• J: Recycled - Water-Closed System; and
• W: Dry Lysimeter/Well.
EPA assumed that the above NODI codes represent "no discharge" for the 2009 annual review
and will continue to use them for subsequent reviews.
EPA evaluated the effect of revising the NODI codes used in the EST=YES calculation
on the category rankings by comparing two sets of annual loads calculated using
PCSLoadCalculator200 7:
1. Annual loads calculated using the NODI codes for previous annual reviews (C, D,
E, F, G, H, K, 5, and 8); and
2. Annual loads calculated using the revised NODI codes for the 2009 annual review
(2, 4, 7, 9, C, I, J, and W).
3-21
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-7 presents a summary of the results of the NODI analysis for the 10 point source
categories showing the highest absolute increase in TWPE from the NODI changes and the total
for PCS-portion ofDMRLoads2007. As shown in Table 3-7, revising the NODI codes changed
the total TWPE for the PCS-portion of DMRLoads2007 by only 0.086 percent (794,000 Ib-eq).
The categories showing the greatest sensitivity to the NODI revisions include Copper Forming
(40 CFR Part 468); Justice, Public Order, and Safety (SIC Group 92), and Grain Mills (40 CFR
Part 406) Categories. Table B-3 in Appendix B presents the pounds and TWPE using the two
NODI code options by pollutant parameters. Pollutant parameters showing the highest sensitivity
to the NODI revisions include mercury, 2,3,7,8-Tetrachlorodibenzo-p-dioxin, and
polychlorinated biphenyls.
Flow Correction. As described in Section 3.2.2.2, the PCS Load Calculator attempts to
identify and correct flows that have misreported units using a two-step process. First, the PCS
Load Calculator assumes that any flow rate that is greater than 5,000 MGD should actually be
reported as GPD, and divides the flow by one million. EPA also reviews reported flows ranging
from 1,300 to 5,000 MGD by comparing reporting units to permit limits to verify the reporting
units and makes corrections on a case-by-case basis. This is a change from the EDS
methodology, which divides all flows that are greater than 1,300 MGD by one million.
The 1,300 MGD cutoff was based on the maximum flow rate identified at the time that
EDS was developed. EPA has identified several facilities that currently discharge wastewater at
flows exceeding 1,300 MGD. The 1,300 MGD cutoff used by EDS would underestimate loads
for these facilities by a factor of one million if the facilities report pollutant discharges as
concentrations in PCS. During the development of PCSLoadCalculator2004 as part of the 2007
annual review, EPA queried the Envirofacts Data Warehouse11 Web page for design flows. The
design flow rate is the average flow, in MGD, that a facility is designed to accommodate. The
highest design flow identified by this query was 4,453 MGD for the DC Water and Sewer
Authority (DC0000221). EPA based the new 5,000 MGD cutoff on this design flow. To be
consistent with the methodology used in the 2007 annual review, EPA used this cutoff again for
the 2009 annual review.
NODI B. The following is a discussion of a methodology change that EPA considered,
but decided not to implement. NODI (no data indicator) is a single character code that indicates
why pollutant measurements are blank for a reporting period. NODI = B means that the pollutant
was measured BDL for that monitoring period. Typically, facilities report BDL measurements by
reporting the detection limit concentration (or a mass quantity that was calculated using the
detection limit concentration) and indicate the measurement is BDL using a less-than (<) sign.
However, some facilities report BDL measurements by leaving the measurement value field
blank and reporting B in the NODI field. Because the detection limit concentration is not
provided in PCS, EPA cannot calculate period loads when the NODI B reporting method is used.
11 Envirofacts is a web-based system that allows the public to access PCS data for recent years.
3-22
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-7. Results of NODI Code Excluded from EST=YES Revision Analysis for PCS
Point Source Category
Copper Forming (40 CFR Part
468)
Justice, Public Order, and Safety
(SIC Group 92)
Grain Mills (40 CFR Part 406)
Nonferrous Metals Manufacturing
(40 CFR Part 421)
Canned and Preserved Seafood
Processing (40 CFR Part 408)
Meat and Poultry Products (40
CFR Part 432)
Metal Molding and Casting
(Foundries) (40 CFR Part 464)
Aluminum Forming (40 CFR Part
467)
Tobacco Products (PNC)
Mineral Mining and Processing
(40 CFR Part 436)
Total DMRLoads2007
PCS Annual
Load, Ib/yr a
2,110,000
1,230,000
21,400,000
152,000,000
9,900,000
53,800,000
5,860,000
13,900,000
10,700
265,000,000
35,800,000,000
PCS Annual Load
with NODI
Revisions, Ib/yr b
2,300,000
1,270,000
27,500,000
139,000,000
13,100,000
80,700,000
5,340,000
14,100,000
9,990
271,000,000
39,700,000,000
Difference in PCS
Annual Load,
Ib/yr
191,000(9.1%)
39,300 (3.2%)
6,130,000(29%)
13,000,000 (8.5%)
3,200,000 (32%)
26,900,000 (50%)
517,000(8.8%)
150,000(1.1%)
755 (7%)
6,200,000 (2.3%)
3,860,000,000
PCS TWPE, Ib-
eq/yr a
77.9
49.5
437
262,000
3,120
445,000
4,940
11,900
2.95
26,700
918,000,000
PCS TWPE with
NODI Revisions,
Ib-eq/yr b
1,080
214
1,600
529,000
5,670
152,000
6,040
13,600
2.53
29,400
918,000,000
Difference in PCS
TWPE, Ib-eq/yr
1,000 (1,290%)
164 (332%)
1,160 (265%)
267,000 (102%)
2,540 (82%)
292,000 (66%)
1,100(22%)
1,740 (15%)
0.422 (14%)
2,710 (10%)
794,000
Source: DMRLoads2007_v3.
a The Total Annual Load and Total TWPE include the revised NODI codes for the EST=YES assumption and were used in generating the category rankings.
b The Total Annual Load and Total TWPE with NODI Revisions include the pre-2009 NODI codes for the EST=YES assumption.
PNC - Potential new category.
3-23
-------
Section 3 - DMRLoads2007: Development and Category Rankings
If the pollutant is measured BDL for all 12 months of the year, then the outcome using
NODI B is the same as the Hybrid Method - the total annual load is zero. However, if the
pollutant is detected at least once during 2007, the EST=YES option will estimate loads for the
months when the pollutant was reported as NODI B based on the detected value. For example, if
a pollutant is reported as NODI B for 11 months but is measured at a concentration above its
detection limit for one month, then the effect of the EST=YES option would be to multiply the
detected concentration by 12 to account for the months when the facility reported NODI B. This
is an overestimation of the Hybrid Method, which would use a concentration equal to half the
detection limit for months when the pollutant was measured BDL.
EPA considered three options for correcting the overestimation of loads for NODI B:
• Option 1: Make no change.
• Option 2: Exclude NODI B from the EST=YES estimation option. The EST
function currently excludes a list of NODI characters that indicate that no
discharge occurred for the monitoring period. Adding NODI B to the list would
result in setting all BDL measurements that use the NODI B reporting method to
zero, which is an underestimation of the Hybrid Method.
• Option 3: Use a concentration of one-half the method detection limit (MDL) for
BDL measurements if the pollutant was detected at least once for 2007. This
option most closely resembles the Hybrid Method, but it would require EPA to
identify MDLs for all pollutant parameters with NODI B values. Based on 2007
data, NODI B was reported for more than 250 parameters.
EPA conducted an analysis to determine the impact of using the EST function for NODI
B on the category rankings. EPA ran the PCS Load Calculator and generated category rankings
first using EST=YES for NODI B and then using EST=NO for NODI B. EPA's analysis found
that estimating for NODI B using EST=YES accounts for 708,000 Ib-eq (0.08 percent) of the
TWPE from facilities mPCSLoadCalculator2007. The top 12 categories generated using
EST=YES for NODI B and using EST=NO for NODI B are identical. Therefore, EPA concluded
that, because using EST=YES for NODI B did not have a significant impact on the screening-
level analysis, no correction was necessary for the NODI B estimation. As a result, EPA did not
make any changes to the EST=YES calculation methodology for NODI B.
NODI Q. Similar to the NODI B analysis discussed above, EPA evaluated the potential
effects of including NODI Q in the EST=YES assumption on the category rankings. NODI Q
means that the measurement was not quantifiable. A measurement is not quantifiable if the
concentration was above the detection limit but the laboratory has determined that the value
cannot be accurately determined. As in the NODI B discussion above, EPA considered three
options for correcting the overestimation due to NODI Q:
• Option 1: Make no change.
• Option 2: Exclude NODI Q from the EST=YES estimation option. The EST
function currently excludes a list of NODI characters that indicate that no
discharge occurred for the monitoring period. Adding NODI Q to the list would
result in setting all BDL measurements that use the NODI Q reporting method to
zero, which is an underestimation of the Hybrid Method.
3-24
-------
Section 3 - DMRLoads2007: Development and Category Rankings
• Option 3: Use a concentration of one-half the method detection limit (MDL) for
BDL measurements if the pollutant was detected at least once for 2007. This
option most closely resembles the Hybrid Method, but it requires EPA to identify
MDLs for over 300 pollutant parameters.
For the same reasons described in the NODI B section above, EPA does not have the detection
limits for NODI Q records and cannot apply the EST=YES assumption using the Hybrid Method
for outfalls in which a pollutant is detected at least once during the year. Because the number of
records reporting NODI Q in PCS was small (0.02 percent of the total), EPA determined that
there should be no change to EDS methodology.
3.2.3 ICIS-NPDES Pollutant Loading Tool
The ICIS-NPDES Pollutant Loading Tool is a web-based application consisting of a user
interface, business logic layer, and an Oracle™ database. The purpose of the pollutant loading
tool is to calculate annual loads, similar to PCS CNVRT and PCSLoadCalculators, but for ICIS-
NPDES data instead of PCS data. The ICIS-NPDES Pollutant Loading Tool contains two
calculation modules:
• A Convert Module that extracts ICIS-NPDES DMR data, processes and formats
the data for loads calculations, and stores the converted data in an Oracle™
database; and
• A Load Calculator Module that queries the Oracle™ database and calculates
annual pollutant loads.
To allow for consistency between the calculated PCS and ICIS-NPDES loads, the Convert
Module mimics the functions of the PCS CNVRT Module while the ICIS-NPDES Load
Calculator Module mimics the methodology EPA developed for the PCS Load Calculator routine
that is used in PCSLoadCalculator2007'.
3.2.3.1 ICIS-NPDES Convert Module Development and Verification
The ICIS-NPDES Convert Module extracts data from ICIS-NPDES tables, stores the
extracted data into five denormalized tables, converts DMR measurements into standardized
units of measure, identifies the statistical basis of the permit limits, and matches DMR
measurements with wastewater flows and permit limits. The following describes the functions of
the ICIS-NPDES Convert Module.
Step 1: Extract Data and Create Denormalized Tables: The Convert Module
downloads and stores the ICIS-NPDES data from 19 extracted tables into the following five
interim tables: DMR, FACILITY, LIMITS, PERMIT FEATURE, and PRAM_CAS
CROSSWALK. Additionally, the Convert Module also creates three lookup tables to perform the
Convert Module functions:
• The UNIT_CONVERSIONS table provides conversion factors for unit codes to
convert concentrations into units of mg/L, loads into kg/day, and flows into
MOD;
3-25
-------
Section 3 - DMRLoads2007: Development and Category Rankings
• The STATS table assigns approximately 160 statistical base codes from the
LIMITS table in ICIS to one of four categories: 1 = Average; 2 = Total; 3 =
Maximum; and 4 = Minimum; and
• The FLOW_PRAM_CODES table identifies 24 parameter codes for wastewater
flow and assigns priorities that the Convert Module uses to match one flow per
outfall and monitoring period for load calculations.
Step 2: Convert to Standard Units: DMR data and permit limits are stored in ICIS-
NPDES in the measurement units specified by facilities' NPDES permits. The ICIS-NPDES
database then converts the DMR measurements and limits into standard units. The Convert
Module verifies the ICIS standard units conversion in the following steps:
• Identify Units of Measure - Unit codes are provided in the DMR and LIMITS
tables. However, the unit code field in the DMR table is blank for most records.
Therefore, the Convert Module selects unit codes from the LIMITS tables if the
DMR unit code fields are blank.
• Verify ICIS Unit Conversions - In this step, the Convert Module back-calculates
the ICIS conversion factors by dividing the standard units values by the original
values. EPA then compared the ICIS conversion factors to conversion factors that
EPA specified in the UNIT_CONVERSIONS look-up table (Table B-4 in
Appendix B). EPA corrected the ICIS-NPDES conversions that did not match the
look-up table. The Convert Module corrected approximately 0.3 percent of the
DMR records in ICIS-NPDES.
Step 3: Assign Statistical Base Codes: ICIS data contain approximately 160 statistical
base codes to describe the statistical basis of the DMR measurements (e.g., 30-day geometric
mean or rolling average).These codes are stored in the LIMITS table. The Convert Module uses
the STATS look-up table (Table B-5 in Appendix B) to assign each statistical base code to one of
five categories:
• 0 = Statistical Base Code is Null;
• 1 = Average;
• 2 = Total;
• 3 = Maximum; and
• 4 = Minimum.
Although specific information regarding the statistical basis of the measurement is lost during
this step, the simplification is necessary for efficient calculation of loads. The Convert Module
creates a STATS code (see Figure 3-2 for an example STATS code) consisting of five characters.
Each character of the STATS code corresponds to one measurement value field.
Step 4: Select Flows: The Convert Module selects DMR measurements for parameters
that are identified as wastewater flows in the FLOW_PRAM_CODES look-up table, and
matches flows with pollutant measurements. EPA identified 24 flows that are appropriate for
loads calculations12. EPA assigned priorities to the PRAM codes in the FLOW PRAM CODES
12 ICIS-NPDES contains other flow parameters such as recirculation flow, flow into well, and flows reported as
percentages or number of occurrences. EPA determined that these flows were not appropriate to use in calculating
mass discharges to receiving streams.
3-26
-------
Section 3 - DMRLoads2007: Development and Category Rankings
look-up table (Table B-6 in Appendix B). If a facility reports more than one type of flow PRAM
for the same outfall, then the Convert Module selects the PRAM code with the highest priority in
the look-up table. As a result, the Convert Module selects only one flow for each outfall,
monitoring location, and monitoring period end date.
Step 5: Select Temperature and pH: The Convert Module creates two new columns in
the CONVERT_DMR table for wastewater stream temperature and pH. ICIS-NPDES contains
two parameter codes for temperature and one parameter code for pH:
• Temperature Degrees C = PRAM 00010;
• Temperature Degrees F = PRAM 00011; and
pH = PRAM 00400.
The Convert Module uses a measurement value selection hierarchy, based on the STATS codes
created in Step 4, to select one pH and one temperature for each permitted feature, monitoring
location, and monitoring period end date.
Step 6: Identify Number of Report: The Convert Module identifies the number of days
per monitoring period using the NMBR_OF_REPORT field and the
MONITORING_PERIOD_END_DATE field. The NMBR_OF_REPORT field indicates the
number of months of discharges represented on each DMR. For example, a
NMBR_OF_REPORT of 1 indicates a monthly report and a NMBR_OF_REPORT of 3 is a
quarterly report (i.e., three months are in a quarter). EPA reviewed the ICIS-NPDES data and
identified the following valid NMBR_OF_REPORT values:
• 1 = Monthly Report;
• 2 = Bi-monthly Report;
• 3 = Quarterly Report;
• 4 = Triannual Report (typically for April, August, and December);
• 6 = Semi-annual Report; and
• 12 = Annual Report.
However, due to a data-entry rule in ICIS, some DMR records have invalid
NMBR_OF_REPORT values, such as 5, 7, 8, 9, 10, 11 or greater than 12. Because ICIS does not
allow users to enter a monitoring period start date that is earlier than the permit effective date,
facilities whose permits are renewed part-way through the year cannot enter valid
NMBR_OF_REPORT values. For example, if a facility submitted a semi-annual DMR in June,
which covered discharges from January to June, but their NPDES permit was renewed in
February, then ICIS will not allow the facility to enter a monitoring period start date earlier than
the effective date of the permit (February). As a result the NMBR_OF_REPORT field in ICIS is
5 instead of 6. The Convert Module addresses this issue by rounding up invalid
NMBR_OF_REPORT values to the next valid value. In addition, NMBR_OF_REPORT values
that are greater than 12 are converted to 12.
Once all NMBR_OF_REPORT have been converted to valid values, the Convert Module
uses Table 3-8 to assign the actual number of days for the monitoring period. This table presents
the actual number of days for all possible MONITORING_PERIOD_END_DATE and
NMBR_OF_REPORT combinations.
3^27
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-8. Actual Number of Days per Monitoring Period
Monitoring Period End Date Month
January
February a
March
April
May
June
July
August
September
October
November
December
Number of Report
1
31
28
31
30
31
30
31
31
30
31
30
31
2
62
59
59
61
61
61
61
62
61
61
61
61
3
92
90
90
89
92
91
92
92
92
92
91
92
4
123
120
121
120
120
122
122
123
122
123
122
122
6
184
181
182
181
182
181
181
184
183
184
183
184
12
365
365
365
365
365
365
365
365
365
365
365
365
"Does not account for the number of days in February during leap years.
As a final step for assigning the number of days per monitoring period, the Convert
Module identifies and corrects monitoring periods with multiple reported measurements. For
example, if a facility's NPDES permit requires them to report wastewater selenium discharges on
both January 15 and January 30, the Loading Tool would overestimate the annual selenium load
because it would multiply both the January 15 and January 31 discharges by 31 days per month
according to Table 3-8. To eliminate this overestimation, the Convert Module divides the
NMBR_OF_REPORT and the NMBR_OF_DAYS by the number of DMRs submitted per
monitoring period. Using the above example, the Convert Module calculates the
NMBR_OF_REPORT and NMBR_OF_DAYS for the January DMRs as follows:
NMBR OF REPORT =
_ NMBR_OF_REPORT (l) _
2 Reports per month
= 0.5
NMBR OF DAYS
NMBR_OF_DAYS (31) _
2 Reports per month
= 15.5
Step 7: Correct Flows: The Convert Module corrects flows. The methodology corrects
all flows exceeding 5,000 MGD, and applies more conservative criteria to correct flows from
1,000 to 5,000 MGD. The Convert Module uses three types of erroneous flow indicators to
correct flows:
1. Type 1: Month-to-Month Variability: In this step, the Convert Module compares
flows reported for each month and identifies variations greater than three orders
of magnitude using the following procedure:
3-28
-------
Section 3 - DMRLoads2007: Development and Category Rankings
a. Create a field that identifies the magnitude of each flow (e.g., 62,800
MOD has a magnitude of 10,000);
b. Group flow magnitudes by LIMIT_ID13;
c. Find the minimum flow magnitude that is >1,000;
d. Find the maximum flow magnitude that is <1,000;
e. Calculate a flow correction factor by dividing Step C/Step D;
f If the correction factor indicates a difference of three orders of magnitude
or more, and the actual measured flow is 1,000 to 5,000 MOD, then
correct the flow as follows:
Corrected Flow (MGD) = Actual Measured Flow (MGD):
Maximum Flow Magnitude < 1,000 ^|
Minimum Flow Magnitude > 1,000 J
g. If the correction factor indicates a difference of one order of magnitude or
more, and the actual measured flow is > 5,000 MGD, then correct the flow
using the equation in step f).
Table 3-9 presents an example of a Type 1 flow correction that the Convert Module identified.
As shown in the table, the September 2007 flow is three orders of magnitude higher than the
flows reported for other monitoring periods. Therefore, the Convert Module divided the
September flow by 1,000.
Table 3-9. Example Type 1 Flow Correction
External Permit
Number
GA0037648
GA0037648
GA0037648
GA0037648
GA0037648
GA0037648
GA0037648
GA0037648
GA0037648
GA0037648
GA0037648
GA0037648
Permitted
Feature
Number
OBO
OBO
OBO
OBO
OBO
OBO
OBO
OBO
OBO
OBO
OBO
OBO
Monitoring Period
End Date
31-Mar-07
30-Apr-07
31-May-07
30-Jun-07
31-Jul-07
31-Aug-07
30-Sep-07
31-Oct-07
31-Dec-07
28-Feb-07
31-Jan-07
30-Nov-07
Original
Flow
0.74
0.54
0.67
1.31
1.02
1.06
2,554.00
1.24
1.29
0.96
1.02
0.85
Maximum Flow Magnitude < 1,000
Minimum Flow Magnitude > 1,000
Correction Factor
Flow
Magnitude
0.1
0.1
0.1
1
1
1
1000
1
1
0.1
1
0.1
1
1000
1000
New
Flow
0.74
0.54
0.67
1.31
1.02
1.06
2.55
1.24
1.29
0.96
1.02
0.85
Correction
Applied?
No
No
No
No
No
No
Yes
No
No
No
No
No
The LIMIT_ID is a unique identifier in the ICIS_LIMITS table. It is a primary key for each unique set of
parameter code, limit set, season, and permitted feature.
3-29
-------
Section 3 - DMRLoads2007: Development and Category Rankings
2. Type 2: Variations from Design Flows and Actual Average Flows in
FACILITIES: The FACILITIES table contains information for facility design
flow and actual average flow in MGD. These fields are not required and therefore
are not populated for all records. However, when populated, these fields can be
used to help evaluate the reasonableness of the flows reported in the DMR data.
The Convert Module compares the design flow and actual average flow in
FACILITIES to the reported flows in CONVERT_DMR using the following
procedure:
a. Use ACTUAL_AVG_FLOW if reported. If ACTUAL_AVG_FLOW is
not reported, then use DESIGN_FLOW.
b. Use similar procedure as Type 1 to calculate the magnitude of the reported
flows and the actual/design flow magnitudes.
c. Divide the reported flow (e.g., FQ1) magnitude by the actual/design flow
magnitude to calculate the correction factor.
d. If the correction factor indicates a difference of three orders of magnitude
or more, and the actual measured flow is 1,000 to 5,000 MGD, then
correct the flow as follows:
Corrected Flow (MGD) = Actual Measured Flow (MGD)
( Design Flow Magnitude ^
^Actual Measured Flow Magnitude J
e. If the correction factor indicates a difference of one order of magnitude or
more, and the actual measured flow is > 5,000 MGD, then correct the flow
using the equation in step d.
Table 3-10 presents an example of a Type 2 flow correction that the Convert
Module identified. As shown in the table, the reported flows (FQ1) were six
orders of magnitude higher than the facility design flow. Therefore, the Convert
Module divided all flows by 1,000,000.
Table 3-10. Example Type 2 Flow Correction
External
Permit
Number
NHO 100692
NHO 100692
NHO 100692
NHO 100692
NHO 100692
NHO 100692
NHO 100692
NHO 100692
NHO 100692
NHO 100692
Permitted
Feature
Number
001
001
001
001
001
001
001
001
001
001
Monitoring
Period End
Date
31-Jan-07
28-Feb-07
31-Mar-07
30-Apr-07
31-May-07
30-Jun-07
31-Jul-07
31-Aug-07
30-Sep-07
31-Oct-07
Original
Flow
250,038
131,243
203,087
308,359
382,444
460,524
308,488
154,491
161,996
158,444
Flow
Magnitude
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
Design
Flow
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
Design Flow
Magnitude
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Correction
Factor
1,000,000
1,000,000
1,000,000
1,000,000
1,000,000
1,000,000
1,000,000
1,000,000
1,000,000
1,000,000
New
Flow
0.25
0.13
0.20
0.31
0.38
0.46
0.31
0.15
0.16
0.16
3-30
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-10. Example Type 2 Flow Correction
External
Permit
Number
NHO 100692
NHO 100692
Permitted
Feature
Number
001
001
Monitoring
Period End
Date
30-Nov-07
31-Dec-07
Original
Flow
183,168
190,775
Flow
Magnitude
100,000
100,000
Design
Flow
0.5
0.5
Design Flow
Magnitude
0.1
0.1
Correction
Factor
1,000,000
1,000,000
New
Flow
0.18
0.19
3. Type 3: Flows Exceeding the 5,000 MGD Cap: If a reported flow exceeds 5,000
MOD and is not identified for the Type 1 or Type 2 corrections, then the Convert
Module assumes that the flow was incorrectly entered in units of GPD and divides
the flow by 1,000,000.
Step 8: Select Effluent Monitoring Location: Permits often require a facility to monitor
at multiple locations. The monitoring location is indicated in the CONVERT_DMR table in the
MONITORING_LOCATION_CODE (MLOC) field. Five monitoring location codes in ICIS-
NPDES represent effluent discharges, seen below. For its screening level review, EPA estimates
annual loads that represent effluent discharges. Like PCSLoadCalculator2007, the ICIS-NPDES
Load Calculator searches the monitoring location field to identify effluent data only. When more
than one type of effluent data are present for an outfall, MLOC is selected in the following
hierarchy:
MLOC 2 - Effluent gross discharge;
MLOC 1 - Effluent net discharge;
MLOC A - After disinfection;
MLOC B - Before disinfection; and
MLOC SC - See comments.
For example, if a facility reports both MLOC 1 and MLOC 2, MLOC 2 is used in preference to
MLOC 1.
tables:
In executing the above steps, the Convert Module creates the following four output
CONVERT DMR;
FACILITY;
PERMIT FEATURE; and
PRAM CAS CROSSWALK.
Figure 3-4 shows the relationship diagram for the Convert Module Output.
The CONVERT_DMR table contains year 2007 DMR measurements for over 10,000
permits, of which approximately 80 percent are individual NPDES permits, 15 percent are
general permits, and the remaining five percent include industrial user permits and state-issued
non-NPDES permits.
3-31
-------
Section 3 - DMRLoads2007: Development and Category Rankings
FACILITY Table
CONVERT DMR Table
PERMIT FEATURE Table
ICIS FACILITY INTEREST ID
EXTERNAL PERMIT NMBR
FACILITY UIN
MAJOR MINOR STATUS FLAG
CITY
STATE CODE
COUNTY CODE
ZIP
HUC_CODE
EPA_REGION_CODE
STATE_WATER_B OD Y_NAME
TOTAL_DE SIGN_FLOW_NMBR
ACTUAL AVERAGE FLOW NMBR
PERMIT TYPE CODE
FACILITY TYPE CODE
GEOCODE LATITUDE
GEOCODE LONGITUDE
-
ACTIVITYJD
4 >
ACTIVITY ID
EXTERNAL PERMIT NMBR
PERM FEATURE ID
PERM FEATURE NMBR
LIMIT SET DESIGNATOR
LIMIT_SET_ID
LQ1
LQ2
LCI
LC2
LC3
LIMIT ID
STAT5
MONITORING PERIOD END DATE
NMBR OF REPORT
NMBR OF SUBMISSION
NMBR OF DAYS
MONITORING LOCATION CODE
PARAMETER CODE
NODI CODE
Ql QUAL
Cl QUAL
C2 QUAL
C3 QUAL
MQ1
MQ2
MCI
MC2
MC3
FSAT5
FQ1
FQ2
FC1
FC2
FC3
AVG_TEMP
AVG_PH
LIMIT_SET_SCHED ULE_ID
DMR_FORM_ID
DMR_EVENT_ID
DMR_FORM_PARAMETER_ID
DMR_PARAMETER_ID
EXTERNAL PERMIT NMBR
PERM FEATURE NMBR
PERM FEATURE ID
DESIGN FLOW NMBR
ACTUAL AVERAGE FLOW NMBR
WATER BODY NAME
REACH ID
STATE WATER BODY NAME
LATTTUDE_MEASURE
LONGITUD E_MEASURE
PARAMETER_CODE
PARAMETER DESC
POLLUTANT DESC
CHEMICAL ABSTRACT SERVICE NMBR
Figure 3-4. Relationship Diagram for Convert Module Output
3-32
-------
Section 3 - DMRLoads2007: Development and Category Rankings
3.2.3.2 ICIS-NPDES Load Calculator Module Annual Load Calculation
This section describes the calculation steps used by the ICIS-NPDES Load Calculator
Module to produce annual loads from the ICIS-NPDES Convert Module output tables. As stated
in Section 3.2.3, the ICIS-NPDES Load Calculator was developed to mimic the methodology
EPA developed for the PCS Load Calculator routine (see Section 3.2.2.2). EPA developed the
ICIS-NPDES Pollutant Loading Tool as a user-guided, web-based module that includes
functions beyond calculating category rankings for the annual review.
The following is a description of the steps taken by the ICIS-NPDES Load Calculator for
selecting monitoring location, flows, and measurement values and calculating annual loads.
Step 1: Measurement Value Selection. The CONVERT_DMR table (depicted in Figure
3-4) stores DMR data extracted from ICIS-NPDES in five measurement value fields. These
include:
Quantity 1 (MQ1);
• Quantity 2 (MQ2);
• Concentration 1 (MCI);
• Concentration 2 (MC2); and
• Concentration 3 (MC3).
These measurement value fields correspond to the five DMR fields where quantity and
concentration data are stored: 1) Average Quantity (Quantity 1 or 2); 2) Maximum Quantity
(Quantity 1 or 2); 3) Minimum Concentration (Concentration 1, 2, or 3); 4) Average
Concentration (Concentration 1, 2, or 3); and 5) Maximum Concentration (Concentration 1, 2, or
3). Note that unlike PCS, the measurement value fields in ICIS-NPDES are not specific to
average, maximum, or minimum. The statistical basis of the measurements in ICIS-NPDES is
determined by the five-digit statistical base code associated with each measurement field.
Facilities may use a variety of measurements to populate the five measurement value
fields. For example, a facility can use a monthly average, daily average, 30-day geometric
average, etc. to represent the average quantity. The CONVERT_DMR table contains a five-digit
statistical base code (STATS). The following codes are used for the types of measurements that
may be reported:
• Average (STAT5=1);
Total (STAT5=2);
Maximum (STAT5=3);
• Minimum (STAT5=4); and
Null (STAT5=0).
Each of the five digits in the STATS field corresponds to one of the five measurement value
fields. Figure 3-5 shows an example of a possible STATS code. In this figure, the measurements
reported for MQ1, MC2, and MC3 are average values, MQ2 is a maximum value, and no value
was reported for MCI. The ICIS-NPDES STATS example shown in Figure 3-5 is the same as
that shown for PCS in Figure 3-2 except that it was updated to show how the STATS digits
correspond to ICIS-NPDES measurement value fields.
3^33
-------
Section 3 - DMRLoads2007: Development and Category Rankings
13011
Quantity 1 Quantity 2 Concentration 1 Concentration 2 Concentration 3
(MQ1) (MQ2) (MCI) (MC2) (MC3)
Figure 3-5. Example STATS Code in ICIS-NPDES CONVERT DMR Table
The ICIS-NPDES Load Calculator selects measurements for loads calculations using a
hierarchy that prioritizes average values and quantities. In the first step, the ICIS-NPDES Load
Calculator attempts to identify an average value (STAT=1) by searching the STATS digits from
left to right. By scanning left to right, the ICIS-NPDES Load Calculator searches the STATS
digits that correspond to measurement fields in the following sequence:
• Quantity 1 (MQ1);
• Quantity 2 (MQ2);
• Concentration 1 (MCI);
• Concentration 2 (MC2); and
• Concentration 3 (MC3).
If the ICIS-NPDES Load Calculator finds an average value (STAT5=1), then it selects the
corresponding measurement for load calculation and performs the following calculations:
• If the selected measurement is a quantity (MQ1 or MQ2):
— Average daily load (kg/day) = MQ
— Average concentration (mg/L) = MQ/(Flow (MOD) x3.785 (L/gal))
— Monitoring Period Load (kg/monitoring period) =
MQ xNMBR OF DAYS
— Monitoring Period Load Over Limit (LOL) = (MQ - LQ (Quantity Limit))
xNMBR OF DAYS
• If the selected measurement is a concentration (MCI, MC2, or MC3):
— Average daily load (kg/day) = MC * Flow * 3.785
— Average concentration (mg/L) = MC
— Monitoring Period Load (kg/monitoring period) = MC x Flow (MGD) x
3.785 (L/gal) x NMBR_OF_DAYS
— Monitoring Period LOL = (MC - LC (Concentration Limit)) x Flow x
3.785 x NMBR_OF_DAYS
If the ICIS-NPDES Load Calculator does not find an average value (STAT=1), then it
scans STATS from left to right for a total value (STAT=2). "Total" values apply only to quantity
measurements, and because these measurements represent the total mass discharge for the
monitoring period, the ICIS-NPDES Load Calculator cannot use the same calculations used for
average, maximum, and minimum values. If the ICIS-NPDES Load Calculator identifies a total
value, it selects the value and performs the following calculations:
3-34
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Average Daily Load (kg/day) = MQ / NMBR_OF_DAYS
• Average Concentration (mg/L) = MQ/(Flow x NMBR_OF_DAYS x 3.785)
• Monitoring Period Load (kg/monitoring period) = MQ
• Monitoring Period LOL = MQ - LQ
If the ICIS-NPDES Load Calculator does not find an average value (STAT=1) or a total
value (STAT=2), then it scans STATS from left to right for a maximum value (STAT=3). If the
Load Calculator identifies a maximum value, then it selects that value and performs the same
calculations used for the average values (STAT=1).
If the ICIS-NPDES Load Calculator does not find an average value (STAT=1), total
value (STAT=2), or maximum value (STAT=3), then it scans STATS from left to right for a
minimum value (STAT=4). If the ICIS-NPDES Load Calculator identifies a minimum value,
then it selects that value and performs the same calculations used for the average values
(STAT=1).
Finally, if all measurement value fields are blank, then the ICIS-NPDES Load Calculator
sets the average daily load, average concentration, monitoring period load, and load-over-limit
fields to null.
Table 3-11 presents the measurement value selection priorities and calculations.
Step 2: Flow Value Selection. The ICIS-NPDES Load Calculator uses a similar
hierarchy for selecting flow rates and the FSTAT5 code. The FSTAT5 code applies the same
concept as the STATS code, and provides information about the statistical basis of wastewater
flow values. Similar to the measurement value selection hierarchy, the flow selection hierarchy
prioritizes average flows.
First, the ICIS-NPDES Load Calculator attempts to find an average flow (FSTAT=1) by
scanning the FSTATS code from left to right. By scanning from left to right, the ICIS-NPDES
Load Calculator searches the F STATS digits corresponding to the flow values in the following
sequence:
• Flow Quantity 1 (FQ1);
• Flow Quantity 2 (FQ2);
• Flow Concentration 1 (FC1)14;
• Flow Concentration 2 (FC2); and
• Flow Concentration 3 (FC3).
If the ICIS-NPDES Load Calculator finds an average value (STAT5=1), then it selects the
corresponding flow for load calculation and performs the following calculations:
• Average Daily Flow (MGD) = Flow
• Monitoring Period Flow (MG/monitoring period) = Flow x NMBR_OF_DAYS
14 A "flow concentration" is a flow measurement that was reported to a concentration measurement field. Facilities
may report flows in any of the five measurement value fields. However, all flows are reported in units of MGD
whether they are reported in a quantity field or a concentration field.
3-35
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-11. Measurement Value Selection Priorities and Calculations
Priority
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Value Type
MQ1
MQ2
MCI
MC2
MC3
MQ1
MQ2
MQ1
MQ2
MCI
MC2
MC3
MQ1
MQ2
MCI
MC2
STAT
Code
1
1
1
1
1
2
2
o
J
o
J
o
J
o
J
o
J
4
4
4
4
Average Daily
Load (KGD)
MQ1
MQ2
MCI x Flow x
3.785
MC2 x Flow x
3.785
MC3 x Flow x
3.785
MQ1/
NMBR_OF_DAYS
MQ2/
NMBR_OF_DAYS
MQ1
MQ2
MCI x Flow x
3.785
MC2 x Flow x
3.785
MC3 x Flow x
3.785
MQ1
MQ2
MCI x Flow x
3.785
MC2 x Flow x
3.785
Average Concentration
(MP MGL)
MQl/(Flow x 3.785)
MQ2/(Flow x 3.785)
MCI
MC2
MC3
MQ1/(MP_MGD x 3.785)
MQ2/(MP_MGD x 3.785)
MQl/(Flow x 3.785)
MQ2/(Flow x 3.785)
MCI
MC2
MC3
MQl/(Flow x 3.785)
MQ2/(Flow x 3.785)
MCI
MC2
Monitoring Period DMR
Load (KGMP)
MQ1 x
NMBR_OF_DAYS
MQ2 x
NMBR_OF_DAYS
MCI x Flow x 3.785 x
NMBR_OF_DAYS
MC2 x Flow x 3.785 x
NMBR_OF_DAYS
MC3 x Flow x 3.785 x
NMBR_OF_DAYS
MQ1
MQ2
MQ1 x
NMBR_OF_DAYS
MQ2 x
NMBR_OF_DAYS
MCI x Flow x 3.785 x
NMBR_OF_DAYS
MC2 x Flow x 3.785 x
NMBR_OF_DAYS
MC3 x Flow x 3.785 x
NMBR_OF_DAYS
MQ1 x
NMBR_OF_DAYS
MQ2 x
NMBR_OF_DAYS
MCI x Flow x 3.785 x
NMBR_OF_DAYS
MC2 x Flow x 3.785 x
NMBR_OF_DAYS
MP LOL
(MQ1-LQ1) x
NMBR_OF_DAYS
(MQ2 - LQ2) x
NMBR_OF_DAYS
(MCI -LCI) x Flow x 3.785 x
NMBR_OF_DAYS
(MC2 -LC2) x Flow x 3.785 x
NMBR_OF_DAYS
(MC3 -LC3) x Flow x 3.785 x
NMBR_OF_DAYS
MQ1-LQ1
MQ2 - LQ2
(MQ1-LQ1) x
NMBR_OF_DAYS
(MQ2 - LQ2) x
NMBR_OF_DAYS
(MCI -LCI) x Flow x 3.785 x
NMBR_OF_DAYS
(MC2 -LC2) x Flow x 3.785 x
NMBR_OF_DAYS
(MC3 -LC3) x Flow x 3.785 x
NMBR_OF_DAYS
(MQ1-LQ1) x
NMBR_OF_DAYS
(MQ2 - LQ2) x
NMBR_OF_DAYS
(MCI - LCI) x Flow x 3.785 x
NMBR_OF_DAYS
(MC2 -LC2) x Flow x 3.785 x
NMBR_OF_DAYS
3-36
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-11. Measurement Value Selection Priorities and Calculations
Priority
17
18
Value Type
MC3
No Data (NODI
is not null)
STAT
Code
4
Any
Average Daily
Load (KGD)
MC3 x Flow x
3.785
NULL
Average Concentration
(MP MGL)
MC3
NULL
Monitoring Period DMR
Load (KGMP)
MC3 x Flow x 3.785 x
NMBR_OF_DAYS
NULL
MP LOL
(MC3 -LC3) x Flow x 3.785 x
NMBR_OF_DAYS
NULL
3-37
-------
Section 3 - DMRLoads2007: Development and Category Rankings
If the ICIS-NPDES Load Calculator does not find an average flow (FSTAT=1), then it
scans FSTAT5 from left to right for a total flow (FSTAT=2). Because "total" flows represent the
total wastewater discharge for the monitoring period, the ICIS-NPDES Load Calculator cannot
use the same calculations used for average, maximum, and minimum flows. If the ICIS-NPDES
Load Calculator identifies a total flow, it selects the value and performs the following
calculations:
• Average Daily Flow (MOD) = Flow/MNBR_OF_DAYS
• Monitoring Period Flow (MG/monitoring period) = Flow
If the ICIS-NPDES Load Calculator does not find an average flow (FSTAT=1) or a total
flow (FSTAT=2), then it scans FSTAT5 from left to right for a maximum flow (FSTAT=3). If
the Load Calculator identifies a maximum flow, then it selects that flow and performs the same
calculations used for the average flows (FSTAT=1).
If the ICIS-NPDES Load Calculator does not find an average flow (FSTAT=1), total
flow (FSTAT=2), or maximum value (FSTAT=3), then it scans FSTAT5 from left to right for a
minimum flow (FSTAT=4). If the ICIS-NPDES Load Calculator identifies a minimum flow,
then it selects that flow and performs the same calculations used for the average flows
(FSTAT=1).
Table 3-12 presents the flow value selection priorities and calculations.
Table 3-12. Flow Value Selection Priorities
Priority
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Value Type
Quantity 1
Quantity 2
Concl
Cone 2
Cone 3
Quantity 1
Quantity 2
Quantity 1
Quantity 2
Concl
Cone 2
Cone 3
Quantity 1
Quantity 2
Cone 1
Cone 2
Cone 3
FSTAT5
1
1
1
1
1
2
2
3
3
3
3
3
4
4
4
4
4
Average Daily Flow
(MGD)
FQ1
FQ2
FC1
FC2
FC3
FQ1 / NMBR_OF_DAYS
FQ2 / NMBR_OF_DAYS
FQ1
FQ2
FC1
FC2
FC3
FQ1
FQ2
FC1
FC2
FC3
Monitoring Period Flow
(MGMP)
FQ1 * NMBR_OF_DAYS
FQ2 * NMBR_OF_DAYS
FC1 * NMBR_OF_DAYS
FC2 * NMBR_OF_DAYS
FC3 * NMBR_OF_DAYS
FQ1
FQ2
FQ1 * NMBR_OF_DAYS
FQ2 * NMBR_OF_DAYS
FC1 * NMBR_OF_DAYS
FC2 * NMBR_OF_DAYS
FC3 * NMBR_OF_DAYS
FQ1 * NMBR_OF_DAYS
FQ2 * NMBR_OF_DAYS
FC1 * NMBR_OF_DAYS
FC2 * NMBR_OF_DAYS
FC3 * NMBR_OF_DAYS
3-38
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Step 3: Detection Limit Options (DL). When pollutants are not detected, their
concentrations are presumed to be below their detection limit (BDL). Permittees may report the
detection limit with a less-than sign (<) to indicate that the pollutant was BDL. The
CONVERT_DMR table stores the less-than signs for nondetects in the data qualifier field that
corresponds to the measurement value (i.e., MQl_Qual, MQ2_Qual, MCl_Qual, MC2_Qual, or
MC3_Qual). If a pollutant is BDL, the pollutant concentration may be between zero and the
detection limit. The ICIS-NPDES Load Calculator calculates three versions of each monitoring
period load and concentration using each of the following assumptions:
• BDL equals zero;
• BDL equals the detection limit; or
• BDL equals one-half the detection limit.
Step 4: Calculate Load-Over-Limit. The ICIS-NPDES Load Calculator performs some
calculations that are not performed on PCS data, but may be used as part of EPA's screening-
level review. The Load-Over-Limit (LOL) compares the monitoring period loads to the NPDES
permit limits on a mass basis. LOL is not used in rankings categories but may be used in future
reviews. LOL is calculated using the following steps:
• Select the limits from the limit value fields that correspond to the selected
measurement value fields;
• Calculate a monitoring period load over limit using one of the following
equations:
— Calculation of monitoring period LOL from mass quantity (MQ1 or
MQ2):
Monitoring Period LOL (kg/period) = [(MQ (kg/day) - LQ (kg/day)] x NMBR_OF_DAYS
— Calculation of monitoring period LOL from concentration and flow (MCI,
MC2, orMC3):
Monitoring Period LOL (kg/period) = [(MC (mg/L) - LC (mg/L)] x Flow (MGD) x 3.735 (L/gal)
x NMBR_OF_DAYS
• Create two options for Monitoring Period Load-Over-Limit Calculations:
— LOL1 - If the Monitoring Period LOL is negative, then set the LOL to
zero;
— LOL2 - If the Monitoring Period LOL is negative, then retain the
calculated negative value.
For example, if the DMR Monitoring Period Load is 200 and the Limit
Monitoring Period Load is 205, then the result for LOL1 would be 0 and the result
for LOL2 would be -5. If the DMR Monitoring Period Load is 210 and the limit is
205 then the results for LOL1 and LOL2 would both equal 5.
In Step 5, the ICIS-NPDES Load Calculator sums the LOL1 and LOL2 values for the year to
calculate the annual load over limit. The LOL1 method provides the total annual kilograms of
pollutant discharges that exceeded permit limits, but does not give credit for monitoring periods
3^39
-------
Section 3 - DMRLoads2007: Development and Category Rankings
where the discharges were below the permit limit. Summing the LOL2 values provides the net
permit limit exceedances for the year.
Step 5: Calculate Annual Totals and Averages. In this step, the ICIS-NPDES Load
Calculator calculates the following annual averages and totals for each facility, outfall,
monitoring location, limit set designator, and parameter:
• Total Annual Pollutant Load (kg/yr) = Sum of Monitoring Period Loads for three
DL Options:
— BDL = 0
— BDL = !/2 DL; and
— BDL = DL;
• Annual Average Pollutant Load (kg/day) = Avg of Average Daily Loads for three
DL Options:
— BDL = 0;
— BDL = !/2 DL; and
— BDL = DL;
• Total Annual Wastewater Flow (MG/yr) = Sum of Monitoring Period Flows;
• Annual Average Wastewater Flow (MGD) = Avg of Average Daily Flows;
• Annual Average Concentration (mg/L) = Avg of Concentrations for three DL
Options:
— BDL = 0;
— BDL = !/2 DL; and
— BDL = DL;
• Total Load-Over-Limit 1 = Sum of Monitoring Period LOL1;
• Total Load-Over-Limit 2 = Sum of Monitoring Period LOL2; and
• Annual Average Temperature and pH.
Step 6: Estimation Function (EST). Like PCSLoadCalculator2007 (see Section
3.2.2.2), the ICIS-NPDES Load Calculator estimates discharges for monitoring periods with
NODI codes that indicate discharge did not occur. The ICIS-NPDES Load Calculator assumes
no discharge for the same NODI codes as PCSLoadCalculator2007:
• 2: Operations shutdown;
• 4: Discharge to Lagoon/Groundwater;
• 7: No Influent;
• 9: Conditional Monitoring;
• C: No discharge;
• I: Land Applied;
• J: Recycled - Water-Closed System; and
• W: Dry Lysimeter/Well.
3^40
-------
Section 3 - DMRLoads2007: Development and Category Rankings
The ICIS-NPDES Load Calculator uses an identical methodology as the EST=YES
function used by PCSLoadCalculator2007; the ICIS-NPDES Load Calculator normalizes the
calculated annual load to 12 months per year using the following equation:
12
Annual Load (kg/yr) = Sum of Monitoring Period Loads x'
Sum of NMBR_OF_REPORT
EPA developed the ICIS-NPDES Load Calculator to allow users to selectively query
loads calculated using this estimation function (EST=YES) or without the estimation function
(EST=NO). Therefore, the ICIS-NPDES Load Calculator differs fromPCSLoadCalculator2007
in that only one set of annual loads is included in the final ICIS-NPDES annual loads output
table. EPA ran the ICIS-NPDES Load Calculator using EST=YES to develop the annual loads
for the 2009 annual review.
Step 7: Parameter Grouping. An NPDES permit may require a facility to measure a
pollutant in more than one way. For example, a facility may report both total lead and dissolved
lead. Because total lead includes dissolved lead, adding the two measurements together
overestimates the mass of lead discharged from the facility. To avoid double counting, the ICIS-
NPDES Load Calculator can group parameters that represent a single pollutant more
accurately15. The ICIS-NPDES Load Calculator grouping function uses a hierarchy to determine
which parameter best represents the total pollutant discharge. For example, copper has eight
parameter codes. If a facility reports multiple parameter codes for copper, the ICIS-NPDES Load
Calculator uses the following "grouping" hierarchy:
1. Total copper;
2. Copper;
3. Total copper per batch;
4. Total recoverable copper;
5. Dry weight copper;
6. Potentially dissolved copper; and
7. Sum of (dissolved copper and suspended copper).
Table B-7 in Appendix B presents the parameter grouping hierarchy.
Load Calculator Module Output. The ICIS-NPDES Pollutant Loading Tool output
contains loads, concentrations, flows, and wastewater stream conditions for each facility, outfall,
monitoring location, and parameter as seen in Table 3-13.
15 EPA also groups parameters in PCS using the same parameter grouping as ICIS-NPDES. The PCS parameter
grouping occurs inDMRLoadsAnalysis2007. See Section 3.2.4.1 for additional details.
3-41
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-13. ICIS-NPDES Load Calculator Module Output
Output
Parameter
EST
Option
DL Option
Used in Category
Rankings?
Purpose
Total Annual Pollutant Loads
KGYOO
KGYEO
KGY10
KGY01
KGYE1
KGY11
No
No
No
Yes
Yes
Yes
BDL = 0
BDL = !/2 DL
BDL = DL
BDL = 0
BDL = !/2 DL
BDL = DL
No
No
No
Yes
Yes
No
Not used in Rankings or Sensitivity Analysis
Not used in Rankings or Sensitivity Analysis
Not used in Rankings or Sensitivity Analysis
Used with KGYE1 to calculate Hybrid
Used with KGY01 to calculate Hybrid
DL Sensitivity Analysis
Annual Average Daily Loads
AVGJCGDO
AVGJCGDE
AVGJCGD1
NA
NA
NA
BDL = 0
BDL = !/2DL
BDL = DL
No
No
No
Not used in Rankings or Sensitivity Analysis
Not used in Rankings or Sensitivity Analysis
Not used in Rankings or Sensitivity Analysis
Annual Average Concentrations
AVG_MGLO
AVG_MGLE
AVG_MGL1
NA
NA
NA
BDL = 0
BDL = !/2DL
BDL = DL
No
No
No
Not used in Rankings or Sensitivity Analysis
Not used in Rankings or Sensitivity Analysis
Not used in Rankings or Sensitivity Analysis
Annual Load-Over-Limit
SUM_LOL1
SUM_LOL2
NA
NA
NA
NA
No
No
Not used in Rankings or Sensitivity Analysis
Not used in Rankings or Sensitivity Analysis
Average Wastewater Stream Conditions
AVG_TEMP
AVG_PH
NA
NA
NA
NA
No
No
Not used in Rankings or Sensitivity Analysis
Not used in Rankings or Sensitivity Analysis
Total Annual Wastewater Flow
MGY
NA
NA
No
Not used in Rankings or Sensitivity Analysis
Annual Average Daily Wastewater Flow
AVG_MGD
NA
NA
No
Not used in Rankings or Sensitivity Analysis
NA = Not Applicable
3.2.4 DMRLoadsAnalysis2007
As depicted in Figure 3-6, the DMRLoadsAnalysis2007 database imports annual load
tables from PCSLoadCalculator2007 and the ICIS-NPDES Load Calculator and facility
information from PCS and ICIS-NPDES. PCSLoadCalculator2007 creates annual loads from
using the hybrid methodology but without grouping the parameters while ICIS-NPDES Load
Calculator groups the parameters but does not apply the hybrid methodology. To create one set
of annual loads from the two data sets, DMRLoadsAnalysis2007 applies the hybrid methodology
to the output from the ICIS-NPDES Load Calculator and groups the parameters in the annual
load output from PCSLoadCalculator2007'. DMRLoadsAnalysis2007 then uses information
facility information from PCSFAC and ICIS Facilities and Chemical Abstract Services (CAS)
numbers to calculate TWPE and create the "combined" annual loads table ("DMR2007") that is
used by DMRLoads2007 to generate category rankings. In 2007, 64 percent of the records in the
DMR2007 table were from PCS, while the remaining 36 percent were from ICIS-NPDES.
Additionally, DMRLoadsAnalysis2007 uses annual loads from the PCSLoadCalculator2007 and
3-42
-------
Section 3 - DMRLoads2007: Development and Category Rankings
ICIS-NPDES Load Calculator outputs to perform a sensitivity analysis on the various calculation
assumptions.
Table 3-14 describes the function of each table inDMRLoadsAnalysis2007.
Table 3-14. Tables Imported or Created in DMRLoadsAnalysis2007
Table Name
PRAM Codes - PCS
PRAM Codes - ICIS
PCS to ICIS-NPDES
Parameter Crosswalk
Point Source
Category/SIC
Crosswalk
Point Source Category
Codes
SIC Codes
SUPERCAS Category
TWFs
ICIS Facilities
PCSFAC
DMRFAC2007
DMR2007
DMR2007 Sensitivity
Analysis
Parameter Groupings
Created or Imported
Imported from PCS
Imported from ICIS-NPDES
Created
Imported from
TRICalculations2007
Imported from
TRICalculations2007
Imported from
TRICalculations2007
Imported from ICIS-NPDES
Imported from
TRICalculations2007
Imported from ICIS-NPDES
Imported from PCS
Created
Created using queries
Created using queries
Imported from ICIS-NPDES
Description
Lists pollutants and corresponding parameter codes used for
them in PCS.
Lists pollutants and corresponding parameter codes used for
them in ICIS-NPDES.
Links PCS parameter codes to ICIS-NPDES parameter
codes
Links SIC codes with point source categories using a
numeric code assigned in the Point Source Category Codes
table.
Assigns a numeric code to industrial categories using their
40 CFR Part number or 2-digit or 4-digit SIC code.
Lists SIC codes and their descriptions.
Links CAS numbers to pollutant parameter codes.
Assigns TWF values to chemicals by CAS number.
Presents information on permitted facilities, such as facility
name, location and major/minor discharge status.
Presents information on permitted facilities, such as facility
name, location, major/minor discharge status, and date of
most recent permit issuance.
Combines PCSFAC and ICISFAC into one table.
Presents the annual loads in pounds per year and TWPE for
each pollutant discharge for each outfall at major permitted
facilities.
Presents the annual loads in pounds per year for each
pollutant discharge for each outfall for the five annual loads
calculated by PCSLoadCalculator2007 and three annual
loads calculated by the ICIS-NPDES Load Calculator (see
Section 3.2.4.2 and Table 3-16).
Lists ICIS-NPDES pollutant parameter codes and their
hierarchies for grouping parameters. This table is used to
group parameters in both PCS and ICIS-NPDES.
3-43
-------
Section 3 - DMRLoads2007: Development and Category Rankings
PCS Annual Loads
NPDES Permit Number
Major Indicator
Discharge Pipe Number
Monitoring Location
Report Designator
Parameter Code
KGY01
KGYHO
KGYH1
NOMAX KGYH1
NOCONC KGYH1
ICIS-NPDES Annual Loads
NPDES Permit Number
Major Indicator
Discharge Pipe Number
Monitoring Location
Report Designator
Parameter Code
KGY01
KGYE1
KGY11
Step 1: Parameter Grouping Hierarchy
PCS to ICIS-NPDES Parameter Crosswalk
PCS Parameter Code
ICIS-NPDES Parameter Code
Parameter Groupings
ICIS-NPDES Parameter Code
Group Code (if grouped)
Grouped
Annual Loads
Step 3: Combine Annual Loads
Combine Loads
LBY = KGYH1 x 2.205
DMR2007
SIC Code
NPDES Permit Number
Major Indicator
Discharge Pipe Number
Monitoring Location
Report Designator
Parameter Code
Total Annual Pounds
TWPE
Major Indicator = M (Majors Only)
Combined Facility Data (DMRFAC2007)
NPDES Permit ID
SIC Code
NAME, CITY, ZIP
Major Facility Indicator
TWPE
KGYH1
Step 2: Hybrid Methodology
Apply Hybrid Methodology
Step 4: Calculate TWPE
SUPERCAS
ICIS-NPDES Parameter/Group Code
Parameter Name
CAS Number
Figure 3-6. DMRLoadsAnalysis2007 Inputs Used to Create DMR2007 Table
3-44
-------
Section 3 - DMRLoads2007: Development and Category Rankings
3.2.4.1 Creation of DMR2007 Annual Loads Table
The following is a description of the steps EPA took to combine the annual loads from
ICIS-NPDES Load Calculator and PCSLoadCalculator2007 into one table, "DMR2007".
Step 1: Apply Parameter Grouping Hierarchy to PCSLoadCalculator2007 Annual
Loads. The first step in applying the parameter grouping hierarchy to the annual loads from
PCSLoadCalculator2007 is to convert the PCS parameter codes to ICIS-NPDES parameter
codes because there are some parameters that have different parameter codes between the
databases. EPA created the PCS to ICIS-NPDES Parameter Crosswalk table, which links the
PCS parameter code to the ICIS-NPDES parameter code. EPA then updated the PCS parameter
codes in the annual loads tables from PCSLoadCalculator2007 to the ICIS-NPDES parameter
codes.
As discussed in Section 3.2.3.2, aNPDES permit may require a facility to measure a
pollutant in more than one way. The annual loads from PCSLoadCalculator2007 include one
load for every parameter reported. To avoid double-counting pollutants, EPA applied the same
hierarchy used in the ICIS-NPDES Load Calculator to group the pollutants in the
PCSLoadCalculator2007 annual loading tables (see Table B-7 Appendix B).
Step 2: Apply Hybrid Methodology to ICIS-NPDES Load Calculator Annual Loads.
As discussed in Section 3.2.2.2, the output from the ICIS-NPDES Load Calculator includes two
annual loads, KGYE1 and KGY01, calculated using BDL=l/2 DL and BDL=0, respectively.
EPA applied the Hybrid Method that was used to calculate the PCSLoadCalculator2007 annual
loads to the KGYE1 and KGY01 annual loads from the ICIS-NPDES Load Calculator. See
Section 3.2.2.3 for more details on the hybrid methodology.
Step 3: Combine Annual Loads into DMR2007. In 2007, annual loads for 72 facilities
were in both PCSLoadCalculator2007 and the ICIS-NPDES Load Calculator. Because states are
currently migrating from PCS to ICIS-NPDES, EPA chose to use the annual loads from ICIS-
NPDES Load Calculator for the overlapping outfalls and pollutants16. Table B-8 in Appendix B
presents the list of facilities, outfalls, and pollutants in both PCS and ICIS-NPDES in 2007. EPA
used this methodology to combine the two sets of loads into the annual loads table, DMR2007.
The DMR2007 table indicates the data source for the calculated annual load.
Step 4: Calculate Toxic Weighted Pound Equivalent. To identify potential impacts on
human health and the environment, EPA estimated toxic equivalent mass discharge through the
use of TWFs. Section 5 of this report discusses TWFs in more detail. Chemicals for which EAD
(Engineering and Analysis Division) has developed TWFs are identified by CAS number. To
assign TWFs to reported discharges, EPA used the "SUPERCAS" table, developed in earlier
work with PCS and TRI data, to link CAS numbers to pollutant parameters reported in PCS.
EPA updated the SUPERCAS table to include ICIS-NPDES parameter codes. EPA has expanded
the SUPERCAS list of chemicals by identifying CAS numbers for priority pollutants and
chemicals that are frequently reported. EPA obtained the CAS numbers from
16Facilities may have some outfalls/pollutants that are in both PCS and ICIS-NPDES and other outfalls/pollutants in
PCS or ICIS-NPDES only. For example, chlorine data for outfall 001 may be in PCS and ICIS-NPDES, but
aluminum data for outfall 001 is only in PCS. In this example, EPA would use the chlorine load reported to ICIS-
NPDES and the aluminum load reported to PCS.
3-45
-------
Section 3 - DMRLoads2007: Development and Category Rankings
www.ChemFinder.com. During the 2009 annual review, as was done during previous annual
reviews, EPA made the following assumptions to assign CAS numbers to pollutant parameters:
• All forms of a pollutant were assigned the same CAS number (e.g., Dissolved
Copper, Total Recoverable Copper, and Total Copper (as Cu) were all assigned
the CAS number for Copper); and
• Chemicals that were reported in different ways were assigned only one CAS
number (e.g., Nitrate (as NOs) and Nitrate (as N) were both assigned the CAS
number for Nitrate.
EPA did not identify CAS numbers for chemicals infrequently reported. In addition, there are no
CAS numbers for non-chemical parameters reported in ICIS-NPDES and PCS (e.g., total
suspended solids, BOD5, COD, etc.).
Once the CAS numbers were assigned to each parameter using the expanded SUPERCAS
table, the TWFs were assigned by matching the CAS numbers. EPA did not assign TWFs to all
parameters reported in ICIS-NPDES and PCS. For the 2009 annual review, EPA continued to
estimate the TWFs for certain parameters that were reported as chemical groups based on
transfers from existing TWFs, as was done during previous annual reviews. Table 3-15 lists these
parameters and the method of TWF assignment.
Table 3-15. TWF Assignment for Chemical Mixtures
Parameter
Code
78216
82699
30383
34034
74105
Parameter Description
Aldrin + Dieldrin
Endrin + Endrin Aldehyde (Sum)
Benzene, Ethylbenzene, Toluene, and
Xylene
Chlorinated Phenols
Phenols, Chlorinated
Method of TWF assignment
Average of aldrin and dieldrin TWFs
Average of endrin and endrin aldehyde TWFs
Average of benzene, ethylbenzene, toluene, and xylene
TWFs
Average of the TWFs for PCS parameters 2,4,6-trichloro-
phenol, pentachlorophenol, 2,4-dichlorophenol, and 2-
chlorophenol (most common chlorinated phenols)
Average of the TWFs for PCS parameters 2,4,6-trichloro-
phenol, pentachlorophenol, 2,4-dichlorophenol, and 2-
chlorophenol (most common chlorinated phenols)
3.2.4.2 Sensitivity Analyses
As described in Section 3.2.2.2, EPA developed queries in PCSLoadCalculator2007 and
used annual loads output from ICIS-NPDES Load Calculator to calculate annual loads using the
DL=0 alternative method. For this method, the Load Calculators assume a discharge of zero for
pollutants that are labeled BDL. EPA combined the annual loads calculated using the DL=0
alternative method from PCSLoadCalculator2007 and ICIS-NPDES Load Calculator in
DMRLoadsAnalysis200 7.
During previous annual reviews EPA also calculated annual loads using the following
alternative methods:
3-46
-------
Section 3 - DMRLoads2007: Development and Category Rankings
• EST=NO. PCSLoadCalculator2007 assumes a discharge of zero for monitoring
periods where discharge data are missing.
• No Maximum (NOMAX). PCSLoadCalculator2007 used an alternative
measurement selection hierarchy, which set maximum concentrations (MCMX)
and maximum quantities (MQMX) to zero during the measurement value
selection process.
• No Concentration (NOCONC). PCSLoadCalculator2007 used an alternative
measurement selection hierarchy, which set average concentrations (MCAV),
minimum concentrations (MCMN), and maximum concentrations (MCMX) to
zero during the measurement value selection process.
EPA did not calculate the annual loads using these alternative methods as part of the 2009 annual
review based on the findings of the previous reviews. Section 6.1.2.3 of the Technical Support
Document for the Preliminary 2008 Effluent Guidelines Program Plan describes the results of
the 2007 annual review sensitivity analyses for EST=NO, NOMAX, and NOCONC (U.S. EPA,
2007).
Table 3-16 compares the assumptions and calculation options that the ICIS-NPDES Load
Calculator and PCSLoadCalculator2007 used to calculate each set of annual loads, including the
alternative loads that were not calculated as part of the 2009 annual review.
Table 3-16. Comparison of Alternative Load Calculation Methods
Annual Load Set
EST Option
DL Option
Measurements Included in
Selection Hierarchy a
Standard Load Calculation
DMR 2007 (PCS and
ICIS-NPDES)
EST=YES
Hybrid (DL=0 or DL=l/2)
MQAV MQMX
MCMN MCAV
MCMX
Alternative Load Calculations
DL=0 (PCS and
ICIS-NPDES)
EST=NO b
NOMAX b
NOCONC b
EST=YES
EST=NO
EST=YES
EST=YES
DL=0
Hybrid
Hybrid
Hybrid
MQAV MQMX
MCMN MCAV
MCMX
MQAV MQMX
MCMN MCAV
MCMX
MQAV MQMX=0
MCMN MCAV
MCMX=0
MQAV MQMX
MCMN=0 MCAV=0
MCMX=0
aFor the standard load calculation and DL=0 alternative load calculation in ICIS-NPDES, the measurements
included in the selection hierarchy are the five ICIS-NPDES measurements (MQ1, MQ2, MCI, MC2, and MC3).
bEPA did not calculate these annual loads as part of the 2009 annual review. Section 6.1.2.3 of the Technical
Support Document for the Preliminary 2008 Effluent Guidelines Program Plan describes the results of these
alternative load calculation methods (U.S. EPA, 2007).
3-47
-------
Section 3 - DMRLoads2007: Development and Category Rankings
EPA examined the impact of each calculation method, shown in Table 3-16, on the
calculated pollutant loads in a series of sensitivity analyses. To conduct each sensitivity analysis,
EPA calculated TWPE for loads calculated with each alternative method, and compared TWPE
calculated using the standard and alternative load calculation methods. EPA made this
comparison for total discharge and for the discharges separated into categories. EPA then
identified categories and individual facilities within a category that show a large difference
between DMR 2007 TWPE and alternative TWPE using the calculations shown below:
Amount of TWPE Based on Calculation Alternative (Ib-eq/yr) = Standard Load TWPE (Ib-eq/yr) -
Alternative Load TWPE (Ib-eq/yr)
Percent of TWPE Based on Calculation Alternative = Amount of TWPE Based on Calculation Alternative
(Ib-eq/yr) / Standard Load TWPE (Ib-eq/yr)
The following sections discuss the results of the DL sensitivity analyses based on combined 2007
PCS and ICIS-NPDES data.
DL Sensitivity Analysis
The purpose of the DL sensitivity analysis is to evaluate the impact of EPA's use of the
Hybrid Method, which estimates loads for some pollutants reported to PCS and ICIS-NPDES as
BDL, on the screening-level analysis. Table 3-17 presents a summary of the results of the DL
analysis for the point source categories showing the highest sensitivity to the DL options and the
total forDMRLoads2007. As shown in Table 3-17, only 0.12 percent (1,110,000 Ib-eq) of the
TWPE in DMRLoads2007 are based on BDL assumptions using the Hybrid Method. The
categories showing the greatest sensitivity to the DL options include Superfund Sites, the Pulp,
Paper and Paperboard Category, and the Petroleum Refining Category. For the complete results
of the DL sensitivity analysis see Table B-9 in Appendix B. Table B-10 in Appendix B presents
the results of the sensitivity analysis by pollutant. Pollutant parameters showing the highest
sensitivity to the DL options include 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD), TCDD
equivalents, and chlorine.
3-48
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-17. Results of DL Sensitivity Analysis
Point Source Category
Sanitary Services (SIC 4959)
Pulp, Paper, and Paperboard
(40 CFR Part 430)
Trucking and Warehousing
(SIC Group 42)
Pesticide Chemicals (40 CFR
Part 455)
Airport Deicing (PNC)
Petroleum Refining (40 CFR
Part 4 19)
Nonferrous Metals
Manufacturing (40 CFR Part
421)
Organic Chemicals, Plastics,
and Synthetic Fibers (40 CFR
Part 4 14)
Non-classifiable Establishments
(SIC Group 99)
Independent and Stand Alone
Labs (PNC)
Total DMRLoads2007
Total Number
of Facilities
2
217
2
147
5
108
36
219
10
6
2,177
Number of
Facilities
Affected by DL
1
78
1
28
2
65
14
89
2
4
1,025
Total Annual Load,
Ib/yr
653,000
2,450,000,000
83,300
3,840,000,000
1,160,000
1,950,000,000
188,000,000
1,480,000,000
24,800,000
465,000
43,100,000,000
Total Annual Load
Based on DL, Ib/yr
9,450 (1.5%)
70,800,000 (2.9%)
613 (0.7%
109,000 (0.002%)
30,000 (2.6%)
153,000,000 (7.8%)
30,700,000 (16%)
114,000,000(7.7%)
1,560 (0.01%)
10,100 (2.2%)
3,830,000,000
Total TWPE, Ib-
eq/yr
2.69
2,730,000
57.6
180,000
265
403,000
343,000
413,000
2,070
5,360
942,000,000
Total TWPE
Based on DL, Ib-
eq/yr
1.35 (50%)
347,000 (13%)
5.24(9.1%)
8,980 (5%)
9.02 (3.4%)
13,000 (3.2%)
10,200 (3%)
6,310(1.5%)
19 (0.9%)
27.7 (0.5%)
1,110,000
Source: DMRLoads2007_v3.
PNC - Potential new category.
3-49
-------
Section 3 - DMRLoads2007: Development and Category Rankings
3.2.5 DMRNutrients2007
DMRNutrients2007 uses the annual loads for nitrogen and phosphorus compounds from
the DMR2007 table to calculate aggregate nitrogen as N and phosphorus as P loads for each
facility outfall. The database sums the aggregate nitrogen and phosphorus loads by facility and
by point source category. Table B-l 1 of Appendix B presents the category rankings for total
nitrogen as N loads and Table B-l2 presents the category rankings for total phosphorus as P
loads.
DMR data include discharges of nitrogen and phosphorus in various chemical forms. For
example, nitrogen may be reported in its elemental form (as N), total Kjeldahl nitrogen (TKN),
organic nitrogen, ammonia as N, ammonia as NH3 or NH4, un-ionized ammonia, nitrite, or
nitrate. EPA developed a series of hierarchies to select the appropriate combination of nitrogen
and phosphorus compounds to calculate the total nitrogen and total phosphorus loads. These
hierarchies, summarized below, are described in detail in "Point Source Category Rankings by
Nitrogen and Phosphorus Loads Calculated Using 2002 PCS Data" (Kandle, 2005a).
use):
Total Nitrogen Load
EPA calculated total nitrogen using one of the following equations (presented in order of
• Total Nitrogen Load = Total Nitrogen as N;
• Total Nitrogen Load = TKN + Nitrite (NO2) + Nitrate (NO3); or
• Total Nitrogen Load = Organic Nitrogen + Ammonia + Nitrite + Nitrate.
Nitrogen compounds that are reported as NH3, NH4, NC>2, or NO3 were converted to N
based on molecular weight, then summed to calculate Total Nitrogen Load. Table 3-18 presents
the conversion factors EPA used for nitrogen compounds.
Table 3-18. Conversion Factors for Nitrogen Compounds
Nitrogen Compound
Ammonia as NH3
Nitrite as NO2
Nitrate as NO3
Conversion Factor
14N/17NH3
14 N / 46 NO2
14 N / 62 NO3
Total Phosphorus Load
Loads for phosphorus parameters were grouped by EPA's grouping hierarchy described
in Section 3.2.2 and assigned to a grouped parameter code. As a result, DMRNutrients2007
includes only two parameters for phosphorus compounds. EPA used the following hierarchy to
calculate total phosphorus load:
• If loads of phosphorus (PRAM PHOSP) were available, EPA used the PRAM
PHOSP load to represent total phosphorus. EPA assumed that the majority of the
loads were reported as phosphorous and did not apply a conversion factor to
calculate pounds of phorphosour.
3^50
-------
Section 3 - DMRLoads2007: Development and Category Rankings
• If loads of phosphorus (PRAM PHOSP) were not available, EPA used loads of
phosphate (PRAMs PO4). EPA multiplied the load by 31/95 to convert the
reported phosphate load to pounds of phosphorous.
3.2.6 DMRLoads2007
As the final step in developing DMRLoads2007, EPA grouped discharges from
DMRLoadAnalysis2007 to create the point source category rankings and to perform other
analyses. Section 3.2.6.1 discusses the tables and table structure ofDMRLoads2007, and Section
3.2.6.2 discusses the SIC/Point Source Category Crosswalk in relation to DMRLoads2007 tables.
3.2.6.1 DMRLoads200 7 Structure
Table 3-19 lists and describes the tables in DMRLoads2007'.
Table 3-19. Tables Imported or Created in DMRLoads2007
Table Name
PRAM Codes
SIC/Point Source
Category Crosswalk
Point Source Category
Codes
SIC Codes
SUPERCAS Category
TWFs
DMRFAC
DMR2007
Manual ICIS Loads
Corrections
Category Rankings -
Nitrogen
Category Rankings -
Phosphorus
SIC Code Rankings
Category Rankings -
Toxic Weight
Created or Imported
Linked from
DMRLoadsAnalysis2007
Linked from
DMRLoadsAnalysis2007
Linked from
DMRLoadsAnalysis2007
Linked from
DMRLoadsAnalysis2007
Linked from
DMRLoadsAnalysis2007
Linked from
TRICalculations2007
Linked from
DMRLoadsAnalysis2007
Linked from
DMRLoadsAnalysis2007
Created
Linked from
DMRNutrietns2007
Linked from
DMRNutrients2007
Created using queries
Created using queries
Description
Lists pollutants and corresponding parameter codes.
Links SIC codes with point source categories using a
numeric code assigned in the Point Source Category Codes
table.
Assigns a numeric code to industrial categories using their
40 CFR part number or 2-digit or 4-digit SIC Code.
Lists SIC codes and their descriptions.
Links CAS numbers to pollutant parameter codes.
Assigns TWF values to chemicals by CAS number.
Presents information on permitted facilities, such as facility
name, location, major/minor discharge status, and date of
most recent permit issuance
Presents the annual loads in pounds per year and TWPE for
each pollutant discharge for each outfall at major permitted
facilities.
Lists ICIS-NPDES loads corrections identified by manual
review.
Presents rankings of categories based on aggregated
nitrogen load.
Presents rankings of categories based on aggregated
phosphorus load.
Presents rankings of SIC codes based on calculated TWPE.
Presents rankings of categories based on calculated TWPE.
3-51
-------
Section 3 - DMRLoads2007: Development and Category Rankings
3.2.6.2 SIC/Point Source Category Crosswalk
DMRLoads2007 assigns a facility's discharge to an industrial category using 4-digit SIC
codes. Point source categories are not generally defined by SIC codes. As a result, EPA
developed a crosswalk that links point source categories to 4-digit SIC codes, described in
Section 1 of this document. EPA has developed ELGs for point source discharges from 56
specific categories. The point source categories, which may be divided into subcategories, are
generally defined in terms of combinations of products made and the processes used to make
these products. Facilities with data in PCS and ICIS-NPDES are identified by SIC code. Thus, to
use the PCS and ICIS-NPDES data to estimate the pollutants discharged by each point source
category, EPA assigned each 4-digit SIC code to an appropriate point source category using the
"SIC/Point Source Category Crosswalk" table. See Section 4.2 for additional information on the
development of the SIC/Point Source Category Crosswalk.
As shown in Figure 3-7, DMRLoads2007 links information from the DMR2007 Table,
DMRFAC, and the SIC/Point Source Category Crosswalk to create point source category
rankings. The SIC codes in the DMR2007 Table are specific to each parameter, discharge pipe
(outfall), and facility (NPDES permit number). This allows EPA to make SIC adjustments to
differentiate between various operations/outfalls at one facility and assign discharges at the
pollutant level to different point source categories, as described in Section 4.2.1.2.
Sum Annual Loads and TWPE by Point Source Category
Category Rankings - Toxic Weight
Point Source Category Code
Category Name
Type of Group
Majors
Total Annual Pounds
Figure 3-7. DMRLoads2007 Database Structure
3.2.7 Database Corrections
EPA reviewed DMRLoads2007 output for reasonableness, as described in Section 3.4.
Also, during previous screening-level analyses, EPA identified facility-specific corrections for
PCS data. Several of these corrections similarly apply to the 2007 DMR data. In addition, EPA's
quality review (see Section 3.4) identified other corrections to the 2007 DMR data, (e.g., units
incorrectly reported as gallons per day were corrected to MOD). Table B-13 in Appendix B lists
all corrections made to the 2007 DMR data. In addition to the facility-specific data corrections,
DMRLoadsAnalysis2007 performs the following modifications to the annual loads:
3-52
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Categorization of Discharges. Section 1 of this report describes the development
of the SIC/Point Source Category Crosswalk, which EPA uses to link between
facility SIC codes and categories with existing ELGs. Because most point source
categories are not defined by SIC code, the relationship between SIC code and
point source category is not a one-to-one correlation. A single SIC code may
include facilities in more than one point source category, and associating an SIC
code with only one category may be an over simplification. Also, many facilities
have operations subject to more than one point source category. Further, facilities
in some categories cannot be identified by SIC code (e.g., Centralized Waste
Treatment facilities). The database changes are summarized below:
— Facility-Level Point Source Category Assignment. For some SIC codes
that include facilities subject to guidelines from more than one point
source category, EPA was able to assign each facility to the category that
best applied to the majority of its discharges. EPA reviewed information
available about each facility to determine which point source category
applied to the facility's operations.
— Pollutant-Level Point Source Category Assignment. Many facilities have
operations subject to more than one point source category. For most of
these facilities, EPA cannot divide the pollutant discharges among the
applicable point source categories. Two exceptions where EPA was able
to assign wastewater discharges of certain chemicals to the appropriate
point source category include Organic Chemicals, Plastics, and Synthetic
Fibers (OCPSF) /Pesticides and MP&M/Metal Finishing (see Section
4.2.1.2 for additional discussion).
Table 3-20 shows the facilities for which EPA revised SIC codes to link to an
appropriate point source category based on known plant operations.
Internal Monitoring. As described in Sections 3.2.2.2 and 3.2.3.2,
PCSLoadCalculator2007 and the ICIS-NPDES Load Calculator calculated loads
only for monitoring locations that are labeled as effluent (MLOC 1 or 2 in PCS
and MLOC 1, 2, A, B, or SC in ICIS-NPDES). For the 2009 annual review, EPA
included only MLOC 1 and 2 for ICIS-NPDES. EPA will include MLOC A, B,
and SC in future annual reviews. As a result, the Load Calculators exclude
discharges for internal monitoring locations such as intake water, influent to
treatment, and intermediate points in the wastewater treatment system. However,
during previous category reviews and detailed studies, EPA identified instances of
double counting that resulted from including certain internal monitoring points in
the loads database. For example, a facility monitors for Pollutant A at the effluent
from its wastewater treatment system (Internal Outfall 101). Outfall 101
wastewater is later combined with other plant discharges at final Outfall 001 and
is discharged to a receiving stream. The facility also monitors for Pollutant A at
Final Outfall 001. Both outfalls are effluent monitoring points identified as
MLOC 1 or MLOC 2; however, Outfall 101 is upstream of the final outfall.
Calculating loads for Pollutant A at both the internal and final outfalls results in
double counting Pollutant A discharges. EPA identified instances where pollutant
discharges are reported for multiple monitoring locations along the same
3^53
-------
Section 3 - DMRLoads2007: Development and Category Rankings
discharge line, and eliminated the discharges for the upstream monitoring
locations. EPA made these corrections in PCSLoadCalculator2007 for the PCS
data and in DMRLoadsAnalysis2007 for the ICIS-NPDES data (see Table B-13 in
Appendix B).
• Intermittent Discharges. As described in Sections 3.2.2.2 and 3.2.3.2, in
PCSLoadCalculator2007 and the ICIS-NPDES Load Calculator EPA assumes
that all discharges in PCS and ICIS-NPDES are continuous. During previous
Annual Reviews, EPA identified facility discharges that are intermittent and
therefore overestimated by the Load Calculator. EPA calculated annual loads for
these discharges based on information obtained from the facility on the frequency
and duration of wastewater discharges. EPA made these corrections in
PCSLoadCalculator2007 for the PCS data and in DMRLoadsAnalysis2007 for the
ICIS-NPDES data (see Table B-13 in Appendix B).
• Pollutant Parameters Excluded from DMRLoads2007. Parameters in PCS and
ICIS-NPDES include water quality parameters (e.g., dissolved oxygen and
temperature), specific chemicals (e.g., phenol), bulk parameters (e.g., biochemical
oxygen demand), and flow. As described in Sections 3.2.2.2 and 3.2.3.2, facilities
report pollutant mass quantities, pollutant concentrations, and wastewater flow
rates to PCS and ICIS-NPDES using a variety of units. EPA's PCS CNVRT
program and the ICIS-NPDES convert module convert the discharges into
standard units of kilograms per day for mass quantities, milligrams per liter for
concentrations, and millions of gallons per day for flow rates. However, some
parameters are reported in units that cannot be converted into kg/day or mg/L
(e.g. temperature and pH). EPA excluded these parameters from the screening-
level analysis. Table B-14 of Appendix B lists the excluded parameters.
• ICIS-NPDES Load Corrections. In some cases, EPA identified that loads are
not estimated correctly because of errors in units, number of reporting periods,
detection limit estimation, improperly-coded outfalls, or other data entry errors.
For data in PCS, EPA made corrections in the PCSLoadCalculator database. For
data from ICIS-NPDES, EPA corrected loads in DMRLoadsAnalysis2007'. Table
B-13 in Appendix B lists these case-by-case ICIS Loads corrections.
3-54
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-20. Case-by-Case Point Source Category Reassignments in DMRLoads2007
NPID
TN0002968
OH0048836
MD0000060
MO0002356
MS0046931
MS0002941
AL0000213
AL0001945
AL0003514
DE0050911
KY0003484
KYO 102083
LA0005231
LA0005983
LA0029769
ME0000639
NV0020923
NY0001635
NY0003328
NY0003336
OHO 11 5401
TN0002461
TX0007412
TX0008150
WI0003565
WV0000108
WV0004359
Name
US DOE-Oak Ridge Y12 Plant
Duke Energy, Ohio, Inc.
Perdue Farms, Inc.
BCP Ingredients, Inc
Scott County
Lawrence County
Occidental Chemical Corp
Olin Chlor Alkali Products
Occidental Chemical Corp
Diamond Shamrock Chemicals
Co.
Westlake Ca&O Corp
USEC PDGDP
Pioneer Chlor Alkali Co., Inc
Occidental Chemical Corp
IMC-Phosphates Company
Holtachem Mfg
Pioneer Americas-BMI Complex
Olin Corp - Niagara Falls Pit
E I Dupont De Nemours & Co,
Inc
Occidental Chemical Corp
Us Enrichment Corp Ports Gaseo
Olin Chemicals Corp.
Deer Park Plant
Oxy Vinyls, Lp, Harris County
Erco Worldwide Usa Inc Pt Edw
Kincaid Enterprises
Ppg Industries, Inc.
Old SIC
Code
9611
4932
2048
2048
2048
2861
2813
2812
2812
2812
2812
2819
2812
2812
2819
2812
2812
2812
2812
2812
2819
2812
2812
2812
2812
2819
2812
Old Point Source Category
Administration Of Economic Programs
Electric, Gas, & Sanitary Services
Food & Kindred Products
Food & Kindred Products
Food & Kindred Products
Gum And Wood Chemicals
Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
Inorganic Chemicals Manufacturing
New SIC Code
a
3499
4911
2048GRAIN
2048GRAIN
2048MPP
2621-2
VCCA
VCCA
VCCA
VCCA
VCCA
2819NMM
VCCA
VCCA
2873
9999
VCCA
VCCA
VCCA
VCCA
2819NMM
VCCA
VCCA
VCCA
VCCA
2879
VCCA
New Point Source Category
Metal Finishing
Steam Electric Power Generating
Grain mills
Grain mills
Meat and Poultry Products
Pulp, Paper And Paperboard
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Nonferrous Metals Manufacturing
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Fertilizer Manufacturing
Non Classifiable Establishments
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Nonferrous Metals Manufacturing
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Pesticide Chemicals
Chlorine And Chlorinated Hydrocarbons
3-55
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-20. Case-by-Case Point Source Category Reassignments in DMRLoads2007
NPID
AR0037800
LA0038245
LA0065501
MOO 108472
TN0074225
TX0005941
TX0030937
TX0091855
ME0001872
IL0001724
TN0002488
TN0003671
AL0002658
AL0026832
CO0042480
MI0004227
NY0006548
OH0004219
OH0004260
OH0098540
AL0054704
DE0000612
IL0001350
IN0002101
Name
Clean Harbors El Dorado, LLC
Clean Harbors Baton Rouge, LLC
Clean Harbors White Castle LLC
Front St Remedial Action
Ettp-Central Neutraliz. Fac
Clean Harbors Deer Park WWTP
Vopak Logistics Services USA
Stolthaven Houston, Inc.
Domtar Maine Corporation
American Nickeloid Co-Peru
State Ind- Ashland Cty
Usa Holston Army Ammo Pit
Area
Anniston Army Depot
Golden Rod Broilers
Eagle Mine Remediation WWTP
Dsc Ltd
Owl Wire & Cable Inc - Rome
Fac
Timken Company - Canton
AK Steel Coshocton Stainless
Reserve Environmental Services
Sabic Innovative Plactics
Formosa Plastics Corporation
Formosa Plastics-Illinois
Sabic Innovative Plastics Mt Ve
Old SIC
Code
4953
4953
4953
4953
4953
4953
4953
4953
2411
3471
3639
9711
9999
9999
9999
9999
9999
9999
9999
9999
2821
2821
2821
2821
Old Point Source Category
Landfills
Landfills
Landfills
Landfills
Landfills
Landfills
Landfills
Landfills
Lumber & Wood Products
Metal Finishing
Metal Finishing
National Security & International
Affairs
Non Classifiable Establishments
Non Classifiable Establishments
Non Classifiable Establishments
Non Classifiable Establishments
Non Classifiable Establishments
Non Classifiable Establishments
Non Classifiable Establishments
Non Classifiable Establishments
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
New SIC Code
a
4953WC
CWT
CWT
SUPER
4953WC
4953WC
CWT
CWT
2411-1
3471CC
3639PE
2892
3795
2015
9999
3316
3351
3562
3312
CWT
VCCA
VCCA
VCCA
VCCA
New Point Source Category
Waste Combustors
Centralized Waste Treatment
Centralized Waste Treatment
Superfund Sites
Waste Combustors
Waste Combustors
Centralized Waste Treatment
Centralized Waste Treatment
Pulp, Paper And Paperboard
Coil Coating
Porcelain Enameling
Explosives Manufacturing
Metal Finishing
Meat and Poultry Products
Non Classifiable Establishments
Iron And Steel Manufacturing
Copper forming
Metal Finishing
Iron And Steel Manufacturing
Centralized Waste Treatment
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
3-56
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-20. Case-by-Case Point Source Category Reassignments in DMRLoads2007
NPID
LA0000761
LA0002933
LA0003301
LA0006149
LA0006220
LA0007129
LA0041025
LA0056171
NJ0004286
NJ0004391
OH0007269
TX0002798
TX0006335
TX0006483
TX0070416
TX0085570
Name
PPG - Lake Charles
Occidental Chemical Corp.
Dow Chemical - Plaquemine
Formosa Plastics Corp
Crompton Manufacturing Co.
Georgia Gulf Corporation
Certainteed Corporation
Occidental Chemical Corporatio
Polyone Corporation
Colorite Polymers Company
Dover Chemical Subsidiary Of I
Wwtp
Oxy Vinyls, Lp
Dow Chemical
Vinyl Chloride Monomer Plant
Formosa Point Comfort Plant
Old SIC
Code
2869
2869
2869
2869
2869
2869
2821
2869
2821
2821
2869
2869
2821
2869
2869
2869
Old Point Source Category
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
Organic Chemicals, Plastics And
Synthetic Fibers
New SIC Code
a
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
VCCA
New Point Source Category
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
3-57
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-20. Case-by-Case Point Source Category Reassignments in DMRLoads2007
NPID
TXO 104876
CT0003212
OK0040827
MS0001309
TX0062677
LA0004847
MS0003115
GA0046973
OK0034321
WI0001848
OR0020834
CO0042064
LA0038245
LA0065501
LA0066214
MOO 108472
NJ0005240
TX0030937
TX0091855
MO0001716
Name
Organic Chemical Manufacturing
Kimberly-Clark Corporation
Kimberly-Clark Corp-Jenks Fac
Adams County
North Regional Treatment Plant
Mosaic Fertilizer, LLC
Jackson County
Fort James Operating Company
Fort James Oprating Co-Muskoge
Georgia Pacific Consumer Prod
St. Helens STP/Boise Cascade
Treatment, Storage & Disposal
Clean Harbors Baton Rouge, LLC
Clean Harbors White Castle LLC
NPC Services-Alsen
Front St Remedial Action
Safety-Kleen - Bridgeport
Vopak Logistics Services USA
Stolthaven Houston, Inc.
Basf Hannibal Plant
Old SIC
Code
2869
2676
2676
2911
2911
2874
2874
2621
2621
2621
4952
4953
4953
4953
4953
4953
4953
4953
4953
5191
Old Point Source Category
Organic Chemicals, Plastics And
Synthetic Fibers
Paper & Allied Products
Paper & Allied Products
Petroleum Refining
Petroleum Refining
Phosphate Manufacturing
Phosphate Manufacturing
Pulp, Paper And Paperboard
Pulp, Paper And Paperboard
Pulp, Paper And Paperboard
Sewerage Systems
Waste Combustors
Waste Combustors
Waste Combustors
Waste Combustors
Waste Combustors
Waste Combustors
Waste Combustors
Waste Combustors
Wholesale Trade- Nondurable Goods
New SIC Code
a
VCCA
2621-2
2611-2
2611-2
CWT
2874FER
2874FER
VCCA
VCCA
VCCA
2621-1
4953L
CWT
CWT
4953L
SUPER
4953L
CWT
CWT
2879
New Point Source Category
Chlorine And Chlorinated Hydrocarbons
Pulp, Paper And Paperboard
Pulp, Paper And Paperboard
Pulp, Paper And Paperboard
Centralized Waste Treatment
Fertilizer Manufacturing
Fertilizer Manufacturing
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Chlorine And Chlorinated Hydrocarbons
Pulp, Paper And Paperboard
Landfills
Centralized Waste Treatment
Centralized Waste Treatment
Landfills
Superfund Sites
Landfills
Centralized Waste Treatment
Centralized Waste Treatment
Pesticide Chemicals
Source: DMRLoads2007_v3.
a Because some point source categories correspond to multiple SIC codes, some changes to SIC codes did not result in a change in point source category
assignment. These SIC changes are not shown in the table.
3-58
-------
Section 3 - DMRLoads2007: Development and Category Rankings
3.3 Results of the Preliminary Analysis
This section presents the results oftheDMRLoads2007 database. Table 3-22 presents the
categories ranked from highest to lowest TWPE. Table B-l of Appendix B presents the four-
digit SIC code rankings by TWPE. Table B-2 of Appendix B presents the total TWPE for
pollutant parameters reported in DMR.
3.4 Data Quality Review
EPA evaluated the quality of the PCS and ICIS-NPDES DMR data for use in the 2009
screening-level review. This evaluation considered data completeness, accuracy, reasonableness,
and comparability. The Quality Assurance Project Plan for the 2009 Annual Screening-Level
Analysis of TRI and PCS Industrial Category Discharge Data (ERG, 2009) describes the quality
objectives in more detail. EPA conducted quality reviews for four stages of the development of
DMRLoads2007: PCS CNVRT program output; ICIS-NPDES Convert Module output;
PCSLoadCalculator2007 and the ICIS-NPDES Pollutant Loading Tool output; and
DMRLoads2007 results. The following discussion provides an overview of the quality review
steps for each stage:
• PCS CNVRT program output. EPA's quality review of the CNVRT output files
included reasonableness checks of pollutant quantities and concentrations. EPA
reviewed the CNVRT program output (i.e., the pollutant discharges stored in PCS
converted into standard units of kg/day and mg/L) to identify possible errors in
recording units of measure. EPA reviewed ranges of pollutant quantities and
concentrations and identified pollutant measurements and flows that were
unreasonably high. EPA then compared these measurements with measurements
available on EPA's Envirofacts web page. If the measurements were similar EPA
concluded that the CNVRT file output was acceptable. This review resulted in
two types of systematic corrections to the CNVRT output:
— Corrections to 1,015 flows ranging from 1,300 MOD to 5,000 MOD17 (see
Section 3.2.2.2); and
— Corrections to 284 mercury concentrations reported to PCS using PRAM
50092 (Mercury Total Low Level) (see Section 3.4.1).
• ICIS-NPDES Convert Module output. EPA conducted an initial quality review
of the extracted ICIS-NPDES DMR data to evaluate its completeness,
reasonableness, and comparability. For completeness, EPA compared the number
of major facilities and the universe of SIC codes in the 2007 ICIS-NPDES DMR
data to the PCS DMR data in 2004, the last complete DMR data set for ICIS-
NPDES states. The 2007 ICIS-NPDES data had at least as many majors and SIC
codes as PCS in 2004. Additionally, EPA verified that, while PCS 2004 had more
parameter codes than ICIS-NPDES in 2007, all commonly reported parameters
are present in the 2007 ICIS-NPDES DMR data.
17 In addition to these systematic flow corrections, EPA determined that all flows between 1,000 and 5,000 MOD
reported by facilities in Ohio were flows in GPD. EPA automatically divided these flows by 1,000,000. However,
because power plants are known to have high flows, EPA made flow corrections to Ohio facilities reporting SIC
code 4911 (Electrical Services) on a case-by-case basis.
3-59
-------
Section 3 - DMRLoads2007: Development and Category Rankings
EPA reviewed the DMR data for reasonableness to identify any data quality
issues, such as misreported units that the ICIS-NPDES Convert Module did not
correct. EPA identified several wastewater flows that exceeded the reasonable
range. EPA reviewed these flows and developed the proposed flow correction
function for the ICIS-NPDES Convert Module (described in Section 3.2.3.1). This
function is designed to identify data entry errors for flows greater than 1,000
MOD. The ICIS-NPDES Convert Module corrects all flows exceeding 5,000
MOD, and applies more conservative criteria to correct flows from 1,000 to 5,000
MOD. The ICIS-NPDES Convert Module made following corrections to ICIS-
NPDES wastewater flows:
— 1,113 corrections based on month-to-month variations;
— 1,605 corrections based on comparing flows to design flows; and
— 142 corrections based on assuming that flows exceeding 5,000 MGD were
reported in units of GPD.
EPA also evaluated the comparability of the extracted 2007 ICIS-NPDES DMR
data to the 2004 PCS data. As shown in Table 3-21, most of the average loads and
concentrations in ICIS-NPDES are within one order of magnitude of the 2004
PCS data. However, the maximum loads and concentrations indicate that there
may be some unreasonable values in the 2007 ICIS-NPDES DMR data. EPA
verified the unit conversions used in the ICIS-NPDES Convert Module and for
this reason concluded that the unreasonable flows and pollutant measurements are
likely the result of data entry errors and are not the result of any errors in the
ICIS-NPDES Convert Module functions.
Table 3-21. Comparison of Load and Concentration Ranges for Common Parameters
Pollutant
Aluminum
Ammonia
BOD
Chlorine
Copper
Iron
Nitrogen
Oil and
Grease
Phosphate
Phosphorus
TKN
TSS
Zinc
Average
2004 PCS
Quantity
(kg/day)
77
154
1,296
0.90
0.42
241
53
41
43
583
65
457
1.89
Maximum
2004 PCS
Quantity
(kg/day)
4,755
1,873
275,456
692
244
51,812
19,985
5,953
5,953
131,464
5,117
131,429
1,046
Average
2004 PCS
Cone.
(mg/L)
19
2.8
1,773
0.5
2.7
2.7
6.9
3.2
3.3
20
15
838
1.3
Maximum
2004 PCS
Cone.
(mg/L)
3,333
116
6,690,000
1,420
1,042
19,450
2,701
9,400
9,400
8,104
3,400
73,500
1,360
Average
2007
ICIS-
NPDES
Quantity
(kg/day)
15
343
802
1.5
0.20
165
818
195
6.5
39
298
574
0.67
Maximum
2007 ICIS-
NPDES
Quantity
(kg/day)
72
7,082
3,490,205
1,647
123
46,530
43,584
10,651
30
152,101
9,403
3,405,402
1,308
Average
2007
ICIS-
NPDES
Cone.
(mg/L)
10
8.8
134
0.46
0.11
3.6
20
5.4
4.4
2.1
2,348
93
0.46
Maximum
2007
ICIS-
NPDES
Cone.
(mg/L)
5,620
147
290,171
60,000
850
2,800
114,598
380
93
20,990
240,000
380,800
769
Sources: PCSLoadCalculator2004 and the ICIS-NPDES Pollutant Loading Tool.
3^60
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-22. DMR 2007 Point Source Category Rankings by TWPE
40CFR
Part
NA
423
433
430
414.1 a
418
420
432
414
419
415
421
440
455
471
NA
410
429
417
NA
444
445
409
436
439
463
422
467
464
428
454
437
469
411
NA
NA
NA
Point Source Category
Superfund Sites
Steam Electric Power Generating
Metal Finishing
Pulp, Paper And Paperboard
Chlorine And Chlorinated Hydrocarbons
Fertilizer Manufacturing
Iron And Steel Manufacturing
Meat and Poultry Products
Organic Chemicals, Plastics And
Synthetic Fibers
Petroleum Refining
Inorganic Chemicals Manufacturing
Nonferrous Metals Manufacturing
Ore Mining And Dressing
Pesticide Chemicals
Nonferrous Metals Forming And Metal
Powders
Drinking Water Treatment
Textile Mills
Timber Products Processing
Soap And Detergent Manufacturing
National Security & International Affairs
Waste Combustors
Landfills
Sugar Processing
Mineral Mining And Processing
Pharmaceutical Manufacturing
Plastics Molding And Forming
Phosphate Manufacturing
Aluminum forming
Metal Molding And Casting (Foundries)
Rubber Manufacturing
Gum And Wood Chemicals
Manufacturing
Centralized Waste Treatment
Electrical And Electronic Components
Cement Manufacturing
Engineering & Management Services
Miscellaneous Foods And Beverages
Independent And Stand Alone Labs
Number of
Facilities
1
550
111
217
40
21
90
44
219
108
55
36
54
139
14
13
48
5
2
35
10
10
21
34
28
6
12
12
7
17
2
6
5
6
1
8
6
Total Pounds
1,331,644
25,138,490,268
77,793,914
2,449,186,965
1,583,223,789
125,646,884
672,371,411
673,799,975
1,484,731,242
1,952,697,634
1,173,945,339
187,923,634
470,835,865
3,843,462,966
5,496,943
1,135,551,072
29,467,857
99,888,774
230,007
92,583,865
19,162,733
18,668,498
698,918,657
264,924,182
43,720,508
88,969,293
62,276,423
15,781,323
6,019,649
8,947,786
838,168
120,470,939
2,665,896
63,110,706
3,284,525
94,045,452
465,432
Total TWPE
909,115,642
20,374,829
3,361,768
2,726,865
1,220,744
1,095,046
730,252
535,913
413,226
402,506
393,523
342,747
184,455
179,697
119,244
119,190
79,934
51,552
47,815
38,983
38,412
35,804
32,520
26,719
24,937
24,626
18,459
12,182
11,271
11,195
10,478
10,403
9,350
8,960
5,978
5,842
5,355
3-61
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-22. DMR 2007 Point Source Category Rankings by TWPE
40CFR
Part
424
408
468
434
NA
406
407
443
461
NA
NA
NA
457
NA
412
NA
NA
426
NA
NA
NA
NA
435
NA
465
NA
405
NA
NA
460
NA
466
425
451
438
NA
4959
NA
Point Source Category
Ferroalloy Manufacturing
Canned And Preserved Seafood
Processing
Copper forming
Coal Mining
Non Classifiable Establishments
Grain mills
Canned And Preserved Fruits And
Vegetables Processing
Paving And Roofing Materials (Tars And
Asphalt)
Battery Manufacturing
Amusement & Recreation Services
Printing & Publishing
Environmental Quality & Housing
Explosives Manufacturing
General Building Contractors
CAFO
Justice, Public Order, & Safety
Educational Services
Glass Manufacturing
Special Trade Contractors
Construction And Development
Lumber & Wood Products
Airport Deicing
Oil & Gas Extraction
Real Estate
Coil Coating
Executive, Legislative, & General
Dairy products processing
Trucking & Warehousing
Wholesale Trade- Durable Goods
Hospital
Pipelines, Except Natural Gas
Porcelain Enameling
Leather Tanning And Finishing
Concentrated Aquatic Animal Production
Metal Products And Machinery
Tobacco Products
Sanitary Services
Transportation Services
Number of
Facilities
3
8
9
9
10
14
11
4
1
1
2
5
5
1
1
9
5
o
6
i
2
1
5
5
9
1
2
o
6
2
2
2
1
1
1
23
2
1
2
1
Total Pounds
7,905,371
124,735,909
2,928,183
44,228,933
24,794,788
28,624,175
7,177,438
494,518
136,061
118,566
1,039,175
5,849
21,980,426
41,817
10,812,796
1,351,009
4,934,978
2,715,981
8,073,573
28,460,736
8,975,046
1,162,405
531,118
4,857,073
445
53,109
262,241
83,278
538,559
9,134
289,497
13,507
33,076
5,310,357
1,187,703
10,740
653,114
713,322
Total TWPE
4,349
3,232
2,310
2,294
2,066
1,984
1,757
1,280
1,096
1,025
999
972
785
645
617
505
410
353
330
324
283
265
256
214
166
77
76
58
30
15
12
11
8
5
3
o
J
3
o
5
3-62
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-22. DMR 2007 Point Source Category Rankings by TWPE
40CFR
Part
NA
459
442
NA
Point Source Category
Photo Processing
Photographic
Transportation Equipment Cleaning
Wholesale Trade- Nondurable Goods
Number of
Facilities
1
1
2
1
Total Pounds
34,136
34,136
326,427
33,166
Total TWPE
1
1
0
0
Source: DMRLoads2007_v3.
a414.1 refers to the chlorinated hydrocarbon segment of the Organic Chemicals, Plastics, and Synthetic Fibers
Category (40 CFR Part) 414 and the Chlor-Alkali Subcategory of the Inorganic Chemicals Manufacturing Category
(40 CFR Part 415).
NA - Not applicable; no existing ELGs apply to discharges.
• Load Calculator routines. EPA's quality review for the Load Calculator routines
included accuracy checks for database queries onPCSLoadCalculator2007. EPA
reviewed the programming code used to develop each query to verify the logic
and verified that the number of records in the output table equaled the number of
records in intermediate queries to ensure that no data were missing and that there
were no duplicate data. EPA also verified the Load Calculator routine in the ICIS-
NPDES Pollutant Loading Tool. EPA created a query-based system and
compared the annual loads calculated by the queries to those calculated by the
ICIS-NPDES Pollutant Loading Tool. The output from the queries was identical
to that of the ICIS-NPDES Pollutant Loading Tool. In addition, EPA performed
hand calculations to verify the accuracy of the PCSLoadCalculator2007 and
ICIS-NPDES Load Calculator Module outputs during reviews of facility
discharges for DMRLoads2007 results.
• DMRLoads2007 results. EPA's quality review of the DMRLoads2007 results
included the following:
— Completeness checks: EPA compared counts of dischargers in
DMRLoads2007 to PCSLoads2004 to describe the completeness of the
database. There were 2,027 facilities that reported a load to
PCSLoads2004 and 2,018 facilities that reported a load to
DMRLoads2007'. Therefore, EPA determined DMRLoads2007 was
complete.
— Accuracy of facility discharges. EPA reviewed the accuracy of facilities'
discharges that had the greatest impact on total category loads and
category rankings to identify possible calculation errors. EPA reviewed
monthly information reported in PCS and ICIS-NPDES, measurement
data available on EPA's Envirofacts Web page, and information from the
facility's NPDES permit. In some cases, EPA contacted facilities to verify
the monthly measurements in their DMR. Section 3.4.2 describes EPA's
review of facility discharges in more detail.
— Accuracy of category discharges. EPA reviewed the accuracy of category
discharges by verifying that pollutant discharges in PCS and ICIS-NPDES
3-63
-------
Section 3 - DMRLoads2007: Development and Category Rankings
were assigned to the appropriate point source category. EPA used
engineering judgment to determine if the pollutant discharge was
reasonably associated with the point source category. Section 3.2.7
discusses facility-level and pollutant-level category assignments.
Accuracy of database queries. EPA's quality review for the development
of DMRLoads2007 included accuracy checks for database queries in
DMRLoadsAnalysis2007, DMRNutrients2007, andDMRLoads2007.
Documentation of accuracy checks is provided in a QC table in each
Microsoft Access database.
Reasonableness of pollutant loads. EPA reviewed the Load Calculator
output (i.e., the calculated kg/year for each pollutant at each discharge
pipe and monitoring location) for those pollutant discharges with the
highest toxic-weighted loads (e.g., dioxins, PCBs, and mercury). To
identify possible errors in recording units of measure, EPA identified
calculated discharges that were orders of magnitude higher than previous
years' discharges and other facilities within the same category. EPA
reviewed quantities or concentrations and flows that the
PCSLoadCalculator2007 and ICIS-NPDES Pollutant Loading Tool
databases used to calculate the annual discharge. EPA compared these
measurements with measurements available on EPA's Envirofacts web
page. If the measurements were similar then EPA concluded that the
output was acceptable. If the data did not match between the databases and
Envirofacts, EPA corrected the data to match Envirofacts. When EPA was
unsure of the correct data, EPA contacted the facility for more information
(see Section 3.4.2).
Reasonableness of facility loads. EPA identified facility discharges with
the highest TWPE and nutrient pollutant loads. EPA identified facilities
for review whose pollutant discharges accounted for more than 95 percent
of the TWPE for its point source category. Similarly, EPA identified
facilities for review whose nitrogen and phosphorus discharges account
for the majority of nutrient discharges in DMRLoads2007'. EPA compared
2007 PCS and ICIS-NPDES data to other available information, such as
information from EPA's Envirofacts web page, the facility's NPDES
permit, and discussion with the facility contact. EPA made several facility-
level corrections, as shown in Table 3-23.
Comparability. EPA compared DMRLoads2007 to PCSLoads2004 and
PCSLoads2002 to identify pollutant discharges or wastewater flows that
differ more than the year-to-year variation of other chemicals and
facilities. EPA used this comparison to determine if quantity,
concentration, or flow corrections were needed for facility discharges with
the highest TWPE. If the comparison was unavailable (e.g., the pollutant
was not previously reported) EPA contacted the facility (see Table 3-23).
3-64
-------
Section 3 - DMRLoads2007: Development and Category Rankings
The following sections discuss EPA's quality review for the development of
DMRLoads2007:
• Section 3.4.1 describes EPA's review of mercury using PRAM 50092 (Mercury
Total Low Level); and
• Section 3.4.2 describes EPA's facility review.
3.4.1 Mercury Discharges Reported Using PRAM 50092
During the reasonableness checks of the PCS CNVRT output, EPA identified unusually
high mercury concentrations reported to PCS by facilities located in Ohio in the PCS CNVT
output. These facilities reported mercury discharges using PRAM 50092 (Mercury Total Low
Level). The PRAM 50092 concentrations in the 2004 CNVRT output ranged from 0.2 to 673
mg/L and from 0.0035 to 260,000 mg/L in the 2007 CNVRT output. EPA contacted the Ohio
Environmental Protection Agency (Ohio EPA) to determine the correct reporting units for
PRAM 50092 (Finseth, 2007c). An Ohio EPA representative explained that Ohio EPA started
requiring low level mercury analyses in 2002. At that time, some facilities had limits in
micrograms per liter (ug/L). Currently, all of the limits are in nanograms per liter (ng/L).
As a result of this contact, EPA concluded that the units for the PRAM 50092
concentrations for the 2004 PCS data should be ng/L, not mg/L. The PRAM 50092
concentrations in the 2007 CNVRT output ranged from 0.0035 to 260,000 mg/L with greater
than 99 percent of these concentrations between 0.5 and 800 mg/L. Based on this distribution,
EPA concluded that the error for the 2004 data persisted in 2007. Therefore, EPA corrected the
concentrations by dividing all concentrations for PRAM 50092 reported by facilities in Ohio in
PCSLoadCalculator2007 by one million. EPA did not make any corrections to the ICIS-NPDES
Pollutant Loading Tool because Ohio 2007 DMR data are only in PCS.
3.4.2 Facility Reviews
EPA reviewed the accuracy of facility discharges that had the greatest impact on total
category loads and category rankings. EPA used the following criteria to select facilities for
review:
• Facilities with the highest toxic-weighted discharges of individual pollutant
parameters;
• Facilities with the highest discharges of nutrients; and
• Facilities with relatively high percent of their discharges based on estimates for
missing DMR data (EST).
For the identified facilities, EPA used the following steps to review the accuracy of the
loads calculated from PCS and ICIS-NPDES data.
1. Reviewed database corrections for PCSLoads2004, PCSLoads2002, and
PCSLoads2000 to determine whether corrections were made during previous
reviews and evaluated whether these corrections should be applied to the 2007
DMR discharges.
3-65
-------
Section 3 - DMRLoads2007: Development and Category Rankings
2. Reviewed 2007 DMR data, hand calculated annual pollutant loads, and compared
results to loads calculated by PCSLoadCalculator2007and the ICIS-NPDES
Pollutant Loading tool, and stored in DMRLoads2007.
3. Reviewed PCS and ICIS-NPDES pipe description information available in PCS,
EPA's on-line Envirofacts data system, ICIS-NPDES supporting tables, or from
the facility's NPDES permit to identify monitored pollutant discharges that are:
a. Intermittent (e.g., tidal, seasonal, or occur after a storm event)
b. Internal monitoring locations from which wastewater is combined with
other waste streams and monitored again, resulting in double counting
loads, and
c. Not representative of category discharges (e.g., storm water runoff from
non-process areas, non-contact cooling water, or wastewater related to
operations in another point source category).
Table 3-23 presents EPA's review of facilities in DMRLoads2007 and the resulting corrections
made to the database.
3-66
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-23. Summary of DMRLoads2007 Facility Review
Facility
Blue Heron Paper Company
Cargill Fertilizer, Inc. - Riv
CF Industries -
Donaldsonville
Clean Harbors White Castle
LLC
Climax Mine
Doe Run Resources Co
Location
Oregon City, OR
Hillsborough
County, FL
Donaldsonville, LA
Iberville Parish, LA
Summit County, CO
Viburnum, MO
Point Source
Category
Pulp and
Paperboard
Phosphate
Manufacturing
Fertilizer
Manufacturing
CWT
Ore Mining and
Dressing
Ore Mining and
Dressing
Pollutant(s) in
Question
Methylmercury
Phosphorous
Nitrogen, Ammonia
Benzidine
Molybdenum
Lead
Review Findings
Methylmercury concentrations in
PCSLoadCalculator2007 are 1,000 times
higher than the concentrations in
Envirofacts. Envirofacts methylmercury
concentrations are in ng/L but were
entered into PCSLoadCalculator2007 as
ug/L. Facility contact verified units
should be ng/L (McCuutchen, 2009).
Facility reports DRID 1 (monthly cone.)
and A (annual quan.) with annual loads
that do not equal. DMR is counting both
DRIDs instead of just one also. Unable to
determine the correct DRID to use based
on Envirofacts.
Maximum quantities are less than
average quantities. Suspect that some
average quantities should be divided by
10. Envirofacts has the same quantities.
The permitted benzidine limit is three
orders of magnitude lower than the
concentrations in
PCSLoadCalculator2007. Facility
contact said that benzidine was ND
(Ourso, 2009).
This is a molybdenum mine. Units are
consistent with Envirofacts and permit
reporting limits. Permit/fact sheet
contains self-monitoring data that agrees
with the values reported to PCS (CO DH,
2004; CO DPHE, 2004).
This is a lead or zinc mine based on SIC
code. Units are consistent with
Envirofacts and permit reporting limits
(0.005 mg/L to 0.8 mg/L).
Action Taken/
Database Correction
Database Change: Correct
methylmercury concentrations
None
None
Database Change: Revise benzidine
concentrations to zero
None
None
3-67
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-23. Summary of DMRLoads2007 Facility Review
Facility
Dyno Nobel, Inc.
Envirosystems Incorporated
Front St. Remedial Action
GE Silicones, LLC
General Electric - Erie
Golden Eagle Refinery
Location
Carthage, MO
Hampton, NH
Kansas City, MO
Friendly, WV
Erie, PA
Martinez, CA
Point Source
Category
Explosives
Manufacturing
Independent And
Stand Alone Labs
Waste
Combustors
OCPSF
Metal Finishing
Petroleum
refining
Pollutant(s) in
Question
Nitrogen, Ammonia
Cadmium
Dioxin
Copper
Mercury
TCDD Equivalents
Review Findings
For pram 00610 (Nitrogen, ammonia
total (as N)), each outfall reports 6
months under DRID B and 6 months
under DRID C. Flows for some months
are 1 ,000 times greater than other
months.
Review of fact sheet shows that facility
incorrectly reported flows in GPD instead
of MGD for certain months (U.S. EPA
Region 1,2006).
Facility is a superfund site, and operated
in the past as both a waste combustor and
CWT. Currently treating groundwater
contaminated by organics and inorganics.
Three of four dioxin concentrations in
2007 were above the detection limit and
the MDL. Concentrations were provided
by permitting authority. Detected dioxin
in Q2 2007 and Q3 and Q4 were ND. Lab
did not analyze wastewater for dioxin for
Ql (Archterlonie, 2009).
Suspected copper concentrations units
error because the permit reporting
requirements are in ug/L instead of mg/L.
Facility confirmed the units error and
provided correct concentrations for two
quarters. Data was reported as ug/L not
mg/L (Martin, 2009a).
Facility reported 3.3 mg/L in December
2007, reported annually. Verified units in
OTIS. Facility said mercury should be
ng/L instead of mg/L (Verderese, 2009).
TCDD Equivalents measurements in
database are 1 ,000 times larger than the
concentrations in Envirofacts. The units
for concentrations in Envirofacts are in
pg/L.
Action Taken/
Database Correction
Database Change: Change DRID B and
D to C for PRAM 00610 and divide
affected flows by 1,000.
Database Change: Correct flows for the
affected monitoring periods
Database Change: Revise SIC code to
link to superfund category
Future Database Change: Revise
copper concentrations
Future Database Change: Revise
mercury concentration
Database Change: Correct TCDD
Equivalents measurements
3-68
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-23. Summary of DMRLoads2007 Facility Review
Facility
IMC - Phosphates
Company
Innovia Films
Jackson County
LAC Minerals
Morgan's Point Plant
Northshore Mining/Silver
BayP
PEPCO-Benning
Location
Donaldsonville, LA
Tecumseh, KS
Pascagoula, MS
Central City, SD
Morgan's Point, TX
Silver Bay, MN
Washington, DC
Point Source
Category
Fertilizer
Manufacturing
Plastics Molding
and Forming
Fertilizer
Manufacturing
Ore Mining
OCPSF
Ore Mining and
Dressing
Steam Electric
Power Generation
Pollutant(s) in
Question
Phosphorous
Carbon Disulfide
Phosphorous
Cyanide
Chlorine
Copper
Arochlor 1260
Review Findings
Highest phosphorous loads are from
outfall 002. Loads are approximately the
same using the quantity and the
concentration calculations. Loads also are
comparable to PCSLoads2004.
One monthly concentration appears to be
100 times higher than the other months in
2007 and 2004. Facility contact provided
corrected concentrations for April and
May that were units errors (Martin,
2009b).
Concentrations in Envirofacts match
concentrations in
PCSLoadCalculator200 7.
A review of the permit and fact sheet
indicated that the outfall SIR is an in-
stream monitoring location and therefore
should be excluded from the facility's
loads (Fuller, 2005).
The monthly average flow for March
2007 was 10,000 times higher than the
monthly maximum flow for that month
and the flows for the rest of the year.
This is a taconite mine. Units are
consistent with Envirofacts and permit
reporting limits. The calculation relies on
only one reported measurement when the
permit shows facility must monitor
monthly.
A review of OTIS data shows that all
PCBs were reported as BDL with "<" and
a concentration. The data in ICIS-NPDES
did not include the less-than signs.
Because all monthly values are BDL,
using the Hybrid Method all PCB loads
should be zero.
Action Taken/
Database Correction
None
Database Change: Correct carbon
disulfide concentrations
None
Database Change: Change MLOC to Z
(excluded from database) outfall STR
Database Change: Correct March 2007
flow
None
Database Change: Zero all PCB
(PRAM codes 39508, 39504, and
39496) loads
3-69
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-23. Summary of DMRLoads2007 Facility Review
Facility
Prasa El Yunque Filtration
Plant
Rhone-Poulenc Basic
Chemicals
Sabic Innovate Plastics
SIGECO FB Cully Station
Tampa Bay Desal
Tosco Refinery (Rodeo)
USA Holston Army Ammo
Plant Area
Location
Rio Grande, PR
Baton Rouge, LA
Ottawa, IL
Newburgh, IN
Tampa Bay, FL
Rodeo, CA
King sport, TN
Point Source
Category
Drinking Water
Treatment
Inorganic
chemicals
manufacturing
OCPSF
Steam Electric
Power Generation
Drinking Water
Treatment
Petroleum
refining
Explosives
Manufacturing
Pollutant(s) in
Question
Copper
Phosphorus, Total
(asP)
Hexachlorobenzene
Aluminum
Chloride
TCDD Equivalents
RDX, Total
Review Findings
Review of the 2007 concentration data in
OTIS indicated that February through
August, November, and December
copper concentrations were reported in
ug/L but were in the ICIS-NPDES
database as mg/L.
A review of the facility's discharges and
Envirofacts data shows the phosphorous
concentrations should be in pg/L rather
than ug/L.
Review of concentration data for OTIS
showed that the data were missing '<'
signs for every month reported for all
parameters except for copper.
For aluminum, the concentration for 10
months is 1,000 times higher than the
Form 2C data (2006) and 2006/2008 data
in OTIS. Silver, arsenic, and cadmium
concentrations are suspected units error
based on the Form 2C data. Corrected
concentrations to correspond to Form 2C
data (SIGECO, 1994).
Previous review identified a mismatch
between flows and concentrations.
NPDES permit fact sheet indicated the
flow is diluted by 70 percent from the
plant outfall to the final outfall (State of
Florida, 2001).
TCDD equivalents measurements in
database are 1 ,000 times larger than the
concentrations in Envirofacts. The units
for concentrations in Envirofacts are in
pg/L.
Facility contact said the December 2007
value was RDX, Total production instead
of effluent concentration. Contact
provided correct concentration (House,
2009).
Action Taken/
Database Correction
Database Change: Revise affected
copper concentrations by 1,000
Database Change: Revise Phosphorus,
Total (As P) concentrations
Database Change: Zero all loads except
for copper
Database Change: Revise metal
concentrations
Database Change: Divide monthly
flows by 70
Database Change: Revise TCDD
equivalents concentrations
Database Change: Revise RDX, Total
December 2007 concentration
3-70
-------
Section 3 - DMRLoads2007: Development and Category Rankings
Table 3-23. Summary of DMRLoads2007 Facility Review
Facility
Westvaco Texas, L.P.
Wise Alloys LLC
Location
Evadale, TX
Muscle Shoals, AL
Point Source
Category
Pulp, paper and
paperboard
Aluminum
Forming
Pollutant(s) in
Question
TCDD Equivalents
Nitrogen, Nitrate
Total (as N)
Review Findings
Concentrations in
PCSLoadCalculator2007 are 1,000 times
larger than the concentrations in
Envirofacts. The units for concentrations
in Envirofacts are in pg/L. Facility
contact also said all quarters were ND,
even though the fourth quarter did not
have a '<' indicator (Davis, 2009).
The facility reported two DRIDs: 1
(monthly cones.) and Q (quarterly quan.).
Unable to determine the difference
between DRIDs. Envirofacts does not
have the permit/fact sheet.
Action Taken/
Database Correction
Database Change: Revise TCDD
equivalents concentrations
Future Database Change: Add <
indicator to fourth quarter 2007 TCDD
equivalents concentration
None
3-71
-------
Section 3 - DMRLoads2007: Development and Category Rankings
3.5 DMRLoads2007 References
1. Auchterlonie, Steve. 2009. Notes from Telephone Conversation between Chris Krejci,
Eastern Research Group, Inc. and Steve Auchterlonie, Front St. Remedial Action. RE:
Verification of Magnitude and Basis of Estimate for Dioxin and Dioxin-Like Compound
Discharges in PCS. (March 13). EPA-HQ-OW-2008-517 DCN 06636.
2. Camp, Meghan. 2009. Memorandum to Carey Johnston, U.S. EPA from Meghan Camp,
Eastern Research Group, Inc. RE: Results of ICIS-NPDES Pollutant Loading Tool
Convert Module Development and DMR Data Review. (May 29). EPA-HQ-OW-2008-
0517 DCN 06357.
3. Colorado Department of Health (CO DH). 2004. NPDES Permit for Climax
Molybdenum Company, Summit County, Colorado. (September 13). EPA-HQ-OW-
2008-0517 DCN 05548.
4. Colorado Department of Public Health and Environment (CO DPHE). 2002. NPDES
Permit for Climax Molybdenum Company, Summit County, CO. (December 31). EPA-
HQ-OW-2008-0517 DCN 05972.
5. Davis, Katherine. 2009. Notes from Telephone Conversation between Elizabeth Sabol,
Eastern Research Group, Inc. and Katherine Davis, Westvaco Texas, L.P. RE: Basis of
TCDD Equivalent Concentrations Reported in 2007. (July 7). EPA-HQ-OW-2008-0517
DCN 06547.
6. Eastern Research Group, Inc. (ERG). 2009. Revised Quality Assurance Project Plan for
the 2009 Annual Screening-Level Analysis ofTRI, ICIS-NPDES, and PCS Industrial
Category Discharge Data. (September). EPA-HQ-OW-2008-0517 DCN 06558.
7. FL DEP. 2001. State of Florida Department of Environmental Protection. Save our Bays,
Air and Canals, Inc. vs Tampa Bay Desal and Department of Environmental Protection.
(October 17). EPA-HQ-OW-2008-0517 DCN 06699.
8. Fuller, Gene. South Dakota Department of Environment and Natural Resources. 2005.
NPDES Permit for LAC Minerals, Inc., Lawrence, SD. (March 22). EPA-HQ-OW-2008-
0517 DCN 06638.
9. House, Nigel. 2009. Notes from Telephone Conversation between Jessica Wolford,
Eastern Research Group, Inc. and Nigel House, USA Holston Army Ammo Plant Area.
RE: Discussion of Total RDX Discharges in PCS for 2007. (April 13). EPA-HQ-OW-
2008-0517 DCN 06696.
10. Kandle, Meghan. 2005. Memorandum to 2006 Effluent Guidelines Program Plan Docket,
EPA Docket Number OW-2004-032. RE: Point Source Category Rankings by Nitrogen
and Phosphorus Loads Calculated Using 2002 PCS Data. (August 9). EPA-HQ-OW-
2004-0032-908.
3-72
-------
Section 3 - DMRLoads2007: Development and Category Rankings
11. Martin, Jason. 2009a. Notes from Telephone Conversation between Elizabeth Sabol,
Eastern Research Group, Inc. and Jason Martin, MPM Silicones LLC. RE: Basis of
Copper (Total Recoverable) Concentrations Reported in 2007. (July 1). EPA-HQ-OW-
2008-0517 DCN 06549.
12. Martin, Tony. 2009b. Notes from Telephone Conversation between Tony Martin, Innovia
Films Inc. and Jessica Wolford, Eastern Research Group, Inc. RE: Discussion of Carbon
Disulfide Discharges for Innovia Films Inc. (March 13). EPA-HQ-OW-2008-0517. DCN
06704.
13. McCuutchen, Kate. 2009. Notes from Telephone Conversation between Elizabeth Sabol,
Eastern Research Group, Inc. and Kate McCuutchen, Blue Heron Paper Co. RE: Basis of
Methylmercury Concentration Reported in 2007 in DMR. (July). EPA-HQ-OW-2008-
0517 DCN 06546.
14. Ourso, Lisa Jo. 2009. Notes from Telephone Conversation between Chris Krejci, Eastern
Research Group, Inc. and Lisa Jo Ourso, Clean Harbors White Castle LLC. RE:
Verification of Magnitude of Benzidine Discharges Reported in PCS. EPA-HQ-OW-
2008-0517 DCN 06403.
15. Southern Indiana Gas & Electric Company (SIGECO). 1994. Form 2C NPDES for
SIGECO's F.B. Culley Generating Station. DCN 04496A34.
16. U.S. EPARegion 1. 2006. Fact Sheet for NPDES Permit NH0022055, EnviroSystems,
Inc., Hampton, NH. (April 11). EPA-HQ-OW-2008-0517 DCN 06635.
17. U.S. EPA. 1997. Guidance and Standards for Calculating Point Source Pollutant Loads
using the Permit Compliance System: Point Source Load Reductions as an Indicator of
Water Quality Improvements. Washington, DC. (August). EPA-HQ-OW-2004-0032-
0891.
18. U.S. EPA. 2001. Permit Compliance System Generalized Retrieval Training Manual.
Washington, DC. (February). EPA-HQ-OW-2003-0074 DCN 00357.
19. U. S. EPA. 2005. 2005 Annual Screening-Level Analysis: Supporting the Annual Review
of Existing Effluent Limitations Guidelines and Standards and Identification of New
Point Source Categories for Effluent Limitations Guidelines and Standards. EPA-821-B-
05-003. Washington, DC. (August). EPA-HQ-OW-2004-0032-0901.
20. U. S. EPA. 2007. Technical Support Document for the Preliminary 2008 Effluent
Guidelines Program Plan. EPA-821-R-07-007. Washington, DC. (October). EPA-HQ-
OW-2004-0032-1410.
21. Verderese, Jim. 2009. Notes from Telephone Conversation between Elizabeth Sabol,
Eastern Research Group, Inc. and Jim Verderese, General Electric Erie. RE: Basis of
Mercury Concentration Reported in December 2007 in DMR. (July). EPA-HQ-OW-
2008-0517 DCN 06548.
3-73
-------
Section 4 - Identification of Point Source Categories
4. IDENTIFICATION OF POINT SOURCE CATEGORIES
The purpose of EPA's screening-level analysis is to use existing environmental data
reported in discharge monitoring reports (DMRs) and in the Toxics Release Inventory (TRI) to
investigate discharges from industrial point source categories and prioritize these categories for
additional review. Specifically, EPA prioritizes its review of the industrial categories currently
regulated by existing effluent limitations guidelines and standards (ELGs) that cumulatively
compose 95 percent of the reported hazard (reported in units of toxic-weighted pound equivalent
or TWPE). EPA focuses its efforts on collecting and analyzing data to identify industrial
categories whose pollutant discharges potentially pose the greatest hazard to human health or the
environment because of their toxicity (i.e., highest estimates of toxic-weighted pollutant
discharges).
The term "point source category" refers to an industry as a whole based on similarity of
product produced or service provided, and is not meant to refer to specific industrial activities or
processes involved in generating the product or service. EPA therefore identifies in its biennial
Effluent Guidelines Program Plan only those new industries that it determines are properly
considered stand-alone "categories" within the meaning of the CWA - not those that are properly
considered potential new subcategories of existing categories based on similarity of product or
service. As part of existing effluent guidelines and pretreatment standard annual reviews, EPA
considers whether there are industrial activities not currently subject to effluent guidelines or
pretreatment standards that should be included with these existing categories, either as part of
existing subcategories or as potential new subcategories.
Pursuant to CWA section 304(b), which requires EPA to establish ELGs for "classes and
categories of point sources," EPA has promulgated ELGs for 56 industrial "categories." Each of
these "categories" consists of a broad array of facilities that produce a similar product or perform
a similar service - and is broken down into smaller subsets, termed "subcategories," that reflect
variations in the processes, treatment technologies, costs and other factors associated with the
production of that product that EPA is required to consider in establishing ELGs under section
304(b).
For example, the Pulp, Paper and Paperboard Point Source Category (40 CFR Part 430)
encompasses a diverse range of industrial facilities involved in the manufacture of a like product
(paper); the facilities range from mills that produce the raw material (pulp) to facilities that
manufacture end-products such as newsprint or tissue paper. EPA's classification of this
"industry by major production processes used many of the statutory factors set forth in CWA
Section 304(b), including manufacturing processes and equipment (e.g., chemical, mechanical,
and secondary fiber pulping; pulp bleaching; paper making); raw materials (e.g., wood,
secondary fiber, non-wood fiber, purchased pulp); products manufactured (e.g., unbleached pulp,
bleached pulp, finished paper products); and, to a large extent, untreated and treated wastewater
characteristics (e.g., BOD loadings, presence of toxic chlorinated compounds from pulp
bleaching) and process water usage and discharge rates."18
Each subcategory reflects differences in the pollutant discharges and treatment
technologies associated with each process. Similarly, the Iron and Steel Manufacturing Point
18 Supplemental Technical Development Document for Effluent Limitations Guidelines and Standards for the Pulp,
Paper, and Paperboard Category, Page 5-3, EPA-821-R-97-011, October 1997 (U.S. EPA, 1997).
4-1
-------
Section 4 - Identification of Point Source Categories
Source Category (40 CFR Part 420) consists of various subcategories that reflect the diverse
range of processes involved in the manufacture of iron and steel, ranging from facilities that
make the basic fuel used in the smelting of iron ore (Subpart A - Cokemaking) to those that cast
the molten steel into molds to form steel products (Subpart F - Continuous Casting). An example
of an industry category based on similarity of service provided is the Transportation Equipment
Cleaning Point Source Category (40 CFR Part 442), which is subcategorized based on the type of
tank (e.g., rail cars, trucks, barges) or cargo transported by the tanks cleaned by these facilities,
reflecting variations in wastewaters and treatment technologies associated with each.
Finally, Section 304(m)(l)(B) of the CWA directs EPA to use the biennial Effluent
Guidelines Program Plans to identify categories of sources discharging non-trivial amounts of
toxic or non-conventional pollutants for which EPA has not published ELGs under section
304(b)(2) or new source performance standards (NSPS) under section 306. EPA uses DMR and
TRI data to assist in the identification of any new point source categories that meet the criteria of
Section 304(m)(l)(B). EPA also uses TRI data to help identify indirect dischargers without
categorical pretreatment standards to identify potential new categories for pretreatment standards
under CWA Sections 304(g) and 307(b). EPA assesses whether industrial operations not
currently regulated by existing effluent guidelines or pretreatment standards should be addressed
as a potential new subcategory under an existing point source category rather than as a new
industrial category.
EPA uses Standard Industrial Classification (SIC) and North American Industrial
Classification System (NAICS) codes to relate discharge data in DMR and TRI to the point
source categories. DMR data are contained in EPA's Permit Compliance System (PCS) and the
Integrated Compliance Information System for the National Pollutant Discharge Elimination
System (ICIS-NPDES). As part of the 2009 annual review, EPA created DMRLoads2007 to
combine DMR data from PCS and ICIS-NPDES (see Section 3). Facilities with data in PCS
and/or ICIS-NPDES are identified by a four-digit SIC code, while facilities with data in TRI are
identified by a six-digit NAICS code. To use the DMR and TRI data to estimate the pollutants
discharged by each industrial point source category, EPA linked each four-digit SIC code and
six-digit NAICS code to an appropriate point source category. EPA's linkages are summarized in
the "SIC/Point Source Category Crosswalk" and "NAICS/Point Source Category Crosswalk"
tables (Tables C-l and C-2 in Appendix C, respectively). These crosswalks are key elements of
both the DMRLoads2007 and TRIReleases2007 databases.
The remainder of this section presents the following information:
• Section 4.1 - Background on NAICS and SIC Codes;
• Section 4.2 - SIC Code to Point Source Category Crosswalk;
• Section 4.3 - NAICS Code to Point Source Category Crosswalk;
• Section 4.4 - Potential New Point Source Categories; and
• Section 4.5 - Crosswalk References.
4.1 Background on NAICS and SIC Codes
Starting in 2006, facilities reporting to EPA's TRI were required to provide the NAICS
code(s) that describe their actions. The NAICS system is the current statistical classification
standard underlying all establishment-based federal economic statistics classified by industry
(U.S. Census Bureau, 2009). The Office of Management and Budget (OMB) first developed the
-------
Section 4 - Identification of Point Source Categories
NAICS system in 1997 to replace SIC codes and streamline economic statistics throughout North
America. The first set of NAICS codes were updated in 2002 (referred to as the 2002 NAICS
system), and the second set were updated in 2007 (referred to as the 2007 NAICS system).
Although it was developed by OMB, the NAICS system is used by other government
agencies, including EPA, to promote data comparability. In the NAICS system, each
establishment is classified according to its primary economic activity, which is determined by its
principal product or group of products. An establishment may have activities in more than one
NAICS code. Some data collection organizations (e.g., the U.S. Economic Census) assign one
NAICS code per establishment.
On June 6, 2006, EPA published a final rule requiring facilities to use 2002 NAICS
codes, instead of SIC codes, for reporting to the 2006 and 2007 TRI (see 71 FRN 32464). EPA is
requiring facilities to use 2007 NAICS codes for reporting to the 2008 TRI and future years (see
73 FRN 32466, June 9, 2008). TRI allows facilities to identify their primary NAICS code and up
to five additional NAICS codes. These codes reflect the principal activity causing environmental
releases at a facility and other activities, respectively.
Facilities with data in PCS and ICIS-NPDES are classified by SIC code. EPA has not
announced plans to change its PCS and ICIS-NPDES databases to NAICS codes. As with the
NAICS system, an establishment may have activities in more than one SIC code (OMB, 1987).
PCS allows facilities to report one SIC code, while ICIS-NPDES allows facilities to report a
primary SIC code and up to two additional SIC codes. The primary SIC code reflects the
principal activity causing the discharge at each facility and the additional SIC codes represent
other activities at the facility.
As part of the 2009 annual review, EPA reviewed its existing SIC/Point Source Category
Crosswalk to determine if revisions were necessary because facilities reported new SIC codes or
additional information about their discharges. Because the TRI data for 2007 is classified by
NAICS code, EPA created a NAICS/Point Source Category Crosswalk for TRIReleases2007. For
a given facility, the operations covered by the SIC code in PCS and/or ICIS-NPDES may differ
from operations covered by the primary NAICS code identified in TRI.
4.2 SIC Code to Point Source Category Crosswalk
EPA first developed the SIC code to point source category crosswalk (SIC/Point Source
Category Crosswalk) as part of the 2003 and 2004 screening-level analyses (U.S. EPA, 2005a).
Since then, EPA has continued to update this crosswalk. Specifically for the 2009 screening-
level review, EPA updated this crosswalk for use with DMRLoads2007'. For the 2009 annual
review, as well as previous reviews, EPA divided the SIC codes into four groups defined as
follows:
• Existing Point Source Category. Discharges from most facilities in the SIC code
meet the applicability requirements of an existing point source category.
• Potential New Subcategory of an Existing Point Source Category. Discharges
from most facilities in the SIC code may be considered part of a potential new
subcategory of an industrial category subject to an existing ELG. EPA based this
4-3
-------
Section 4 - Identification of Point Source Categories
determination on the similarity of processes and operations at facilities in the SIC
code to those at facilities in the existing category.
• Potential New Point Source Category. Discharges from facilities in the SIC
code are similar to each other but do not meet the applicability requirements of
and are not similar to a point source category subject to an existing ELG.
• Category Not Identifiable. Facilities in the SIC code engage in a variety of
industrial operations and likely meet the applicability requirements of several
existing point source categories. However, EPA is not able to identify a coherent
stand-alone point source category based on the SIC code description.
Most SIC codes reported by facilities with DMR discharge information meet the
applicability of an existing point source category and fall into the first group. The following
sections describe the development and review of the SIC/Point Source Category Crosswalk.
4.2.1 SIC Codes Related to Existing Point Source Categories
As part of its 2003 and 2004 screening-level analyses, EPA related SIC codes to existing
point source categories. During the development of the existing ELGs for these categories, EPA
studied demographic and economic data, including SIC code data, for the facilities to which the
ELGs apply. EPA developed the relationship, or "crosswalk," between SIC codes and point
source categories by consulting, as necessary, the documentation for the development on the
existing ELGs. This crosswalk is included as Table C-l in Appendix C.
Because most point source categories are not defined by SIC code, the relationship
between SIC code and point source category is not a one-to-one correlation. A single SIC code
may include facilities in more than one point source category, so associating an SIC code with
only one category may be an over simplification. Also, many facilities have operations subject to
more than one point source category. Further, facilities in some categories cannot be identified
by SIC code. The following subsections discuss how EPA reconciled these inconsistencies to
cross-reference appropriate point source categories to specific SIC codes. EPA reviewed each of
these inconsistencies as part of the 2009 annual review and further refined the SIC/Point Source
Category Crosswalk.
4.2.1.1 SIC Codes Counted in More than One Point Source Category
A single SIC code may include facilities subject to more than one point source category.
For example, SIC code 3357, Drawing and Insulating of Nonferrous Wire, includes facilities that
draw wire made from aluminum, copper, and other nonferrous metals such as nickel and silver.
Depending on the type of metal, ELGs from three categories may apply to the discharges from
these operations. EPA included the loads discharged by facilities in SIC code 3357 in each of the
three applicable categories: Aluminum Forming, Copper Forming, and Nonferrous Metals
Forming. In order to make a "worst case" estimate of the TWPE discharged by every category,
EPA included the loads from SIC codes associated with multiple point source categories in the
load for each associated category, double- or triple-counting the loads from these SIC codes.
Table 4-1 presents the SIC codes associated with multiple point source categories, and identifies
the applicable point source categories.
-------
Section 4 - Identification of Point Source Categories
Table 4-1. SIC Codes Counted in Multiple Point Source Categories
SIC
Code
3357
3363
3482
3483
3463
4953
7221
7335
7336
7384
SIC Description
Drawing and Insulating of
Nonferrous Wire
Aluminum Die Casting
Small Arms Ammunition
Ammunition, Except for Small Arms
Nonferrous Forgings
Refuse Systems
Photographic Studios, Portrait
Commercial Photography
Commercial Art and Graphic Design
Photofinishing Laboratories
Applicable Point Source Categories
Aluminum Forming (40 CFR 467),
Copper Forming (40 CFR 468), and
Nonferrous Metals Forming & Metal Powders (40 CFR 471)
Aluminum Forming (40 CFR 467) and
Nonferrous Metals Forming & Metal Powders (40 CFR 471)
Metal Finishing (40 CFR 433) and
Nonferrous Metals Forming & Metal Powders (40 CFR 471)
Metal Finishing (40 CFR 433) and
Nonferrous Metals Forming & Metal Powders (40 CFR 471)
Aluminum Forming (40 CFR 467),
Copper Forming (40 CFR 468), and
Nonferrous Metals Forming & Metal Powders (40 CFR 471)
Landfills (40 CFR 445) and
Waste Combustors (40 CFR 444)
Photographic (40 CFR 459) and
Photoprocessing (2005 Annual Review Potential New
Subcategory) a
Photographic (40 CFR 459) and
Photoprocessing (2005 Annual Review Potential New
Subcategory) a
Photographic (40 CFR 459) and
Photoprocessing (2005 Annual Review Potential New
Subcategory) a
Photographic (40 CFR 459) and
Photoprocessing (2005 Annual Review Potential New
Subcategory) a
a As part of the Final 2006 Plan, EPA determined that categorical pretreatment standards were not warranted for the
Photoprocessing industry (U.S. EPA, 2006b).
4.2.1.2 SIC Codes Divided Among Point Source Categories
As noted previously, some facilities are subject to regulations from more than one point
source category. EPA was able to assign discharges from some of these SIC codes to the
appropriate category and avoid double counting. EPA made some of these assignments at the
facility level and some at the pollutant level, as discussed below.
Facility-Level Point Source Category Assignment
For some SIC codes with facilities subject to more than one point source category, EPA
was able to assign each facility to a category. EPA reviewed information available about each
facility to determine which point source category applied to the facility's operations. As part of
the 2005 and 2006 annual reviews, EPA contacted facilities to understand which facility
operations were the source of reported wastewater discharges if publically available information
did not indicate the appropriate category. For example, for the 2005 annual review, EPA located
information about facilities in SIC codes associated with both the Porcelain Enameling and
Metal Finishing Categories. EPA used this information to determine the category most likely to
-------
Section 4 - Identification of Point Source Categories
apply to each facility's discharge (Wolford, 2005). As part of the 2009 screening-level review,
EPA updated the SIC/Point Source Category Crosswalk based on this review (see Section
4.2.1.2). Facilities reporting these SIC codes for the first time in 2007 (e.g., previously reported a
different SIC code or are new facilities) were reviewed to link the facility's discharges to the
appropriate point source category as part of the 2009 annual review. Table 4-2 presents the SIC
codes that EPA assigned to point source categories at the facility level. In future databases, as
new facilities report SIC codes that do not link directly to a point source category (e.g., SIC code
2048 does not link to a point source category), EPA will review facility's operations and identify
the appropriate point source category. EPA will also review operations of new facilities with
significant TWPE in each of these SIC codes to determine if they are assigned to the appropriate
point source category.
As part of the 2009 screening-level review, EPA reviewed available information about
pollutant loads and facility information for facilities reporting SIC code 4953, Refuse Systems,
to determine if the facility's discharges were primarily associated with operations regulated by
the Landfill Category (40 CFR Part 444) or by the Waste Combustor Category (40 CFR Part
445). EPA incorporated these changes into the SIC/Point Source Category Crosswalk in the 2007
DMR database based on this review (see Section 4.2.1.2). In future databases, as new facilities
report SIC code 4953, EPA will individually review their operations to determine the category
that most likely applies to the facility's discharges.
Table 4-2. Facility-Level Point Source Category Assignment SIC Codes
SIC
2048: Prepared Feed and
Feed Ingredients for
Animals and Fowl, Except
Dogs and Cats
2819: Industrial Inorganic
Chemicals, NEC
2874: Phosphatic
Fertilizers
3341: Secondary Smelting
and Refining of Nonferrous
Metals
3431: Metal Sanitary Ware
3469: Metal Stampings,
NEC
3471: Plating and Polishing
Primary Associated Point
Source Category
None. In future databases as new
facilities report this SIC code,
EPA will review facility's
operations and identify the
appropriate point source category.
Inorganic Chemicals
Manufacturing (40 CFR Part 415)
Phosphate Manufacturing (40
CFR Part 422)
Nonferrous Metals Manufacturing
(40 CFR Part 421)
Metal Finishing (40 CFR Part
433)
Metal Finishing (40 CFR Part
433)
Metal Finishing (40 CFR Part
433)
Other Associated Point Source
Categories
Grain Mills Manufacturing (40
CFR Part 406)
Meat and Poultry Products (40
CFR Part 432)
Pharmaceutical Manufacturing
(40 CFR Part 43 9)
Nonferrous Metals Manufacturing
(40 CFR Part 471)
Phosphate Manufacturing (40
CFR Part 422)
Fertilizer Manufacturing (40 CFR
Part 418)
Coil Coating (40 CFR Part 465)
Porcelain Enameling (40 CFR
Part 467)
Porcelain Enameling (40 CFR
Part 467)
Coil Coating (40 CFR Part 465)
Expanded
SIC Code
(Assigned at
Facility
Level)
2048GRAIN
2048MPP
2048PH
2819NMM
2819PHOS
2874FER
3341CC
343 1PE
3469PE
3471CC
4-6
-------
Section 4 - Identification of Point Source Categories
Table 4-2. Facility-Level Point Source Category Assignment SIC Codes
SIC
3624: Carbon and Graphite
Products
3633: Household Laundry
Equipment
3639: Household
Appliances, NEC
4953: Refuse Systems
Primary Associated Point
Source Category
Metal Finishing (40 CFR Part
433)
Metal Finishing (40 CFR Part
433)
Metal Finishing (40 CFR Part
433)
Landfills (40 CFR 445) and
Waste Combustors (40 CFR 444)
a
Other Associated Point Source
Categories
Carbon Black Manufacturing (40
CFR Part 458)
Porcelain Enameling (40 CFR
Part 467)
Porcelain Enameling (40 CFR
Part 467)
Landfills (40 CFR Part 445)
Waste Combustors (40 CFR Part
444)
Expanded
SIC Code
(Assigned at
Facility
Level)
3624CB
3633PE
3639PE
4953L
4953WC
NEC - Not elsewhere classified.
a In future databases as new facilities report this SIC code, EPA will review facility's operations and identify the
appropriate point source category.
EPA is currently considering revisions to a subset of the ELGs for Organic Chemicals,
Pesticides, and Synthetic Fibers (OCPSF) (40 CFR 414) and the Inorganic Chemicals
Manufacturing (40 CFR 415) for facilities that produce chlorine and chlorinated hydrocarbons
(CCH). Because the CCH rulemaking is underway, for the 2009 annual review and previous
reviews, EPA assigned the SIC code "VCCA" to the CCH facilities in the SIC/Point Source
Category Crosswalk to separately identify these facilities (note VCCA, Vinyl Chloride and
Chloralkali, is the previous name for the CCH rulemaking). The list of CCH facilities in
DMRLoads2007 is included in Table C-3 in Appendix C.
As part of the Pulp, Paper, and Paperboard Category (Pulp and Paper Category) (40 CFR
Part 430) Detailed Study (U.S. EPA, 2006a), EPA reviewed the operations of facilities reporting
SIC codes 2611: Pulp Mills, 2621: Paper Mills, and 2631: Paperboard Mills to determine the
applicable subpart for each facility. A 1988 legal suit obligated EPA to address discharges of
poly chlorinated dibenzo-(p)-dioxins and poly chlorinated dibenzofurans from 104 bleaching pulp
mills, including nine dissolving pulp mills. During its response to the 1988 legal suit, EPA
decided to review and revise the Pulp and Paper Category regulations in three phases. EPA
addressed Phase I first, chose not to revise the ELGs for Phase II, and chose to support NPDES
permit writers individually in developing permit-specific effluent limitations to control
discharges of these chemicals from the remaining operating mills in Phase III.
Because the Pulp and Paper Category regulations regulate facilities by process used and
product produced they do not correspond to SIC codes. Therefore, EPA added "-1" to the SIC
codes of facilities that met the applicability of Phase I:
• Subpart B (Bleached Papergrade Kraft and Soda); and
• Subpart E (Papergrade Sulfite).
EPA added "-2" to the SIC codes of facilities that met the applicability of Phase II:
4-7
-------
Section 4 - Identification of Point Source Categories
• Subpart C (Unbleached Kraft);
• Subpart F (Semi-Chemical);
• Subpart G (Groundwood, Chemic-Mechanical, and Chemi-Thermo-Mechanical);
• Subpart H (Non-Wood Chemical Pulp);
• Subpart I (Secondary Fiber Deink);
• Subpart J (Secondary Fiber Non-Deink);
• Subpart K (Fine and Lightweight Papers from Purchased Pulp); and
• Subpart L (Tissue, Filter, Non-Woven and Paperboard from Purchased Pulp).
EPA added "-3" to the SIC codes of facilities that met the applicability of Phase III:
• Subpart A (Dissolving Kraft); and
• Subpart D (Dissolving Sulfite).
As part of the 2009 annual review, EPA continued incorporating the updated SIC codes
identified during previous annual reviews. EPA did not review operations for new facilities
reporting the SIC codes 2611, 2621, and 2631 to assign the regulatory phase to the facilities.
Outfall-Level Point Source Category Assignment
EPA was able to divide the pollutant discharges for selected facilities that discharge
wastewater subject to more than one point source category by outfall. As part of the 2007 annual
review, EPA reviewed discharges, permits, and permit fact sheets for facilities with high TWPE.
EPA determined that one , Radford Army Ammunition Plant in Montgomery County, VA, had
selected outfall that were regulated under OCPSF Category (40 CFR Part 414) while other
outfall were regulated but the Explosives Manufacturing Category (40 CFR Part 457). EPA
assigned the outfalls associated with OCPSF manufacturing to the OCPSF category by
appending "OCPSF" to the facility's outfall-level SIC codes. EPA continued this assignment as
part of the 2009 annual review.
Pollutant-Level Point Source Category Assignment
For most facilities that discharge wastewater subject to more than one point source
category, EPA was not able to divide the pollutant discharges between applicable point source
categories. The following subsections discuss two exceptions where EPA was able to assign
wastewater discharges of certain chemicals to the appropriate point source category.
Organic Chemicals, Plastics, and Synthetic Fibers/Pesticides
The OCPSF ELGs (40 CFR Part 414) may apply to discharges from facilities in the
following SIC codes:
• 2821: Plastics Materials, Synthetic Resins, and Nonvulcanizable Elastomers;
• 2823: Cellulosic Manmade Fibers;
• 2824: Manmade Organic Fibers, Except Cellulosic;
• 2865: Cyclic Organic Crudes and Intermediates, and Organic Dyes and Pigments;
and
• 2869: Industrial Organic Chemicals, Not Elsewhere Classified.
-------
Section 4 - Identification of Point Source Categories
In addition, EPA is considering including operations from the following five SIC codes as
potential new subcategories of the OCPSF Category:
• 2842: Specialty Cleaning, Polishing, and Sanitation Preparations;
• 2844: Perfumes, Cosmetics, and Other Toilet Preparations;
• 2891: Adhesives and Sealants;
• 2899: Chemicals and Chemical Preparations, Not Elsewhere Classified; and
• 5169: Chemicals and Allied Products, Not Elsewhere Classified.
Some facilities in the regulated SIC codes and SIC codes of the potential new subcategory
manufacture and/or formulate pesticides as well as other organic chemicals. Regulations for the
Pesticide Chemicals Category (40 CFRPart 455) control discharges from pesticide operations.
For the 2009 screening-level analysis of discharges from existing categories, and previous
reviews, EPA subtracted all pesticide discharges from OCPSF and counted them as discharges
from the Pesticides Chemicals Category, by appending a "P" to the facility's pollutant-level SIC
code (e.g., EPA revised pesticide discharges from SIC code 2869 to SIC code 2869P).
EPA used a table containing a list of pesticides and their CAS numbers to identify the
pesticide releases from the OCPSF Category for both the DMR and TRI databases. In developing
the list of pesticides, EPA started with the list of 272 pesticide active ingredients that was created
during the most recent pesticides rulemaking. Some of the pesticides in the list of 272 active
ingredients were multiple compounds, for example "2,4 D salts and esters" and "organo-tin
pesticides," and were not identified by CAS number. EPA identified individual chemicals and
CAS numbers for active ingredients in these groups and added them to the pesticides list. All of
the chemicals identified from the list of 272 pesticide active ingredients were included in the
pesticides list, except for biphenyl and dichlorobenzene. Biphenyl and dichlorobenzene were not
included because EPA determined that OCPSF facilities use these chemicals for specific
manufacturing uses not related to pesticides.
EPA identified additional pesticide active ingredients by using the 1988 FIFRA and
TSCA Enforcement System (FATES) Database and a list created in 2003 by the Office of
Pesticide Programs (OPP). EPA combined the two lists and determined which of the pesticide
active ingredients facilities reported discharging to the DMR databases in 2007. For reported
discharges, EPA determined whether the pesticide active ingredient had significant non-pesticide
related manufacturing uses. EPA did not add chemicals, such as acrolein, trichlorofuoromethane,
silver, and sulfuric acid, whose primary use was non-pesticide-related, to the list, while EPA
added chemicals whose primary purpose was pesticide-related to the list. The list of chemicals
reported in the DMR and TRI databases that EPA considered pesticides for the purpose of its
screening-level analysis of discharges from existing categories contains 415 chemicals.
MP&M/Metal Finishing
Regulations for the Metal Finishing Category (40 CFR Part 433) may apply to discharges
from facilities in 179 SIC codes for which discharges were reported in DMR in 2007.
Regulations for the Metal Products and Machinery (MP&M) Category (40 CFR Part 438) may
apply to some of the pollutants directly discharged by facilities in 136 of these SIC codes. The
final MP&M rule at 40 CFR Part 438. l(b) specifically excludes both metal-bearing wastewaters
and wastewaters subject to other effluent guidelines (e.g., Metal Finishing). For its screening-
level analysis of discharges from existing categories, EPA developed methodologies to apportion
-------
Section 4 - Identification of Point Source Categories
pollutant loads between the MP&M and Metal Finishing Categories. EPA applied this
methodology to the 2009 screening-level analysis and previous reviews.
The MP&M rule as promulgated regulates oil and grease (O&G) and total suspended
solids (TSS) in direct discharges from certain facilities that generate oily wastewater; it does not
specifically regulate any other chemicals. EPA used the list of organic "pollutants of concern" it
had developed for the MP&M rule and identified 103 pollutants in the DMR databases, including
O&G and TSS. For the 2009 screening-level analysis, EPA counted all discharges of these
pollutants from the 136 MP&M SIC codes in DMRLoads2007 as MP&M discharges. EPA
counted discharges of all other chemicals from these facilities in the Metal Finishing Category in
DMRLoads2007'. EPA believes that the identified pollutants are those that are most likely
associated with the non-metal bearing oily waste streams subject to the MP&M regulations, and
that this apportionment, which avoids double counting pollutant loads, is a reasonable approach
for screening-level analysis of discharges from existing categories.
Table C-4 in Appendix C lists the 88 organic "pollutants of concern" for the MP&M rule.
For the 2009 Annual Review, as for previous reviews, EPA matched DMR pollutants to the list
of 88 MP&M chemicals using CAS numbers and the SUPERCAS table (described in Section 3).
Using the SUPERCAS table, EPA matched 104 pollutant parameters to the list of 88 organic
"pollutants of concern" for the MP&M rule that are discharged by facilities in the 136 MP&M
SIC codes. EPA identified these 104 pollutant parameters as "Controlled by MP&M." Table C-5
in Appendix C presents the list of DMR parameters allocated to MP&M for the 2009 Annual
Review.
4.2.1.3 Categories Not Identified By SIC Code (Centralized Waste Treaters)
The Centralized Waste Treaters (CWT) Category (40 CFR Part 437) is not linked to
specific SIC codes; therefore, the SIC/Point Source Category Crosswalk does not assign any SIC
codes to the CWT Category. As part of the 2008 annual review, EPA reviewed the list of CWTs
developed as part of the CWT rulemaking and assigned these facilities the SIC code of "CWT"
and linked it to Part 437 in the SIC/Point Source Category Crosswalk. EPA also reviewed the
facilities reporting SIC code 4953, Refuse Systems, and assigned CWT facilities reporting this
SIC code the SIC code of "CWT" that links to Part 437. As part of the 2009 annual review, EPA
reviewed the operations of all facilities reporting SIC code 4953 that were not previously
assigned the "CWT" SIC code to determine if their operations were applicable to the CWT
Category, Landfills Category (40 CFR Part 444), or Waste Combustors Category (40 CFR Part
445).
4.3 NAICS Code to Point Source Category Crosswalk
The 2007 TRI data was the first reporting year that facilities were required to report
NAICS codes rather than SIC codes. Therefore, as part of the 2009 screening-level analyses,
EPA developed the NAICS code to point source category crosswalk (NAICS/Point Source
Category Crosswalk) to link NAICS codes to appropriate point source categories for use with
TRI data. EPA divided the NAICS codes into four groups, the same four groups as EPA used to
develop the SIC/Point Source Category Crosswalk:
4-10
-------
Section 4 - Identification of Point Source Categories
• Existing Point Source Category. Discharges from most facilities in the NAICS
code meet the applicability requirements of an existing point source category.
• Potential New Subcategory of an Existing Point Source Category. Discharges
from most facilities in the NAICS code may be considered part of a potential new
subcategory of an industrial category subject to an existing ELG. EPA based this
determination on the similarity of processes and operations at facilities in the
NAICS code of concern to those at facilities in the existing category.
• Potential New Point Source Category. Discharges from facilities in the NAICS
code are similar to each other but do not meet the applicability requirements of
and are not similar to a point source category subject to an existing ELG.
• Category Not Identifiable. Facilities in the NAICS code engage in a variety of
industrial operations and likely meet the applicability requirements of several
existing point source categories. However, EPA is not able to identify a coherent
stand-alone point source category based on the NAICS code description.
Most NAICS codes reported by facilities in TRI meet the applicability of an existing point
source category and fall into the first group.
4.3.1 NAICS Codes Related to Existing Point Source Categories
As part of its 2009 screening-level analysis, EPA related NAICS codes to existing point
source categories. EPA developed this crosswalk by using TRI facilities' point source category
assignments from previous years of review. For example, for the 2005 annual review SIC code
2023 is linked to the Dairy Products Processing Category (40 CFR Part 405), shown in Table
4-3. In 2007 TRI, Dietrich's Milk Products reported its pollutant discharges were from activities
in the NAICS code 311514. EPA thus assigned NAICS 311514 to the Dairy Products Processing
Category in the NAICS/Point Source Category Crosswalk. Using this method, EPA assigned the
point source categories to all but 46 of the NAICS codes reported in TRIReleases2007'. These 46
NAICS codes were reported by facilities that did not report to TRI in 2005. For these, EPA
assigned these NAICS codes to the appropriate point source category based on NAICS
descriptions and point source category applicability. The resulting NAICS/Point Source
Category Crosswalk is included as Table C-2 in Appendix C.
Table 4-3. Example NAICS/Point Source Category Crosswalk Development
Facility Name
Dietrich's Milk Products
2005 SIC Code
2023 : Condensed and
Evaporated Milk
Point Source Category
Dairy Products Processing
(40 CFR Part 405)
2007 NAICS Code
3 115 14: Dry, Condensed,
and Evaporated Dairy
Product Manufacturing
Because most point source categories are not defined by NAICS code, the relationship
between NAICS code and point source category is not a one-to-one correlation. This is also the
case for the SIC codes (see Section 4.2.1.1). A single NAICS code may include facilities in more
than one point source category, and associating a NAICS code with only one category may be an
over simplification. Also, many facilities have operations subject to more than one point source
-------
Section 4 - Identification of Point Source Categories
category. Further, some categories cannot be identified by NAICS code. The following
subsections discuss how EPA reconciled these inconsistencies to cross-reference appropriate
point source categories to specific NAICS codes. As part of previous annual reviews, EPA
remedied some of these issues in the TRI databases. EPA carried these fixes over to the 2007
TRI database as part of the 2009 annual review.
4.3.1.1 NAICS Codes Counted in More than One Point Source Category
A single NAICS code may include facilities subject to more than one point source
category. For example, NAICS code 562211, Hazardous Waste Treatment and Disposal,
includes facilities that operate treatment and/or disposal facilities for hazardous waste.
Depending on the type of treatment and/or disposal, ELGs from three different categories may
apply to the discharges from these operations. EPA included the loads discharged by facilities in
NAICS code 562211 in each of the three applicable categories: Centralized Waste Treatment,
Waste Combustors, and Landfills. In order to make a "worst case" estimate of the TWPE
discharged by every category, EPA included the loads from NAICS codes associated with
multiple point source categories in the load for each category, double- or triple-counting the
loads from these NAICS codes. Table 4-4 presents the NAICS codes associated with multiple
point source categories, and identifies the applicable point source categories.
Table 4-4. NAICS Codes Counted in Multiple Point Source Categories
NAICS
Code
331521
332112
332992
332993
332999
335921
335929
562211
NAICS Description
Aluminum Die-Casting Foundries
Nonferrous Forging
Small Arms Ammunition
Manufacturing
Ammunition (except Small Arms)
Manufacturing
All Other Miscellaneous
Fabricated Metal Product
Manufacturing
Fiber Optic Cable Manufacturing
Other Communication and Energy
Wire Manufacturing
Hazardous Waste Treatment and
Disposal
Applicable Point Source Categories
Nonferrous Metals Manufacturing (40 CFR Part 421) and
Aluminum Forming (40 CFR Part 467)a
Aluminum Forming (40 CFR Part 467),
Copper Forming (40 CFR Part 468), and
Nonferrous Metals Forming and Metal Powders (40 CFR Part 471)b
Metal Finishing (40 CFR Part 433) and
Nonferrous Metals Forming and Metal Powders (40 CFR Part 471)
Metal Finishing (40 CFR Part 433) and
Nonferrous Metals Forming and Metal Powders (40 CFR Part 471)
Metal Finishing (40 CFR Part 433),
Aluminum Forming (40 CFR Part 467),
Copper Forming (40 CFR Part 468), and
Nonferrous Metals Forming and Metal Powders (40 CFR Part 471)°
Glass Manufacturing (40 CFR Part 426) and
Plastics Molding and Forming (40 CFR Part 463)
Aluminum Forming (40 CFR Part 467),
Copper Forming (40 CFR Part 468), and
Nonferrous Metals Forming and Metal Powders (40 CFR Part 471)
Centralized Waste Treatment (40 CFR Part 437),
Waste Combustors (40 CFR Part 444), and
Landfills (40 CFR Part 445)d
4-12
-------
Section 4 - Identification of Point Source Categories
Table 4-4. NAICS Codes Counted in Multiple Point Source Categories
NAICS
Code
562219
NAICS Description
Other Nonhazardous Waste
Treatment and Disposal
Applicable Point Source Categories
Centralized Waste Treatment (40 CFR Part 437),
Waste Combustors (40 CFR Part 444), and
Landfills (40 CFR Part 445)d
aEPA reviewed publicly available information for these facilities and determined that some reporting this NAICS
code have operations applicable to the Metal Molding and Casting Category (40 CFR Part 464). EPA assigned these
facilities the NAICS code 331521MMC to link to the Metal Molding and Casting Category in the NAICS/Point
Source Category Crosswalk (see Section 4.3.1.2).
b EPA reviewed publicly available information for these facilities and determined that some reporting this NAICS
code have operations applicable to the Iron and Steel Manufacturing Category (40 CFR Part 420) and the Metal
Finishing Category (40 CFR Part 433). EPA assigned the NAICS code 332112IRON to facilities generating
wastewater to which the Iron and Steel ELGs apply and the NAICS code 332112MF to facilities generating
wastewater to which the Metal Finishing ELGs apply (see Section 4.3.1.2).
0 EPA reviewed publicly available information for these facilities and determined that some reporting this NAICS
code have operations that are applicable to the Metal Finishing Category (40 CFR Part 433) and the Nonferrous
Metals Forming and Metal Powders Category (40 CFR Part 471). EPA assigned the NAICS code 332999DC to
these facilities (see Section 4.3.1.2). EPA determined that some facilities reporting this NAICS code have
operations that are applicable to the Aluminum Forming Category (40 CFR Part 467), Copper Forming Category (40
CFR Part 468), and Nonferrous Metals Forming and Metal Powders Category (40 CFR Part 471). EPA assigned the
NAICS code 332999TC to these facilities (see Section 4.3.1.2).
d EPA reviewed publicly available information for these facilities and identified facilities with operations applicable
to the Centralized Waste Treatment Category (40 CFR Part 437) with the NAICS code CWT. EPA assigned the
NAICS code WC to facilities with operations applicable to the Waste Combustor Category (40 CFR Part 444). EPA
assigned the NAICS code LNDFLL to facilities with operations applicable to the Landfills Category (40 CFR Part
445). Facilities with multiple operations were counted in all the applicable categories.
4.3.1.2 NAICS Codes Divided Among Point Source Categories
As noted previously, some facilities are subject to regulations from more than one point
source category. EPA was able to assign some of these discharges to the appropriate category
and avoid double counting by carrying over changes made during previous annual reviews. EPA
made some of these assignments at the facility level, the pollutant level, and the discharge level,
as discussed below.
Facility-Level Point Source Category Assignment
For NAICS codes that include facilities subject to more than one point source category,
EPA reviewed available information about pollutant loads and manufacturing operations to
assign each facility to the category that applied to its discharges. Table 4-5 presents the NAICS
codes that EPA assigned to point source categories at the facility level. In future databases, EPA
will review facilities with significant TWPE in each of these NAICS codes to determine if they
are assigned to the appropriate point source category.
4-13
-------
Section 4 - Identification of Point Source Categories
Table 4-5. Facility-Level Point Source Category Assignment NAICS Codes
NAICS
311119: Other Animal Food
Manufacturing
31 1225: Fats and Oils Refining
and Blending
31 1999: All Other Miscellaneous
Food Manufacturing
315992: Glove and Mitten
Manufacturing
324199: All Other Petroleum and
Coal Products Manufacturing
325120: Industrial Gas
Manufacturing
325188: All Other Basic Inorganic
Chemical Manufacturing
325510: Paint and Coating
Manufacturing
Primary Associated Point
Source Category
Food and Kindred Products
Potential New Point Source
Category a
Miscellaneous Foods and
Beverages Potential New Point
Source Category a
Miscellaneous Foods and
Beverages Potential New Point
Source Category a
Textile Mills (40 CFR Part
410)
Petroleum Refining (40 CFR
Part 4 19)
Inorganic Chemicals
Manufacturing (40 CFR Part
415)
Inorganic Chemicals
Manufacturing (40 CFR Part
415)
Paint Formulating (40 CFR Part
446)
Other Associated Point Source
Categories
Grain Mills (40 CFR Part 406)
Meat and Poultry Products (40 CFR Part
432)
Pharmaceutical Manufacturing (40 CFR
Part 439)
Fertilizer Manufacturing Category (40
CFR Part 4 18)
Dairy Products Processing (40 CFR Part
405)
Grain Mills (40 CFR Part 406)
Organic Chemicals, Plastics, and
Synthetic Fibers (40 CFR Part 414)
Meat and Poultry Products (40 CFR Part
432)
Rubber Manufacturing (40 CFR Part
428)
Apparel and Other Textile Products
Potential New Subcategory of Textile
Mills (40 CFR Part 4 10)
Organic Chemicals, Plastics, and
Synthetic Fibers (40 CFR Part 414)
Organic Chemicals, Plastics, and
Synthetic Fibers (40 CFR Part 414)
Organic Chemicals, Plastics, and
Synthetic Fibers (40 CFR Part 414)
Soap and Detergent Manufacturing (40
CFR Part 4 17)
Nonferrous Metals Manufacturing (40
CFR Part 421)
Phosphate Manufacturing (40 CFR Part
422)
Copper Forming (40 CFR Part 468)
Pharmaceutical Manufacturing (40 CFR
Part 439)
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Cement Manufacturing (40 CFR Part
411)
Electroplating (40 CFR Part 4 13)
Organic Chemicals, Plastics, and
Synthetic Fibers (40 CFR Part 414)
Inorganic Chemicals Manufacturing (40
CFR Part 415)
Expanded
NAICS Code
(Assigned at
the Facility
Level)
311119GRAIN
311119MPP
311119PH
311225FER
311999DPP
311999GRAIN
3119990CPSF
311999MPP
315992RUB
315992AP
324199OCPSF
3251200CPSF
325188OCPSF
325188SD
325188NMM
325188PHOS
325188COP
325188PH
325188NMF
325510CEM
325510ELEC
3255100CPSF
325510INORG
4-14
-------
Section 4 - Identification of Point Source Categories
Table 4-5. Facility-Level Point Source Category Assignment NAICS Codes
NAICS
32561 1 : Soap and Other Detergent
Manufacturing
325998: All Other Miscellaneous
Chemical Product and Preparation
Manufacturing
326199: All Other Plastics Product
Manufacturing
3311 11: Iron and Steel Mills
331221: Rolled Steel Shape
Manufacturing
331314: Secondary Smelting and
Alloying of Aluminum
331423: Secondary Smelting,
Refining, and Alloying of Copper
331491: Nonferrous Metal (except
Copper and Aluminum) Rolling,
Drawing, and Extruding
331492: Secondary Smelting,
Refining, and Alloying of
Nonferrous Metal (except Copper
and Aluminum)
331521: Aluminum Die-Casting
Foundries b
Primary Associated Point
Source Category
Soap and Detergent
Manufacturing (40 CFR Part
417)
Organic Chemicals, Plastics,
and Synthetic Fibers (40 CFR
Part 414)
Plastics Molding and Forming
(40 CFR Part 463)
Iron and Steel Manufacturing
(40 CFR Part 420)
Iron and Steel Manufacturing
(40 CFR Part 420)
Nonferrous Metals
Manufacturing (40 CFR Part
421)
Nonferrous Metals
Manufacturing (40 CFR Part
421)
Nonferrous Metals Forming
and Metal Powders (40 CFR
Part 471)
Nonferrous Metals
Manufacturing (40 CFR Part
421)
Nonferrous Metals
Manufacturing (40 CFR Part
421) and
Aluminum Forming (40 CFR
Part 467)
Other Associated Point Source
Categories
Organic Chemicals, Plastics, and
Synthetic Fibers (40 CFR Part 414)
Business Services Potential New Point
Source Category
Inorganic Chemicals Manufacturing (40
CFR Part 415)
Soap and Detergent Manufacturing (40
CFR Part 4 17)
Petroleum Refining (40 CFR Part 419)
Metal Finishing (40 CFR Part 433)
Pharmaceutical Manufacturing (40 CFR
Part 439)
Pesticide Chemicals (40 CFR Part 455)
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Electroplating (40 CFR Part 4 13)
Organic Chemicals, Plastics, and
Synthetic Fibers (40 CFR Part 414)
Glass Manufacturing (40 CFR Part 426)
Metal Finishing (40 CFR Part 433)
Metal Finishing (40 CFR Part 433)
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Electroplating (40 CFR Part 4 13)
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Metal Finishing (40 CFR Part 433)
Metal Molding and Casting (40 CFR
Part 464)
Aluminum Forming (40 CFR Part 467)
Metal Molding and Casting (40 CFR
Part 464)
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Metal Finishing (40 CFR Part 433)
Copper Forming (40 CFR Part 468)
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Metal Molding and Casting (40 CFR
Part 464)
Expanded
NAICS Code
(Assigned at
the Facility
Level)
32561 10CPSF
325998BS
325998INORG
325998SD
325998PR
325998MF
325998PH
325998P
325998NMF
326199ELEC
326199OCPSF
326199GLASS
326199MF
331111MF
331111NMF
331221ELEC
331221NMF
331314MF
331314MMC
331314AL
331423MMC
331423NMF
331491MF
331492COP
331492NMF
331521MMC
4-15
-------
Section 4 - Identification of Point Source Categories
Table 4-5. Facility-Level Point Source Category Assignment NAICS Codes
NAICS
3321 12: Nonferrous Forging b
332618: Other Fabricated Wire
Product Manufacturing
332813: Electroplating, Plating,
Polishing, Anodizing, and
Coloring
332999: All Other Miscellaneous
Fabricated Metal Product
Manufacturing
336340: Motor Vehicle Brake
System Manufacturing
336360: Motor Vehicle Seating
and Interior Trim Manufacturing
337215: Showcase, Partition,
Shelving, and Locker
Manufacturing
Primary Associated Point
Source Category
Nonferrous Metals
Manufacturing (40 CFR Part
421);
Aluminum Forming (40 CFR
Part 467); and
Copper Forming (40 CFR Part
468)
Metal Finishing (40 CFR Part
433)
Electroplating (40 CFR Part
413)
Metal Finishing (40 CFR Part
433)
Metal Finishing (40 CFR Part
433)
Textile Mills (40 CFR Part
410)
Metal Finishing (40 CFR Part
433)
Other Associated Point Source
Categories
Iron and Steel Manufacturing (40 CFR
Part 420)
Metal Finishing (40 CFR Part 433)
Iron and Steel Manufacturing (40 CFR
Part 420)
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Printing and Publishing Potential New
Point Source Category a
Iron and Steel Manufacturing (40 CFR
Part 420)
Metal Finishing (40 CFR Part 433)
Plastics Molding and Forming (40 CFR
Part 463)
Aluminum Forming (40 CFR Part 467)
Printing and Publishing Potential New
Point Source Category a
Metal Finishing (40 CFR Part 433) and
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Aluminum Forming (40 CFR Part 467);
Copper Forming (40 CFR Part 468); and
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Electroplating (40 CFR Part 4 13)
Metal Finishing (40 CFR Part 433)
Timber Products Processing (40 CFR
Part 429)
Expanded
NAICS Code
(Assigned at
the Facility
Level)
332112IRON
332112MF
33261 8IRON
33261 8NMF
33261 8PP
33281 3IRON
332813MF
33281 3PMF
332813AL
33281 3PP
332999DC b
332999TC b
336340ELEC
336360MF
33721 5TIM
4-16
-------
Section 4 - Identification of Point Source Categories
Table 4-5. Facility-Level Point Source Category Assignment NAICS Codes
NAICS
339999: All Other Miscellaneous
Manufacturing
Primary Associated Point
Source Category
Metal Finishing (40 CFR Part
433)
Other Associated Point Source
Categories
Organic Chemicals, Plastics, and
Synthetic Fibers (40 CFR Part 414)
Mineral Mining and Processing (40 CFR
Part 436)
Pesticide Chemicals (40 CFR Part 455)
Plastics Molding and Forming (40 CFR
Part 463)
Nonferrous Metals Forming and Metal
Powders (40 CFR Part 471 )
Expanded
NAICS Code
(Assigned at
the Facility
Level)
3399990CPSF
339999MIN
339999P
339999PMF
339999NMF
a As part of the Final 2006 Plan, EPA determined that categorical pretreatment standards were not warranted for the
these industries (U.S. EPA, 2006b).
b A single NAICS code may include facilities subject to more than one point source category. EPA included the
loads from NAICs codes associated with multiple point source categories in the load for each category, double- or
triple-counting the loads from these NAICS codes (see Section 4.3.1.1).
EPA is currently considering revisions to a subpart of the ELGs for Organic Chemicals,
Pesticides, and Synthetic Fibers (OCPSF) (40 CFR 414) and Inorganic Chemicals Manufacturing
(40 CFR 415) for facilities that produce chlorine and chlorinated hydrocarbons (CCH). Because
the CCH rulemaking is underway, EPA assigned the NAICS code "VCCA" to the CCH facilities
in the NAICS/Point Source Category Crosswalk to separately identify these facilities (Note:
VCCA, vinyl chloride and chlor alkali, is the former name of the CCH rulemaking). The list of
CCH facilities in TRIReleases2007 is included in Table C-6 in Appendix C.
As part of the Pulp, Paper, and Paperboard Category (Pulp and Paper Category) (40 CFR
Part 430) Detailed Study (U.S. EPA, 2006a), EPA reviewed the operations of facilities reporting
SIC codes 2611: Pulp Mills, 2621: Paper Mills, and 2631: Paperboard Mills to determine the
applicable subpart for each facility. A 1988 legal suit obligated EPA to address discharges of
poly chlorinated dibenzo-(p)-dioxins and poly chlorinated dibenzofurans from 104 bleaching pulp
mills, including nine dissolving pulp mills. During its response to the 1988 legal suit, EPA
decided to review and revise the Pulp and Paper Category regulations in three phases. EPA
addressed Phase I first, chose not to revise the ELGs for Phase II, and chose to support NPDES
permit writers individually in developing permit-specific effluent limitations to control
discharges of these chemicals from the remaining operating mills in Phase III. Because the Pulp
and Paper Category regulations are subcategorized by process used and product produced they
do not correspond to SIC codes. Therefore, EPA added "-1" to the SIC codes of facilities that
met the applicability of Phase I:
• Subpart B (Bleached Papergrade Kraft and Soda); and
• Subpart E (Papergrade Sulfite).
EPA added "-2" to the SIC codes of facilities that met the applicability of Phase II:
Subpart C (Unbleached Kraft);
Subpart F (Semi-Chemical);
4-17
-------
Section 4 - Identification of Point Source Categories
• Subpart G (Groundwood, Chemic-Mechanical, and Chemi-Thermo-Mechanical);
• Subpart H (Non-Wood Chemical Pulp);
• Subpart I (Secondary Fiber Deink);
• Subpart J (Secondary Fiber Non-Deink);
• Subpart K (Fine and Lightweight Papers from Purchased Pulp); and
• Subpart L (Tissue, Filter, Non-Woven and Paperboard from Purchased Pulp).
EPA added "-3" to the SIC codes of facilities that met the applicability of Phase III:
• Subpart A (Dissolving Kraft); and
• Subpart D (Dissolving Sulfite).
EPA carried the facility-specific changes from the 2004 and 2005 screening-level reviews to the
TRIReleases2007 database by appending "-1", "-2", and "-3" to the NAICS codes for facilities in
the Pulp and Paper Category.
Discharge-Level Point Source Category Assignment
Regulations for the Electroplating Category (40 CFR Part 413) apply to discharges from
indirect discharging facilities, while direct discharging electroplating operations are regulated by
the Metal Finishing Category (40 CFR Part 433). EPA determined facilities reporting the
following NAICS codes may have electroplating operations:
• 325510: Paint and Coating Manufacturing;
• 326199: All Other Plastics Products Manufacturing;
• 331221: Rolled Steel Shape Manufacturing;
• 332813: Electroplating, Plating, Polishing, Anodizing, and Coloring; and
• 336340: Motor Vehicle Brake System Manufacturing.
In TRIReleases2007 facilities can report direct and indirect discharges. Therefore, as part of the
2009 screening-level review, for facilities reporting the above NAICS codes EPA assigned direct
discharges to the Metal Finishing Category by appending "MF" to the facility's discharge-level
NAICS code (e.g., EPA revised direct discharges from NAICS code 332813 to 332813MF).
Pollutant-Level Point Source Category Assignment
For most facilities that discharge wastewater subject to more than one point source
category, EPA was not able to divide the pollutant discharges between applicable point source
categories. The following subsections discuss two exceptions where EPA was able to assign
wastewater discharges of certain chemicals to the appropriate point source category.
Organic Chemicals, Plastics, and Synthetic Fibers/Pesticides
The OCPSF ELGs (40 CFR Part 414) may apply to discharges from facilities in the
following NAICS codes:
• 325132: Synthetic Organic Dye and Pigment Manufacturing;
• 325192: Cyclic Crude and Intermediate Manufacturing;
• 325199: All Other Basic Organic Chemical Manufacturing;
-------
Section 4 - Identification of Point Source Categories
• 325211: Plastics Material and Resin Manufacturing;
• 325221: Cellulosic Organic Fiber Manufacturing; and
• 325222: Noncellulosic Organic Fiber Manufacturing.
In addition, EPA is considering including operations from the following 16 NAICS codes as
potential new subcategories of the OCPSF Category:
• 311999OCPSF: All Other Miscellaneous Food Manufacturing;
• 324199OCPSF: All Other Petroleum and Coal Products Manufacturing;
• 325510: Petrochemical Manufacturing;
• 325120OCPSF: Industrial Gas Manufacturing;
• 325188OCPSF: All Other Basic Inorganic Chemical Manufacturing;
• 325193: Ethyl Alcohol Manufacturing;
• 325510OCPSF: Paint and Coating Manufacturing;
• 325520: Adhesive Manufacturing;
• 325611 OCPSF: Soap and Other Detergent Manufacturing;
• 325612: Polish and Other Sanitation Good Manufacturing;
• 325620: Toilet Preparation Manufacturing;
• 325998: All Other Miscellaneous Chemical Product and Preparation
Manufacturing;
• 326199OCPSF: All Other Plastics Product Manufacturing;
• 339999OCPSF: All Other Miscellaneous Manufacturing;
• 424690: Other Chemical and Allied Products Merchant Wholesalers; and
• 562920: Materials Recovery Facilities.
Some facilities in the regulated NAICS codes and NAICS codes of the potential new subcategory
manufacture and/or formulate pesticides as well as other organic chemicals. Regulations for the
Pesticide Chemicals Category (40 CFRPart 455) control discharges from pesticide operations.
For the screening-level analysis of discharges from existing categories, EPA therefore subtracted
all pesticide discharges from OCPSF and counted them as discharges from the Pesticides
Chemicals Category, by appending a "P" to the facility's pollutant-level SIC code (e.g., EPA
revised pesticide discharges from NAICS code 325199 to NAICS code 325199P). EPA
developed this methodology as part of the 2005 annual review for use with the PCS and TRI
data. EPA did not change the methodology for the 2009 annual review.
EPA used a table containing a list of pesticides and their CAS numbers in order to
identify the pesticide releases from the OCPSF Category for both the DMR and TRI databases.
In developing the list of pesticides, EPA started with the list of 272 pesticide active ingredients
that was created during the most recent pesticides rulemaking. Some of the pesticides in the list
of 272 active ingredients were multiple compounds, for example "2,4 D salts and esters" and
"organo-tin pesticides," and were not identified by CAS number. EPA identified individual
chemicals and CAS numbers for active ingredients in these groups and added them to the
pesticides list. All of the chemicals identified from the list of 272 pesticide active ingredients
were included in the pesticides list, except for biphenyl and dichlorobenzene. Biphenyl and
dichlorobenzene were not included because EPA determined that OCPSF facilities use these
chemicals for specific manufacturing uses not related to pesticides.
4-19
-------
Section 4 - Identification of Point Source Categories
EPA identified additional pesticide active ingredients by using the 1988 FIFRA and
TSCA Enforcement System (FATES) Database and a list created in 2003 by the Office of
Pesticide Programs (OPP). EPA combined the two lists and determined which of the pesticide
active ingredients facilities reported having discharged in TRI in 2007. For releases reported in
the 2007 TRI, EPA determined whether the pesticide active ingredient had significant non-
pesticide related manufacturing uses. Chemicals, such as acrolein, trichlorofuoromethane, silver,
and sulfuric acid, whose primary use was non-pesticide-related were not added to the list, while
chemicals whose primary purpose was pesticide-related were added to the list. The list of
chemicals reported in TRI and DMR that EPA considered pesticides for the purpose of its
screening-level analysis of discharges from existing categories contains 415 chemicals.
MP&M/Metal Finishing
Regulations for the Metal Finishing Category (40 CFR Part 433) may apply to discharges
from facilities in 198 NAICS codes for which discharges were reported in TRI in 2007.
Regulations for the Metal Products and Machinery (MP&M) Category (40 CFR Part 438) may
apply to some of the pollutants directly discharged by facilities in 165 of these NAICS codes.
The final MP&M rule at 40 CFR Part 438. l(b) specifically excludes both metal-bearing
wastewaters and wastewaters subject to other effluent guidelines (e.g., Metal Finishing). For its
screening-level analysis of discharges from existing categories, EPA developed methodologies to
apportion pollutant loads between the MP&M and Metal Finishing Categories.
The MP&M rule as promulgated regulates oil and grease (O&G) and total suspended
solids (TSS) in direct discharges from certain facilities that generate oily wastewater; it does not
specifically regulate any other chemicals. EPA used the list of organic "pollutants of concern" it
had developed for the MP&M rule and identified 48 pollutants in the TRI databases, including
O&G and TSS. For the 2009 screening-level analysis, EPA counted all discharges of these
pollutants from the 165 MP&M NAICS codes in TRIReleases2007 as MP&M discharges. EPA
counted discharges of all other chemicals from these facilities in the Metal Finishing Category in
TRIReleases2007. EPA believes that the identified pollutants are those that are most likely
associated with the non-metal bearing oily waste streams subject to the MP&M regulations, and
that this apportionment, which avoids double counting pollutant loads, is a reasonable approach
for screening-level analysis of discharges from existing categories.
For the 2009 annual review, as for previous reviews, EPA matched TRI pollutants to the
list of 88 MP&M chemicals using CAS numbers and the SUPERCAS table (described in Section
1). Using the SUPERCAS table, EPA matched 48 chemicals to the list of 88 organic "pollutants
of concern" for the MP&M rule that are discharged by facilities in the 165 MP&M NAICS
codes. EPA identified these 48 chemicals as "Controlled by MP&M." Table C-4 in Appendix C
lists the 88 organic "pollutants of concern" for the MP&M rule. Table C-7 in Appendix C
presents the list of TRI chemicals allocated to MP&M for the 2009 annual review.
4.4 Potential New Point Source Categories
Concurrent with its review of existing point source categories, EPA also reviews
industries not currently subject to effluent guidelines to identify potential new point source
categories. EPA conducts a "crosswalk" analysis based on data in DMR and TRI. Facilities with
data in DMR and TRI are identified by a four-digit SIC code or six-digit NAICS code (Section
4.1 provides more details on SIC and NAICS codes, respectively). EPA links each four-digit SIC
4^20
-------
Section 4 - Identification of Point Source Categories
code and six-digit NAICS code to an appropriate industrial category (i.e., "the crosswalk").19
This crosswalk identifies SIC codes and NAICS codes that EPA associated with industries
subject to an existing guideline. The crosswalk also identifies SIC and NAICS codes not
associated with an existing guideline. In addition to the crosswalk analysis, EPA relies on
stakeholder comments to identify potential new point sources categories.
For each industry identified through the crosswalk analysis or stakeholder comments,
EPA evaluates whether it constitutes a potential new category subject to identification in the plan
or whether it is properly considered a potential new subcategory of an existing point source
category. To make this determination, EPA generally looks at whether the industry produces a
similar product or performs a similar service as an existing category. If so, EPA generally
considers the industry to be a potential new subcategory of that category. If, however, the
industry is significantly different from existing categories in terms of products or services
provided, EPA considers the industry as a potential new stand-alone category subject to
identification in the plan.
4.4.1 Direct Discharges
Because the CWA has different requirements for potential new categories of direct and
indirect dischargers, EPA examines potential new categories to determine if the category
comprises mostly indirect dischargers or if it comprises both direct and indirect dischargers. If a
category consists largely of indirect dischargers, EPA evaluates the pass-through and
interference potential of the category discharges (see Section 3.4 of the Technical Support
Document for the Preliminary 2010 Effluent Guidelines Program Plan (U. S. EPA, 2009)). If a
category consists largely of direct dischargers, EPA evaluates the type of pollutants discharged
by facilities in the category.
4.4.2 Indirect Discharges
For potential new categories with primarily indirect dischargers, EPA evaluates the
potential for the wastewater discharges to "interfere with, pass through, or [be] otherwise
incompatible with" the operation of POTWs. See 33 U.S.C. § 1371(b)(l). Using available data,
EPA reviews the types of pollutants in an industry's wastewater. Then, EPA reviews the
likelihood of those pollutants to pass through a POTW. For most categories, EPA evaluated the
"pass through potential" as measured by: (1) the total annual TWPE discharged by the industrial
sector; and (2) the average TWPE discharge among facilities that discharge to POTWs. EPA also
assesses the interference potential of the discharge. Finally, EPA considers whether the pollutant
discharges are already adequately controlled by general pretreatment standards and/or local
pretreatment limits.
4.5 Identification of Point Source Category References
1. Office of Management and Budget (OMB). 1987. Standard Industrial Classification
Manual. (Unknown). EPA-HQ-TRI-2008-0564-0070.
19 For additional information on "the crosswalk," see Section 4.0 of the 2009 Technical Support Document for the
Annual Review of Existing Effluent Guidelines and Identification of Potential New Point Source Categories (U.S.
EPA, 2009).
4-21
-------
Section 4 - Identification of Point Source Categories
2. U.S. Census Bureau. 2009. North American Industry Classification System (NAICS)
Introduction. (Unknown). EPA-HQ-OW-2007-0571 DCN 06730.
3. U. S. EPA. 2005. 2005 Annual Screening-Level Analysis: Supporting the Annual Review
of Existing Effluent Limitations Guidelines and Standards and Identification of New
Point Source Categories for Effluent Limitations and Standards. EPA-821 -B-05-003.
Washington, DC. (August). EPA-HQ-OW-2004-0032-0901.
4. U. S. EPA. 2006a. Final Report: Pulp, Paper, andPaperboard. EPA-821R-06-016.
Washington, DC. (November). EPA-HQ-OW-2004-0032-2249.
5. U.S. EPA. 2006b. Technical Support Document for the 2006 Effluent Guidelines
Program Plan. EPA-82l-R-06-018. Washington, DC. (December). EPA-HQ-OW-2004-
0032-2782.
6. U. S. EPA. 2009. Preliminary 2010 Effluent Guidelines Program Plan Technical Support
Document. Washington, D.C. EPA-82l-R-09-009. (October). EPA-HQ-OW-2008-0517
DCN 06703.
7. Wolford, Jessica. Eastern Research Group, Inc. 2005. Memorandum to 2006 Effluent
Guidelines Program Plan Docket. RE: Identification of Facilities for the Porcelain
Enameling Point Source Category. (August). EPA-HQ-OW-2004-0032-0945.
4-22
-------
Section 5 - Toxic Weighting Factors
5. Toxic WEIGHTING FACTORS
DMRLoads2007 and TRIReleases2007 provide chemical discharge information in the
form of mass loads. In order to rank the human health and environmental hazard potential of
these loads, EPA estimates toxic-equivalent mass discharges using toxic weighting factors
(TWFs). EPA's Engineering and Analysis Division (EAD) developed TWFs for use in its
effluent limitations guidelines and standards (ELGs) development program to allow comparison
of pollutants with varying toxicities. The toxic-weighted pound equivalent (TWPE) is the mass
of a pollutant or chemical discharged that accounts for its relative toxicity. EPA calculates
TWPE by multiplying the mass (in pounds) of the chemical by its TWF. The remainder of this
section is divided into the following subsections:
• Section 5.1 - TWF background and development;
• Section 5.2 - New TWFs developed during the 2009 Annual Review;
• Section 5.3 - Chemicals for which EPA has not developed TWFs; and
• Section 5.4 - TWF References.
5.1 TWF Background and Development
In developing ELGs, EPA developed a wide variety of tools and methodologies to
evaluate effluent discharges. EPA's Office of Water, EAD maintains a Toxics Database
compiled from over 100 references for more than 1,900 pollutants. The Toxics Database includes
aquatic life and human health toxicity data, as well as physical and chemical property data. The
pollutants in this database are identified by a unique Chemical Abstract Service (CAS) number.
EPA calculates TWFs from these data to account for differences in toxicity across pollutants and
to provide the means to compare mass loadings of different pollutants. In its analyses, EPA
multiplies a mass loading of a pollutant in pounds per year (Ib/yr) by a pollutant-specific
weighting factor to derive a "toxic-equivalent" loading (Ib-equivalent/yr). Throughout this
document, the toxic-equivalent is also referred to as toxic-weighted pound equivalents, or
TWPE. The development of TWFs is discussed in detail in the Draft and Final TWF
Development Documents (U.S. EPA, 2005; U.S. EPA, 2006).
EPA derives TWFs from chronic aquatic life criteria (or toxic effect levels) and human
health criteria (or toxic effect levels) established for the consumption offish. In the TWF method
for assessing water-based effects, these aquatic life and human health toxicity levels are
compared to a benchmark value that represents the toxicity level of a specified pollutant. EPA
selected copper, a metal commonly detected and removed from industrial effluent, as the
benchmark pollutant. The Final TWF Development Document contains details on how EPA
developed its TWFs (U.S. EPA, 2006). Table D-l in Appendix D lists the TWFs for those
chemicals in the DMRLoads2007 and TRIReleases2007 databases for which EPA has developed
TWFs.
5.2 New Toxic Weighting Factors Developed During the 2009 Annual Review
During the 2009 annual review, EPA revised the TWF for boron to reflect updated
information. EPA did not revise any other TWFs or develop TWFs for any chemicals that had
not previously had TWFs as part of the 2009 annual review (Abt, 2008). Table 5-1 lists the
revised boron TWF. Boron is reported in both DMRLoads2007 and TRIReleases2007'.
-------
Section 5 - Toxic Weighting Factors
Table 5-1. Revised Boron TWF
Pollutant
Boron
CAS Number
7440428
Old TWF
0.177
New TWF
0.0083
Source: Memorandum to Josh Hall, U.S. EPA. Subject: Revised Draft - Updating the Boron TWF (Abt, 2008).
5.3 Chemicals without Toxic Weighting Factors
BAD has not yet developed TWFs for all chemicals in the DMRLoads2007 and
TRIReleases2007 databases. Table 5-2 lists the 29 chemicals in TRIReleases2007 that do not
have TWFs. The total discharge of the chemicals in Table 5-2 for TRIReleases2007 is
17,100,000 pounds. Table 5-3 lists the chemicals inDMRLoads2007that do not have TWFs. The
total discharge of the chemicals in Table 5-3 for DMRLoads2007 is 9.52 billion pounds. Of these
discharges, 3 percent relate to nitrogen- and phosphorous- containing compounds that may act as
nutrients. TWFs are not good indicators of the impact of nutrients on water quality. While
nutrients may have toxic effects that can be reflected in TWFs, their more important effect on
water quality occurs through their promotion of eutrophication20. EPA conducted a screening-
level analysis of nutrient discharges, which ranked point source categories based on
DMRLoads2007 nitrogen and phosphorous compound loads. The results of this analysis are
presented in Section 3.2.5.
Table 5-2. Chemicals with no TWFs in TRIReleases2007
CAS Number
872504
N503
7782414
N120
306832
191242
1344281
75456
149304
2837890
554132
94360
N583
64755
28407376
Chemical Name
N-METHYL-2-PYRROLIDONE
NICOTINE AND SALTS
FLUORINE
DIISOCYANATES
2,2-DICHLORO- 1,1,1 -TRIFLUOROETHANE
BENZO(G,H,I)PERYLENE
ALUMINUM OXIDE (FIBROUS FORMS)
CHLORODIFLUOROMETHANE
2-MERCAPTOBENZOTHIAZOLE
2-CHLORO- 1,1,1 ,2-TETRAFLUOROETHANE
LITHIUM CARBONATE
BENZOYL PEROXIDE
POLYCHLORINATED ALKANES
TETRACYCLINE HYDROCHLORIDE
C.I. DIRECT BLUE 2 18
Total Pounds Released a
13,999,796
2,818,643
97,777
38,774
37,940
34,819
34,495
33,565
20,573
17,219
11,444
2,996
2,705
804
302
Eutrophication occurs when nitrogen, phosphorous, and other nutrients in a body of water stimulate the growth of
algae. Nutrients flow through ecosystems constantly and eutrophication is a natural process that gradually turns
ponds into wetlands and wetlands into meadows. However, when human activity introduces additional nutrients to
the natural system, algal growth can become extreme and overwhelm the ecosystem's capacity. This over-
fertilization can cause increased turbidity, nuisance, or toxic, algal blooms, changes in biota, and anoxia. All of these
effects reduce the level and value of ecosystem services provided by water bodies.
5-2
-------
Section 5 - Toxic Weighting Factors
Table 5-2. Chemicals with no TWFs in TRIReleases2007
CAS Number
422560
924425
79947
764410
612839
71751412
354143
26628228
26471625
75683
1928434
98884
7664939
7647010
Chemical Name
3 ,3 -DICHLORO- 1,1,1 ,2,2-PENTAFLUOROPROPANE
N-METHYLOLACRYL AMIDE
TETRABROMOBISPHENOL A
1 ,4-DICHLORO-2-BUTENE
3 ,3 '-DICHLOROBENZIDINE DIHYDROCHLORIDE
ABAMECTIN
1 , 1 ,2,2-TETRACHLORO- 1 -FLUOROETHANE
SODIUM AZIDE
TOLUENE DIISOCYANATE (MIXED ISOMERS)
1 -CHLORO- 1 , 1 -DIFLUOROETH ANE
2,4-D 2-ETHYLHEXYL ESTER
BENZOYL CHLORIDE
SULFURIC ACID (1994 AND AFTER "ACID AEROSOLS"
ONLY)
HYDROCHLORIC ACID (1995 AND AFTER "ACID
AEROSOLS" ONLY)
Total
Total Pounds Released a
239
158
23
21
9
7
5
5
0.4
0.02
0.0002
0
0
0
17,100,000
Source: TRIReleases2007_v2.
a Includes transfers to POTWs and accounts for POTW removals.
Table 5-3. Chemicals with no TWFs in DMRLoads2007
CAS Number
00900
TSS
00515
BODS
79855
78470
00340
78477
51503
81017
00300
00335
CARBN
80103
CBOD
00181
PHOSP
PRAM Code
471341
7727379
10043524
7782447
7440440
Chemical Name
Hardness, total (as CaCO3)
Total Suspended Solids
Residue, totfltrble (dried at 105 C)
BOD, 5-day
Adsorbable organic halides (AOX)
Nitrogen, sludge, tot, dry wt. (as N)
Oxygen demand, chem. (high level) (COD)
Solids, sludge, tot, dry weight
Calcium Chloride
Chemical Oxygen Demand (COD)
Oxygen, dissolved (DO)
Oxygen demand, chem. (low level) (COD)
Total Carbon
Chemical oxygen demand (COD)
Carbonaceous BOD, 5-day
Oxygen demand, ultimate
Phosphorus
Total Pounds Released
3,043,209,486
3,039,742,485
1,069,455,416
384,518,697
281,109,733
270,173,870
269,728,827
212,062,831
173,744,369
158,007,360
138,838,601
134,291,135
115,774,628
36,770,720
30,325,446
24,249,650
21,673,603
5-3
-------
Section 5 - Toxic Weighting Factors
Table 5-3. Chemicals with no TWFs in DMRLoads2007
CAS Number
03594
TKN
00410
SIO2
00440
46570
80108
78240
00341
NOX
00343
34044
ORGN
80087
32017
51450
78115
00640
00319
TTC1A
51404
71872
34045
78733
70353
PO4
PO4
82209
HC
00415
51360
39942
78157
U238
00551
03773
PO4ASP
61194
51521
PRAM Code
7727379
471341
7631869
71523
7727379
7727379
7647145
7727379
13863417
14265442
308067530
98486
63231516
1338245
7440611
14265442
335671
Chemical Name
Halogens, adsorbable organic
Total Kjeldahl Nitrogen
Alkalinity, total (as CaCO3)
Silica
Bicarbonate ion- (as HCO3)
Hardness, Ca Mg Calculated (mg/L as CaCO3)
Chemical oxygen demand (COD)
Metals, total
Oxygen demand, chem. (COD), dissolved
Nitrogen, oxidized
Oxygen demand, total (tod)
Oxidants, total residual
Nitrogen, organic
BOD, carbonaceous, 20 day, 20 C
Sodium chloride (salt)
Nitrite Plus Nitrate Total
Halogen, total organic
Nitrogen, inorganic total
BOD, (ult. all stages)
Static 4Day Chronic Selen. Capricornutum
Solids, total suspd. non-volatile
Bromine chloride
Oxidants, free available
Volatile fraction organics (EPA 624)
Organic halides, total
Phosphate
Phosphate
Chlorides & sulfates
Total Hydrocarbons
Alkalinity, phenolphthaline method
m-Benzenedisulfonic acid
Hydrocarbons, aromatic
Naphthenic acid
Uranium 23 8
Hydrocarbons, in H2O, IR, CC14 extractible
chromatograph
Chlorine produced oxidants
Phosphate as P
Halogen, total residual
Perfluorooctanoic Acid
Total Pounds Released
11,634,478
7,890,284
7,490,718
5,617,674
5,265,556
5,048,590
4,916,094
4,099,529
3,137,358
3,027,222
2,893,253
1,802,491
1,456,316
1,448,164
1,077,614
854,958
557,377
343,917
330,952
318,731
271,229
259,617
220,001
189,068
185,054
173,386
173,386
165,812
111,168
108,716
107,605
106,144
93,603
78,333
66,577
60,594
58,401
56,810
46,552
5-4
-------
Section 5 - Toxic Weighting Factors
Table 5-3. Chemicals with no TWFs in DMRLoads2007
CAS Number
45501
CFA
04370
FLORB
77517
51340
82214
31667
80279
78218
80996
80126
00740
71845
78239
72035
00314
71870
50008
H2O2
49922
51065
04251
00664
78724
51526
03604
22456
47021
51523
00141
77066
51522
82180
01210
78221
51525
85789
DMDS
PRAM Code
479618
16872110
98113
98679
8002059
999
14265453
14798039
24959679
7722841
3825261
41663847
999
130498292
497267
7440053
754916
1563388
624920
Chemical Name
Petrol hydrocarbons, total recoverable
Chlorophyll A
Sum BOD and ammonia, water
Fluoroborates
Benzenesulphonic acid
p-Phenolsulfonic acid
pH change (range)
Oil petroleum, total recoverable
CBOD5/NH3-N
Phenolic compounds, unchlorinated
Spray irrigation
BOD, carbonaceous, 5 day, 5 C
Sulfite (as SO3)
Nitrogen, ammonia total (as NH4)
Metals, tox priority pollutants, total
Pump hours
BOD, nitrogen inhib 5-day (20 deg. C)
Bromide (as Br)
Priority pollutants total effluent
Hydrogen peroxide
Diesel range organics diesel, total, wtr
Ammonium perfluoroctanoate
Clamtrol CT-1, Total Water a
Dock discharge of phosphorus b
4-Nitro-N-methylphthalimide, total
Perfluorooctanesulfonate
Total phenols
Polynuc aromatic HC per Method 610
Methylene blue active substances
Perfluorobutanoicsulfonate
Solids, total susp per production
2-Methyl- 1 ,3 -dioxolane
Perfluorobutanoic Acid
Hydrocarbons, petroleum
Palladium, total (as Pd)
Organic pesticide chemicals (40 CFR455)
Perfluorooctanesulfonamide
2,2-Dimethyl-2,3-dihydro-7-benzofuranol
Dimethyl disulfide
Total Pounds Released
40,436
33,343
32,830
30,938
30,034
30,034
29,986
23,635
20,650
18,149
18,004
15,707
14,980
12,154
11,888
11,177
9,199
8,510
7,946
5,713
3,556
3,129
2,600
2,267
2,057
1,867
1,784
1,599
1,475
1,426
1,287
1,224
1,185
1,082
842
809
792
689
649
5-5
-------
Section 5 - Toxic Weighting Factors
Table 5-3. Chemicals with no TWFs in DMRLoads2007
CAS Number
01277
01142
39117
51493
00988
49875
HCCB
82560
78456
51524
CLPHN
00144
74052
34283
78155
81559
85795
39084
84085
U308
77247
85812
49491
00741
74053
77102
34521
ABS
82195
51051
00696
77226
77672
51437
51438
51165
78143
81512
77542
PRAM Code
7440213
999
5131668
27154445
1336352
39638329
30498352
683534
7440611
2809214
14265453
872504
191242
42615292
108678
120616
211578040
88164
95169
87854
Chemical Name
Total agg concentration #1
Silicon, total
Phthalate esters
Phenolic Compounds, Total
Iron and manganese, soluble
Propylene glycol monobutyl ether
Hexachlorocyclohexane
Total pesticides
Halomethanes, sum
Perfluorobutanesulfonamide
Chlorinated phenols
Combined metals sum
Chlorinated hydrocarbons, general
Bis(2-chloroisopropyl) ether
Dichlorobenzyl trifluoride
Bromodichloroethane
Xylene, meta & para in combination
Total purgeable halocarbons
Volatile organics detected
Uranium 308
Benzoic acids, total
1 -Hy droxy ethy lidene
BTEX
Sulfite (as S)
Pesticides, general
N-Methyl-2-pyrrolidone
Benzo(ghi)perylene
Alkyl benzene sulfonates
Thiocarbamates
Tin, tri-organo-
Nitrofurans
1 ,3 ,5-Trimethylbenzene
Dimethyl terephthalate
N-Hexane
SAS-310, Total
SAS - 305, total
Monochlorobenzyl trifluoride
Benzothiazole
Hexamethylbenzene
Total Pounds Released
624
588
520
507
497
496
484
465
374
321
313
282
276
223
157
155
137
122
96
77
74
60
41
37
35
32
24
24
16
16
13
12
12
11
11
8
6
5
4
5-6
-------
Section 5 - Toxic Weighting Factors
Table 5-3. Chemicals with no TWFs in DMRLoads2007
CAS Number
51202
71910
01168
34102
45097
78721
81611
73525
49702
38579
70027
39379
84103
77086
51009
82181
01279
76025
34679
00143
00987
00973
01289
00177
01278
01288
00148
78232
77625
77666
77676
77889
77983
78028
61916
78222
77081
78237
78732
PRAM Code
7440575
7440746
628966
98839
26523648
1338234
131748
108996
136677093
74931
1318098
103333
77929
30583336
706785
29797408
12408105
497187
144627
Chemical Name
Sulfide-hydrogen sulfide (undissociated)
Gold, total (as Au)
Indium
Ethylene glycol dinitrate
Methylstyrene
Phthalates, total
Trichlorotrifluoroethane
2-Butanone peroxide
Ammonium picrate
Benzene, halogenated
COD, 25N K2Cr207, tot
DDT/DDD/DDE, sum of p,p' & o,p' isomers
Dioxin laboratory - alpha code
3 -Methy Ipy ridine
RDX+HMX
Hydrocarbons, total petroleum
Total agg concentration #3
Chlorinated dibenzo-p-dioxins, effluent
2,3,7,8-TCDD TEC
Methyl mercaptan
Iron and manganese, total
Asbestos, total amphibole
Biocides
Oxygen demand, dissolved
Total agg concentration #2
Foaming agents
Herbicides, total
Total toxic organics (TTO) (40 CFR469)
Azobenzene
Citric acid
Trichlorotoluene
Octachlorocyclopentene
Dichlorotoluene
Tetrachlorobenzene
1,3-Diaminourea
Organic active ingredients (40 CFR455)
Oxalic acid
Organics, volatile (NJAC reg. 7:23-17e)
Volatile compounds, (GC/MS)
Total Pounds Released
4
4
4
4
3
2
2
2
1
1
0.3
0.2
0.1
0.1
0.004
0.001
0.0003
0.0001
0.000000004
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5-7
-------
Section 5 - Toxic Weighting Factors
Table 5-3. Chemicals with no TWFs in DMRLoads2007
CAS Number
79817
81328
82080
82602
DDAC
78171
51497
32015
34103
34730
38925
45670
49699
49886
77540
51132
77295
51539
51540
61026
70015
76028
76029
03768
51030
PRAM Code
95772
25323302
7173515
576249
13560899
84764
193700059
583788
108805
108430
4901513
Chemical Name
3 ,4-Dichlorophenol
Dichloroethene, total
Trihalomethane, tot.
Produced sand, weight
CalgonH-130M
Aromatics, total purgeable
Spectrus OX 1200
Base/neutral compounds
Benzene, toluene, xylene in combination
2,3-Dichlorophenol, total
Dechlorane plus
Dinonyl phthalate
Betz slimicide C-31, total
Betz clam-trol CT-2
2,5-Dichlorophenol
Cyanuric acid
3-Chlorophenol
Nonpurgeable Organic Halides
Purgeable Organic Halides
2,3,4,5-Tetrachlorophenol
Freon, total
Base neutrals & acid (Method 625), efflnt
Organics, tot purgeables (Method 624)
Purgeable hydrocarbons, Meth. 601
Spectrus CT 1300
Total
Total Pounds Released
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
9,520,000,000
Source: DMRLoads2007_v3.
aFrom the ZM Control Guide (Sprecher, 2000), Clam-Ttol CT-1 is a liquid substance that is 8 percent n-alkyl (C12-
40 percent, C14-50 percent, C16-10 percent) dimethylbenzyl ammonium chloride and 5 percent dodecylguanidine
hydrochloride.
b Dock discharge of phosphorous is required for facilities that operate a ship dock used primarily for loading and
unloading solids containing some compound of phosphorus (e.g., phosphate rock, ammoniated phosphates) (State of
Louisiana, 2004).
TEC - Total equivalent concentration.
5.4 Toxic Weighting Factor References
1. Abt Associates Inc. 2008. Memorandum to Josh Hall, U.S. EPA. RE: Revised Draft -
Updating the Boron TWF. Cambridge, MA. (December 5). EPA-HQ-OW-2008-0517
DCN 06729.
5-8
-------
Section 5 - Toxic Weighting Factors
2. Sprecher, Susan, Kurt Getsinger. 2000. Zebra Mussel Chemical Control Guide.
ERDC/EL TR-00-1. U.S. Army Engineer Research and Development Center. Vicksburg,
MS. (January). EPA-HQ-OW-2004-0032-2171.
3. State of Louisiana Department of Environmental Quality (LA DEQ). 2004. NPDES
Permit for IMC Phosphates Co, Faustina Plant. (May 24). EPA-HQ-OW-2004-0032-1134
and 1135.
4. U.S. EPA. 2005. Draft Toxic Weighting Factor Development in Support ofCWA 304(m)
Planning Process. Washington, DC. (June). EPA-HQ-OW-2004-0032-0857.
5. U.S. EPA. 2006. Toxic Weighting Factor Development in Support ofCWA 304(m)
Planning Process. Washington, DC. (June). EPA-HQ-OW-2004-0032-1634.
5-9
-------
Section 6 - Quality Review
6. QUALITY REVIEW
EPA's screening-level analysis involves the collection and use of existing environmental
data for purposes other than those for which they were originally collected. Pollutant Compliance
System (PCS) and Integrated Compliance Information System for the National Pollutant
Discharge Elimination System (ICIS-NPDES) were designed to automate entering, updating, and
retrieving NPDES data and to track permit issuance, permit limits and monitoring data, and other
data pertaining to facilities regulated under NPDES. The primary purpose of the Toxic Release
Inventory (TRI) is to collect and make public annual data on releases and transfers of certain
toxic chemicals from industrial facilities to inform communities and citizens of chemical hazards
in their areas. Sections 2 and 3 of this report describe how EPA used the data in PCS, ICIS-
NPDES, and TRI to calculate annual pollutant loadings to prioritize industrial category
discharges for further review.
To use data in PCS and ICIS-NPDES, EPA first combined the two datasets to form
DMRLoads2007', as described in Section 3. This section describes the quality review steps that
EPA uses to determine if the DMRLoads2007 and TRIReleases2007 data are suitable for EPA's
use in a screening-level analysis. The remainder of this section is divided into the following
subsections:
• Section 6.1- Overview of Quality Review Steps;
• Section 6.2 - Summary of DMRLoads2007 Quality Review;
• Section 6.3 - Summary of TRIReleases2007 Quality Review; and
• Section 6.4 - Quality Review References.
6.1 Overview of Quality Review Steps
EPA considered the following factors in its quality review of the PCS, ICIS-NPDES, and
TRI data:
• Completeness. The following information is needed to analyze the toxic weighted
pollutant loadings discharged by industrial categories:
— Facility identity,
— Industrial category under which the facility is regulated,
— Identity of parameters discharged and corresponding toxic weighting
factors (TWFs),
— Mass of pollutants discharged (or pollutant concentration and discharge
flow, from which the mass can be calculated), and
— Understanding of how available information represents the discharger
population and pollutant population.
• Accuracy. Analyzed data should accurately categorize and aggregate the
underlying database.
• Reasonableness. Pollutant identities must be reasonably related to the operations
in the category. Reported or calculated loads and facility wastewater flows should
reflect the range of flows and loads known to exist in the United States.
The following subsections discuss each of these factors in more detail.
-------
Section 6 - Quality Review
6.1.1 Completeness Checks
In previous years of review, EPA compared the number of facilities listed in the 2007
U.S. Economic Census to the number of facilities in the PCS and TRI databases, as described in
the report entitled 2005 Annual Screening-Level Analysis: Supporting the Annual Review of
Existing Effluent Limitations Guidelines and Standards and Identification of New Point Source
Categories for Effluent Limitations Guidelines and Standards to determine the extent to which
the facilities in the databases represent the entire industry (U.S. EPA, 2005). In 2009 for
categories selected for preliminary category review21, EPA compared the following statistics in
DMRLoads2007 and TRIReleases2007 to the 2002 U.S. Economic Census: the total number of
facilities, the number of facilities reporting wastewater discharges (direct or indirect) in TRI, and
the number of major and minor facilities in DMR. Table 6-1 lists EPA's findings for the 2007
databases.
Table 6-1. Number of Facilities in Categories Selected for Preliminary Category Review
Point Source Category
Fertilizer Manufacturing (40 CFR
Part 4 18)
Inorganic Chemicals Manufacturing
(40 CFR Part 4 15)
Nonferrous Metals Manufacturing
(40 CFR Part 421)
Ore Mining and Dressing (40 CFR
Part 440)
Organic Chemicals, Plastics and
Synthetic Fibers (40 CFR Part 414)
Petroleum Refining (40 CFR Part
419)
Pulp, Paper and Paperboard (40
CFR Part 430)
NAICS Codes
311225FER, 325312, 325311,
325314
325120, 325131, 325181, 325188,
325998INORG, 331311,
325510INORG
325188NMM, 331312, 331314,
331411, 331419, 331423, 331492,
331521
21220, 212234, 212231, 212221,
212222, 212291, 212299, 213114
311999OCPSF, 324199OCPSF,
325110, 325120OCPSF, 325132,
325188OCPSF, 325192, 325193,
325199, 325211, 325221, 325222,
3255 10OCPSF, 325520,
32561 1OCPSF, 325612, 325620,
325998, 326199OCPSF,
339999OCPSF, 424690, 562920
324110,324191,324199,
325998PR, 474710, 486110
321113-1, 322110, 322121, 322122,
322130, 322211, 322212, 322214,
322215, 322221, 322222, 322224,
322231,322291,322299
2002 U.S.
Economic
Census
>723
>1,335
>937
510
>17,125
>5,785
>4,876
2007
DMRa
85
394
114
57
903
1,393
448
2007 TRI b
110
414
338
76
2,032
780
464
Source: U.S. Economic Census, 2002 (U.S. Census, 2002), TRIReleases2007_v2, DMRLoads2007_v3.
a - Major and minor dischargers. Also, DMR data is reported by SIC code; therefore EPA used an NAICS to SIC
crosswalk for comparison purposes.
b - Releases to any media.
See Section 5.3 of the Technical Support Document for the Preliminary 2010 Effluent Guidelines Program Plan
(U.S. EPA, 2009) for information on how the categories were selected for preliminary category review.
6-2
-------
Section 6 - Quality Review
EPA also considered the pollutant discharges that do not contribute to the category
rankings. As discussed in Section 5, EPA identified and profiled the pollutant parameters that do
not have an assigned TWF. Table D-l in Appendix D lists the TWFs for those chemicals in the
DMRLoads2007 and TRIReleases2007 databases for which EPA has developed TWFs. Table 5-2
and Table 5-3 list the chemicals in the databases that do not have assigned TWFs, as well as the
total pounds of these pollutants estimated as discharged. This quality review showed that 52.6%
of 18.1 billion pounds of pollutants discharged in DMRLoads2007 are not currently assigned
TWF, while 22.2% of 195 million pounds of pollutants released in TRIReleases2007 are not
currently assigned a TWF.
6.1.2 Accuracy Checks
EPA verified the accuracy of database queries used to analyze DMRLoads2007 and
TRIReleases2007 data and generate output tables. As one team member created queries, a second
team member reviewed the logic of the programming code, and compared the number of records
in the output table to the number of records in intermediate queries. This ensured that no data
were missing and that there were no duplicate records. EPA documented the quality checks in a
database table that describes the function of each query created, the quality checks that were
performed, the name of the reviewer, the date the query was reviewed, and any errors that were
identified. Tables A-5 in Appendix A and B-15 in Appendix B present the quality check tables
for the TRIReleases2007 and DMRLoads2007 databases.
6.1.3 Reasonableness Checks
EPA ranked pollutant discharges and facilities by toxic weighted loadings to identify
discharges and loadings that are unusually high. EPA then conducted reasonableness checks on
the unusually high pollutant discharges and facility loads to determine if the unusual values were
misreported or miscalculated. The reasonableness checks are described in the following
subsections.
6.1.3.1 Pollutant Identity
EPA ranked the pollutants discharged from each point source category and, using
engineering understanding of industrial processes, verified that the pollutants composing the
majority of the load could be reasonably related to operations in the industry. For unexpected
results, EPA compared the reported releases to information in the facility's NPDES permit and
other available resources, such as facility descriptions and discussions with the facility contacts.
EPA corrected errors in DMRLoads2007 and TRIReleases2007 and documented the corrections.
For example, in the quality review of the TRIReleases2007 database, EPA identified a petroleum
refining facility that was reporting dioxin discharges that resulted in a large discharge estimate,
in terms of toxic weighted pound-equivalents (TWPE). EPA contacted the facility to verify that
the estimated discharge was based on actual measured dioxin, instead of measurements below
detection limits, because facilities often use non-detect values when estimating dioxin
discharges. The facility verified that the reported discharges were actually an overestimate of
their actual dioxin discharge, which follows EPA's guidance on TRI reporting22. This method is
The Office of Pollution Prevention and Toxic Substances provides guidance on how to report dioxin to TRI in the
document entitled Guidance for Reporting Toxic Chemicals within the Dioxin and Dioxin-like Compounds Category
(http://www.epa.gov/tri/guide docs/index.htm#chemicalsp). "For purposes of threshold determinations and the
6-3
-------
Section 6 - Quality Review
appropriate for TRI reporting purposes, but for the screening-level databases, EPA adjusted the
estimated dioxin discharge to represent the actual dioxin measured in wastewater.
6.1.3.2 Facility Loads
EPA reviewed the toxic weighted loadings of facilities to ensure that they compose a
reasonable percentage of the total national discharge. Facilities that compose a very high
percentage of the national discharge have a large impact on the point source category rankings.
EPA reviewed NPDES permit data or other available data to identify where a facility may have
made a calculation error or reported the incorrect units of measure, and contacted facilities to
confirm suspected errors. EPA corrected confirmed errors and documented the corrections. For
example, in the quality review of the DMRLoads2007 database, EPA identified a facility whose
calculated TWPE for dioxin was over a billion pound-equivalents. EPA contacted the facility's
regulatory authority and learned there were units errors as well as misinterpreted laboratory data.
The units error caused DMRLoads2007 to overestimate the dioxin load by a factor of 162
(Auchterlonie, 2009).
6.2 Quality Review of the DMRLoads2007 Database
As discussed in Section 3, to identify potential anomalous loads, EPA ranked
DMRLoads2007 facilities by total TWPE. For those facilities that ranked the highest for total
TWPE, EPA reviewed them carefully to verify the accuracy of the database. The
DMRLoads2007 review included the following tasks:
• Comparison of DMRLoads2007 to PCSLoads2004;
• Comparison of DMRLoads2007 loads to TRIReleases2007;
• Review of flow and concentration data for units errors;
• Review of reported discharge data and the estimated load for missing data;
• Review of permit limits;
• Verification of proper SIC code/point source category classification;
• Review of NPDES permit or fact sheet where available; and
• Discussion with facility contacts.
These steps were taken for each facility that seemed to have an unusually high TWPE.
Once a possible mistake was identified through the process listed above, EPA contacted the
facilities for verification of changes made to the database. Table B-13 in Appendix B presents
the resulting corrections identified.
reporting of releases and other waste management quantities for dioxin and dioxin-like compounds under EPCRA
Section 313, either with monitoring data, or by using the emission factor approach, non-detects are treated as 'zero'
if that is how the method being used treats non-detects (e.g., Method 1613, Method 23). However, facilities should
use their best readily available information to report, so if a facility has better information than provided by these
methods then that information should be used. For example, if a facility is not detecting dioxin or a particular
dioxin-like compound using a particular method but has information that shows that they should be detecting them
the facility should use this other information and it may be appropriate to estimate quantities using one half the
detection limit." This guidance results in many facilities using one-half the detection limit to estimate discharges for
years where no dioxin were detected in wastewater.
6-4
-------
Section 6 - Quality Review
6.3 Quality Review of the TRIReleases2007 Database
EPA ranked TRI facilities by total TWPE released to surface waters to identify potential
anomalous loads. The review of TRIReleases2007 included the following:
• Comparison of TRIReleases2007 loads to TRIReleases2006;
• Comparison of TRIReleases2007 loads to DMRLoads2007;
• Review of reported discharge data and the estimated load for missing data;
• Review of the basis of estimate used for reporting the pollutant discharge;
• Review of reported dioxin congener distributions;
• Verification of proper NAICS code/point source category classification;
• Discussions with facility contacts.
This review process was carried out for each facility that ranked among the highest for
total TWPE released to surface waters or transferred to POTWs in 2007. Comparing databases
and publically available discharge information made it possible to identify potential errors in the
database, which would result in a high TWPE for a facility. Facilities were contacted to verify
that the correct change to the data was taking place. Table A-3 in Appendix A presents the
resulting corrections identified.
6.4 Quality Review References
1. Auchterlonie, Steve. 2009. Notes from Telephone Conversation between Chris Krejci,
Eastern Research Group, Inc. and Steve Auchterlonie, Front St. Remedial Action. RE:
Verification of Magnitude and Basis of Estimate for Dioxin and Dioxin-Like Compound
Discharges in PCS. (March 13). EPA-HQ-OW-2008-517 DCN 06636.
2. U.S. Economic Census. 2002. Available online at: http://www.census.gov/econ/census02.
3. U. S. EPA. 2005. 2005 Annual Screening-Level Analysis: Supporting the Annual Review
of Existing Effluent Limitations Guidelines and Standards and Identification of Potential
New Categories for Effluent Limitations Guidelines and Standards. EPA-821-B-05-003.
Washington, DC. (August). EPA-HQ-OW-2004-0032-0901.
6-5
-------
Section 7 - Results of 2009 Screening-Level Analysis
7.
RESULTS OF 2009 SCREENING-LEVEL ANALYSIS
This section describes the results of the 2009 screening-level analysis and the
methodology used by EPA to prioritize categories for further review. This section also discusses
the identification of categories warranting detailed studies. The remainder of this section is
divided into the following subsections:
• Section 7.1 - Preliminary Results of the Screening-Level Review;
• Section 7.2 - Prioritization of Categories; and
• Section 7.3 - Identification of Categories for Further Review.
7.1 Preliminary Results of the Screening-Level Review
The purpose of the screening-level review is to evaluate the amount and toxicity of the
pollutants in an industrial category's discharges. Using TRIReleases2007 and DMRLoads2007,
EPA ranked point source categories according to their discharges of toxic and non-conventional
pollutants (reported in units of toxic-weighted pound equivalent or TWPE). As described earlier
in this report, EPA multiplied the pounds of pollutants discharged by toxic weighting factors
(TWFs) resulting in an estimate of TWPE. Discharges were assigned to industrial categories on
the basis of facility Standard Industrial Classification (SIC) and North American Industrial
T\
Classification System (NAICS) codes . Categories included both facilities subject to the
existing effluent guidelines for the category and those belonging to potential new subcategories
of existing effluent guidelines and to potential new categories.
Table 7-1 and Table 7-2 present, for categories for which EPA has promulgated effluent
guidelines and pretreatment standards (ELGs), the preliminary rankings using TRIReleases2007
and DMRLoads2007, respectively. Discharges from facilities that produce chlorine or
chlorinated hydrocarbons (CCH) are listed on these tables as a separate category. See Section
7.2.1 for further discussion. Table 7-1 and Table 7-2 include discharges associated with facilities
subject to the point source category applicability, as well as facilities that are associated with
potential new subcategories of existing categories.
Table 7-1. TRIReleases2007Point Source Category Rankings
Rank
1
2
3
4
5
6
7
40CFR
Part
414.1
414
423
430
419
420
433
Point Source Category
Chlorine And Chlorinated Hydrocarbons
Organic Chemicals, Plastics And Synthetic Fibers
Steam Electric Power Generating
Pulp, Paper And Paperboard
Petroleum Refining
Iron And Steel Manufacturing
Metal Finishing a
TWPE
7,270,000
575,000
542,000
460,000
172,000
104,000
62,000
23 DMR data from PCS and ICIS-NPDES in the DMRLoads2007 has facility SIC codes, while TRI data has NAICS
codes. See Section 5 - Identification of Point Source Categories for additional information on how EPA linked SIC
and NAICS codes to point source categories.
7-6
-------
Section 7 - Results of 2009 Screening-Level Analysis
Table 7-1. TRIReleases2007Point Source Category Rankings
Rank
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
40CFR
Part
415
440
421
432
458
455
429
417
471
463
439
428
425
469
NA
464
468
NA
418
437
413
407
467
436
405
410
406
461
438
426
434
411
424
422
443
465
408
466
Point Source Category
Inorganic Chemicals Manufacturing
Ore Mining And Dressing
Nonferrous Metals Manufacturing
Meat and Poultry Products
Carbon Black Manufacturing
Pesticide Chemicals
Timber Products Processing
Soap And Detergent Manufacturing
Nonferrous Metals Forming And Metal Powders
Plastics Molding And Forming
Pharmaceutical Manufacturing
Rubber Manufacturing
Leather Tanning And Finishing
Electrical And Electronic Components
Miscellaneous Foods And Beverages
Metal Molding And Casting (Foundries)
Copper forming
Tobacco Products
Fertilizer Manufacturing
Centralized Waste Treatment
Electroplating
Canned And Preserved Fruits And Vegetables Processing
Aluminum forming
Mineral Mining And Processing
Dairy products processing
Textile Mills
Grain mills
Battery Manufacturing
Metal Products And Machinery
Glass Manufacturing
Coal Mining
Cement Manufacturing
Ferroalloy Manufacturing
Phosphate Manufacturing
Paving And Roofing Materials (Tars And Asphalt)
Coil Coating
Canned And Preserved Seafood Processing
Porcelain Enameling
TWPE
54,700
44,400
38,900
35,900
32,400
24,700
16,300
14,600
8,830
8,780
8,000
7,860
7,800
7,550
6,580
6,110
4,950
4,760
4,460
3,790
3,210
2,960
2,710
2,420
2,400
2,390
2,080
1,640
917
546
493
452
340
250
249
241
234
164
7-7
-------
Section 7 - Results of 2009 Screening-Level Analysis
Table 7-1. TRIReleases2007Point Source Category Rankings
Rank
46
47
48
49
50
51
52
53
54
55
56
40CFR
Part
446
NA
445
454
444
NA
409
447
457
406
NA
Point Source Category
Paint Formulating
Printing And Publishing
Landfills
Gum And Wood Chemicals Manufacturing
Waste Combustors
Independent And Stand Alone Labs
Sugar Processing
Ink Formulating
Explosives Manufacturing
Hospitals
Drinking Water Treatment
TOTAL
TWPE
140
110
83
55
40
30
26
20
14
1
0
9,550,000
Source: TRIReleases2007_v2.
NA - Not applicable. These are potential new categories.
Table 7-2. DMRLoads2007 Point Source Category Rankings
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
40CFR
Part
423
433
430
414.1
418
420
432
414
419
415
421
440
455
NA
471
410
429
417
444
445
Point Source Category
Steam Electric Power Generating a
Metal Finishing b
Pulp, Paper And Paperboard °
Chlorine And Chlorinated Hydrocarbons
Fertilizer Manufacturing
Iron And Steel Manufacturing
Meat and Poultry Products
Organic Chemicals, Plastics And Synthetic Fibers d
Petroleum Refining
Inorganic Chemicals Manufacturing
Nonferrous Metals Manufacturing
Ore Mining And Dressing
Pesticide Chemicals
Drinking Water Treatment
Nonferrous Metals Forming And Metal Powders
Textile Mills
Timber Products Processing
Soap And Detergent Manufacturing
Waste Combustors
Landfills
Total TWPE
20,400,000
3,360,000
2,730,000
1,220,000
1,100,000
730,000
536,000
413,000
403,000
394,000
343,000
184,000
180,000
119,000
119,000
79,900
51,600
47,800
38,400
35,800
7-8
-------
Section 7 - Results of 2009 Screening-Level Analysis
Table 7-2. DMRLoads2007 Point Source Category Rankings
Rank
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
40CFR
Part
409
436
439
463
422
467
464
428
454
437
469
411
NA
NA
424
408
468
434
406
407
443
461
NA
457
412
426
NA
NA
435
465
405
460
466
425
451
NA
438
NA
Point Source Category
Sugar Processing
Mineral Mining And Processing
Pharmaceutical Manufacturing
Plastics Molding And Forming
Phosphate Manufacturing
Aluminum forming
Metal Molding And Casting (Foundries)
Rubber Manufacturing
Gum And Wood Chemicals Manufacturing
Centralized Waste Treatment
Electrical And Electronic Components
Cement Manufacturing
Miscellaneous Foods And Beverages
Independent And Stand Alone Labs
Ferroalloy Manufacturing
Canned And Preserved Seafood Processing
Copper forming
Coal Mining
Grain mills
Canned And Preserved Fruits And Vegetables Processing
Paving And Roofing Materials (Tars And Asphalt)
Battery Manufacturing
Printing And Publishing
Explosives Manufacturing
CAFO
Glass Manufacturing
Construction And Development
Airport Deicing
Oil & Gas Extraction
Coil Coating
Dairy products processing
Hospital
Porcelain Enameling
Leather Tanning And Finishing
Concentrated Aquatic Animal Production
Tobacco Products
Metal Products And Machinery
Photo Processing
Total TWPE
32,500
26,700
24,900
24,600
18,500
12,200
11,300
11,200
10,500
10,400
9,350
8,960
5,840
5,360
4,350
3,230
2,310
2,290
1,980
1,760
1,280
1,100
999
785
617
353
324
265
256
166
76
15
11
8
5
3
3
1
7-9
-------
Section 7 - Results of 2009 Screening-Level Analysis
Table 7-2. DMRLoads2007 Point Source Category Rankings
Rank
59
60
40CFR
Part
459
442
Point Source Category
Photographic
Transportation Equipment Cleaning
TOTAL
Total TWPE
1
0
32,700,000
Source: DMRLoads2007_v3.
a EPA corrected a suspected units error in DMRLoads2007_v3 for FB Culley Station in Newburgh, IN (IN0002259)
in the Steam Electric Power Generating Category. EPA attempted to contact the facility but the facility never
returned calls. Therefore, EPA was unable to verify the correction.
bEPA contacted General Electric in Erie, PA (PA0000183) in the Metal Finishing Category and identified a units
error in DMRLoads2007_v3 (Verderese, 2009). The new LEY and TWPE reported for this facility were recalculated
and are now 0.024 and 2.790, respectively. The new Metal Finishing Category TWPE is 571,500.
0 For the Pulp, Paper, and Paperboard Category, EPA contacted facilities to verify the concentrations of dioxin and
dioxin-like compounds in PCS and ICIS-NPDES. EPA found that, for all facilities contacted, there were either units
errors (e.g., reported as ng/L but in the database as mg/L) or missing non-detect indicators. The new Pulp, Paper,
and Paperboard Category total TWPE is 252,163. See Section 12.2.2.1 in the Technical Support Document for the
Preliminary 2010 Effluent Guidelines Program Plan (U.S. EPA, 2009) for additional details on the facilities-specific
corrections.
dEPA contacted GE Silicones in Friendly, WV (WV0000094), in the OCPSF Category and identified a units error in
DMRLoads2007_v3 (Martin, 2009). The new LEY and TWPE reported for this facility were recalculated and are
now 158 and 100.3, respectively. The new OCPSF Category total TWPE is 308,721.
NA - Not applicable. These are potential new categories.
7.2 Prioritization of Categories
For the 2009 screening-level review, EPA combined the results of the TRIReleases2007
and the DMRLoads2007 databases, which are presented in Sections 2 and 3 of this document,
respectively. When combining the results of these databases, EPA made adjustments to the
rankings for the following:
• Discharges from industrial categories for which EPA is currently developing or
revising ELGs;
• Discharges from point source categories for which EPA has recently promulgated
or revised ELGs; and
• Discharges from facilities determined not to be representative of their category.
Sections 7.2.1 through 7.2.4 discuss the rationale for these decisions. The final combined
database rankings represent the results of the 2009 screening-level review and are presented in
Section 7.2.5.
7.2.1 Categories for Which EPA is Currently Developing or Revising ELGs
EPA is currently considering revisions to ELGs for Organic Chemicals, Pesticides, and
Synthetic Fibers (OCPSF) (40 CFR 414) and the Inorganic Chemicals Manufacturing (40 CFR
415) Point Source Categories for facilities that produce Chlorine and Chlorinated Hydrocarbons
(CCH). Because the CCH rulemaking is underway, EPA excluded discharges from these
facilities from further consideration under the current planning cycle. EPA subtracted the Toxic
Weight Pollutant Equivalent (TWPE) loads from facilities that produce chlorine or chlorinated
hydrocarbons from the Organic Chemicals, Pesticides, and Synthetic Fibers (OCPSF) and
-------
Section 7 - Results of 2009 Screening-Level Analysis
Inorganic Chemicals Manufacturing Point Source Category loads. Because facilities that produce
chlorine and chlorinated hydrocarbons are only a subset of the OCPSF and Inorganic Chemicals
Manufacturing Categories, EPA included loads for all other facilities in these two categories in
the prioritization of categories for further review24.
7.2.2 Categories for Which EPA Recently Promulgated or Revised ELGs
For the 2009 annual review and development of category rankings, EPA excluded point
source categories for which ELGs were recently established or revised but not yet fully
implemented, or were recently reviewed in a rulemaking context where EPA decided to
withdraw the proposal and select the "no action" option. In general, EPA removes a category
from further consideration during a review cycle if EPA established, revised, or reviewed the
category's ELGs within seven years prior to the annual review. This seven-year period allows
time for the ELGs to be incorporated into NPDES permits. For the 2009 annual review EPA
excluded from the development of category rankings any categories with ELGs established,
revised, or recently reviewed after August 2002. Table 7-3 lists these categories.
Removing a point source category from further consideration in the development of the
rankings does not mean that EPA eliminates the category from annual review. In cases where
EPA is aware of the growth of a new segment within such category, or where new concerns are
identified for previously unevaluated pollutants discharged by facilities in the category, EPA
would apply closer scrutiny to the discharges from the category in deciding whether to consider
it further during the current review cycle. For example, EPA conducted the detailed study of the
coal mining industry based on comments received on the 2006 Preliminary Plan, although the
coal mining ELGs were revised in January 2002.
Table 7-3. Point Source Categories That Have Undergone a Recent Rulemaking or Review
40 CFR Part
Number
122 and 4 12
451
432
413, 433, and 438
420
Point Source Category
Concentrated Animal Feeding Operations (CAFOs)
Concentrated Aquatic Animal Production (or Aquaculture)
Meat and Poultry Products
Metal Products and Machinery
(including Metal Finishing and Electroplating)
Iron and Steel Manufacturing
Date of Rulemaking
November 20, 2008
August 23, 2004
September 8, 2004
May 13, 2003
October 17, 2002
Source: "Guidelines: Final, Proposed, and Under Development" at http://www.epa.gov/waterscience/guide.
7.2.3 Discharges Not Categorizable
EPA identified discharges that are not categorizable into new point source categories or
subcategories. In particular, due to the high TWPE discharges EPA reviewed reported discharges
from a Superfund site (Auchterlonie, 2009).25 Direct discharges from Superfund sites, whether
24 EPA is also currently revising the concentrated animal feeding operations ELG (Part 412); however, the TWPE
associated with this category is low and does not affect the prioritization of categories based on TWPE. For more
information on industries currently undergoing rulemakings, see http://www.epa.gov/guide/industry.html.
25 The Comprehensive Environmental Response, Compensation, and Liability Act, commonly known as Superfund,
was enacted by Congress on December 11, 1980.
7-11
-------
Section 7 - Results of 2009 Screening-Level Analysis
made onsite or offsite, are subject to NPDES permitting requirements (U.S. EPA, 1988a; U.S.
EPA, 1988b). For the reasons discussed below EPA determined that these discharges do not
represent a point source category and excluded these TWPE from the point source category
rankings.
EPA identified that discharges from Superfund sites are too varied to be categorized into
a point source category. In particular, these discharges vary by:
• Contaminants (e.g., metals, pesticides, dioxin);
• Treatment technologies (e.g., air stripping, granular activated carbon,
chemical/ultra-violet oxidation, aerobic biological reactors, chemical
precipitation); and
• Types of facilities causing groundwater contamination (e.g., wood treatment
facilities, metal finishing and electroplating facilities, drum recycling facilities,
mine sites, mineral processing facilities, radium processing facilities).
Moreover, the duration and volume of these direct discharges vary significantly due to
differences in aquifer characteristics and the magnitude, fate, and transport of contaminants in
aquifers and vadose zones. Currently at Superfund sites, permit writers determine technology-
based effluent limits using their best professional judgment (BPJ). EPA selects the remedial
technology and derives numerical effluent discharge limits. The permit must also contain more
stringent effluent limitations when required to comply with state water quality standards. EPA
finds that the current site-specific BPJ approach is workable and flexible within the context of a
Superfund cleanup.
7.2.4 Categories with One Facility Dominating the TWPE
EPA identified point source categories with significant TWPE where only one facility
was responsible for most of the TWPE reported to be discharged (i.e., where one facility's
TWPE accounted for more than 95 percent of the category TWPE, but was not the only facility
reporting discharges for the category). Table 7-4 lists these categories. EPA identified 10
facilities that dominated the TWPE in the category to which they belonged. EPA investigated
these facilities to determine if their discharges were representative of the category. If they were
not, EPA subtracted the facility's TWPE from the total category TWPE and recalculated the
category's ranking. EPA performed this analysis separately for both of the databases. Based on
EPA's knowledge of these industries and the review of the pollutant discharges for these
facilities, EPA determined that all of the pollutant discharges are representative of the industry
and therefore, EPA did not remove the discharges from the category.
7.2.5 Combining the Final DMR and TRI Rankings
After adjusting the category TWPE totals and rankings as described in Sections 7.2.1
through 7.2.4, EPA consolidated the DMRLoads2007 and TRIReleases2007 rankings into one set
using the following steps:
7-12
-------
Section 7 - Results of 2009 Screening-Level Analysis
Table 7-4. Point Source Categories with One Facility Dominating the TWPE Discharges
Point Source Category
Textile Mills (Part 4 10)
Independent and Stand Alone
Labs (Potential New
Category)
Canned and Preserved
Seafood Processing (Part
408)
Plastics Molding and
Forming (Part 463)
Timber Products Processing
(Part 429)
Soap and Detergent
Manufacturing (Part 417)
Ferroalloy Manufacturing
(Part 424)
Construction and
Development (Potential New
Category)
Soap and Detergent
Manufacturing (Part 417)
Tobacco Products (Potential
New Category)
Facility with Over
95% of Category
TWPE
Deroyal Textiles
Brookhaven National
Laboratory
Campbell Soup
Company
Innovia Films, Inc
Stimson Lumber Co
Bonner Mill
Stepan Company -
Elwood
Eramet Marietta Inc
Aeroquip - Vickers
Crodia Inc
Philip Morris Park
500 Site
Facility
Location
Camden,
SC
Upton, NY
Napoleon,
OH
Topeka, KS
Bonner,
MT
Elwood, IL
Marietta,
OH
Joplin, MO
New
Castle, DE
Chester,
VA
Data
Source
DMR
2007
DMR
2007
DMR
2007
DMR
2007
DMR
2007
DMR
2007
DMR
2007
DMR
2007
TRI 2007
TRI 2007
Pollutant Driving
TWPE
Aldrin
PCBs
Hexavalent
Chromium
Carbon Disulfide
Chlorine
Hexachlorobenzene
Cadmium
Cadmium
Bis(2-chloroethyl)
Ether
Chlorine
Facility
TWPE
76,469
5,166
3,123
24,219
51,374
47,795
4,349
324
14,453
4,730
Percent of
Total
Category
TWPE
95.6%
96.5%
96.6%
98.3%
99.7%
99.96%
99.99%
99.99%
99.1%
99.4%
Action
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Did not remove load
from category TWPE
Source: TRIReleases2007 v2\DMRLoads2007 v3.
7-13
-------
Section 7 - Results of 2009 Screening-Level Analysis
• EPA combined the two lists of point source categories by adding each category's
DMRLoads2007 TWPE and TRIReleases2007 TWPE26.
• EPA then ranked the point source categories based on total DMRLoads2007 and
TRIReleases2007 TWPE.
Table 7-5 presents the combined DMRLoads2007 and TRIReleases2007 rankings. These
are the final category rankings accounting for all corrections made to the databases during the
2009 screening-level review and removal of any categories and discharges as discussed in
Sections 7.2.1 through 7.2.4.
7.3 Identification of Categories With Existing Effluent Guidelines for Further Review
After completing the development of the prioritized list, shown in Table 7-5, EPA
selected for further review the point source categories that cumulatively discharge 95 percent of
the total DMRLoads2007 and TRIReleases2007 TWPE. The cutoff point is shown as a bold line
in Table 7-5.
EPA performed detailed studies on three point source categories as part of its 2009
annual review based on the results of its 2007 and 2008 annual reviews. Because EPA data
collection was not finished in 2008, EPA continued detailed studies of the Steam Electric
Generating Category (Part 423), Oil and Gas Extraction (Part 435) (to assess whether to revise
the limits to include coalbed methane extraction as a new subcategory), and the Health Care
Industry (includes Hospitals (Part 460)). EPA did not identify additional categories for detailed
study as part of the 2009 annual review.
EPA's detailed studies generally examine the following: (1) wastewater characteristics
and pollutant sources; (2) the pollutants driving the toxic-weighted pollutant discharges; (3)
availability of pollution prevention and treatment; (4) the geographic distribution of facilities in
the industry; (5) any pollutant discharge trends within the industry; and (6) any relevant
economic factors. First, EPA attempts to verify the screening-level results and fill in data gaps.
Next, EPA considers costs and performance of applicable and demonstrated control technology,
process change, or pollution prevention alternatives that can effectively reduce the pollutants
remaining in the industrial category's wastewater. Last, EPA considers the affordability or
economic achievability of the technology, process change, or pollution prevention measures
identified above.
Types of data sources that EPA may consult in conducting its detailed studies include,
but are not limited to: (1) the U.S. Economic Census; (2) TRI, PCS, and ICIS-NDPES data; (3)
trade associations and reporting facilities to verify reported releases and facility categorization;
(4) regulatory authorities (states and EPA regions) to understand how category facilities are
permitted; (5) NPDES permits and their supporting fact sheets; (6) EPA effluent guidelines
technical development documents; (7) relevant EPA preliminary data summaries or study
reports; and (8) technical literature on pollutant sources and control technologies.
26 EPA notes that this may result in "double-counting" of chemicals a facility reported to both PCS/ICIS-NPDES
and TRI, and "single-counting" of chemicals reported in only one of the databases. The combined databases do not
count chemicals that may be discharged but are not reported to PCS/ICIS-NPDES or TRI.
-------
Section 7 - Results of 2009 Screening-Level Analysis
Table 7-5. Final DMRLoads2007 and TRIReleases2007 Combined Point Source Category Rankings
40CFR
Part
423
430
418
414
419
415
421
440
455
471
410
429
417
444
445
463
439
409
458
436
428
422
464
469
467
437
Point Source Category
Steam Electric Power Generating
Pulp, Paper And Paperboard
Fertilizer Manufacturing
Organic Chemicals, Plastics And Synthetic Fibers
Petroleum Refining
Inorganic Chemicals Manufacturing
Nonferrous Metals Manufacturing
Ore Mining And Dressing
Pesticide Chemicals
Nonferrous Metals Forming And Metal Powders
Textile Mills
Timber Products Processing
Soap And Detergent Manufacturing
Waste Combustors
Landfills
Plastics Molding And Forming
Pharmaceutical Manufacturing
Sugar Processing
Carbon Black Manufacturing
Mineral Mining And Processing
Rubber Manufacturing
Phosphate Manufacturing
Metal Molding And Casting (Foundries)
Electrical And Electronic Components
Aluminum forming
Centralized Waste Treatment
DMRLoads2007
TWPE
20,374,829 a
2,726,865 b
1,095,046
413,226 c
402,506
393,523
342,747
184,455
180,117
119,244
79,934
51,552
47,815
38,412
35,804
24,626
24,937
32,520
26,719
11,195
18,459
11,271
9,350
12,182
10,403
TRIReleases2007
TWPE
541,508
459,959
4,462
574,742
171,756
54,657
38,885
44,437
24,693
8,834
2,389
16,301
14,585
40
83
8,781
7,996
26
32,375
2,416
7,864
250
6,115
7,551
2,707
3,785
Total TWPE
20,916,337
3,186,823 b
1,099,509
987,968 c
574,262
448,181
381,632
228,892
204,810
128,077
82,323
67,852
62,401
38,451
35,887
33,407
32,934
32,545
32,375
29,135
19,059
18,709
17,386
16,902
14,889
14,189
Cumulative Percent
of Total TWPE
72.64
83.71
87.53
90.96
92.96
94.51
95.84
96.63
97.35
97.79
98.08
98.31
98.53
98.66
98.79
98.90
99.02
99.13
99.24
99.34
99.41
99.47
99.54
99.59
99.65
99.69
Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
7-15
-------
Section 7 - Results of 2009 Screening-Level Analysis
Table 7-5. Final DMRLoads2007 and TRIReleases2007 Combined Point Source Category Rankings
40CFR
Part
NA
454
411
425
468
NA
NA
407
424
406
408
434
461
405
443
NA
426
457
465
435
466
446
447
460
NA
Point Source Category
Miscellaneous Foods And Beverages
Gum And Wood Chemicals Manufacturing
Cement Manufacturing
Leather Tanning And Finishing
Copper forming
Independent And Stand Alone Labs
Tobacco Products
Canned And Preserved Fruits And Vegetables
Processing
Ferroalloy Manufacturing
Grain mills
Canned And Preserved Seafood Processing
Coal Mining
Battery Manufacturing
Dairy products processing
Paving And Roofing Materials (Tars And Asphalt)
Printing & Publishing
Glass Manufacturing
Explosives Manufacturing
Coil Coating
Oil & Gas Extraction
Porcelain Enameling
Paint Formulating
Ink Formulating
Hospital
Photo Processing
DMRLoads2007
TWPE
5,842
10,478
8,960
8
2,310
5,355
3
1,757
4,349
1,984
3,232
2,294
1,096
76
1,280
999
353
785
166
256
11
15
1
TRIReleases2007
TWPE
6,576
55
452
7,802
4,951
30
4,756
2,960
340
2,084
234
493
1,642
2,402
249
110
546
14
241
164
140
20
Total TWPE
12,418
10,532
9,412
7,809
7,261
5,385
4,759
4,717
4,689
4,068
3,467
2,787
2,738
2,479
1,529
1,109
899
798
407
256
175
140
20
15
1
Cumulative Percent
of Total TWPE
99.74
99.77
99.81
99.83
99.86
99.88
99.89
99.91
99.93
99.94
99.95
99.96
99.97
99.98
99.99
99.99
99.99
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
Rank
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
7-16
-------
Section 7 - Results of 2009 Screening-Level Analysis
Table 7-5. Final DMRLoads2007 and TRIReleases2007 Combined Point Source Category Rankings
40CFR
Part
459
442
Point Source Category
Photographic
Transportation Equipment Cleaning
Total
DMRLoads2007
TWPE
1
0
26,719,348
TRIReleases2007
TWPE
2,073,457
Total TWPE
1
0
28,792,806
Cumulative Percent
of Total TWPE
100.00
100.00
Rank
52
53
Source: TRIReleases2007_v2; DMRLoads2007_v3.
a EPA corrected a suspected units error in DMRLoads2007_v3 for FB Culley Station in Newburgh, IN (IN0002259) in the Steam Electric Power Generating
Category. EPA attempted to contact the facility but the facility never returned calls. Therefore, EPA was unable to verify the correction.
b For the Pulp, Paper, and Paperboard Category, EPA contacted facilities to verify the concentrations of dioxin and dioxin-like compounds in PCS and ICIS-
NPDES. EPA found that, for all facilities contacted, there were either units errors (e.g., reported as ng/L but in the database as mg/L) or missing non-detect
indicators. The new Pulp, Paper, and Paperboard Category total DMR TWPE is 252,163, while the new DMR and TPJ combined total TWPE is 712,122. See
Section 12.2.2.1 in the Technical Support Document for the Preliminary 2010 Effluent Guidelines Program Plan (U.S. EPA, 2009) for additional details on the
facilities-specific corrections.
dEPA contacted GE Silicones in Friendly, WV (WV0000094), in the OCPSF Category and identified a units error in DMRLoads2007_v3 (Martin, 2009). The
new LEY and TWPE reported for this facility were recalculated and are now 158 and 100.3, respectively. The new OCPSF Category total DMR TWPE is
308,721, while the new DMR and TRI combined total TWPE is 883,463.
NA - Not applicable; no existing ELGs apply to discharges.
7-17
-------
Section 7 - Results of 2009 Screening-Level Analysis
Preliminary category reviews are similar to detailed studies and have the same purpose.
During preliminary reviews, EPA generally examines the same items listed above for detailed
studies. However, EPA's preliminary review of a category and available pollution prevention
and treatment options is less rigorous than its detailed studies. While EPA collects and analyzes
hazard and technology-based information on categories undergoing preliminary review, it
assigns a higher priority to investigating categories undergoing detailed studies.
EPA identified for preliminary review those industrial categories currently regulated by
existing effluent guidelines that cumulatively compose more than 95 percent of the combined
DMRLoads2007 and TRIReleases2007 total TWPE. EPA also reviewed the Ore Mining and
Dressing Category (40 CFR Part 440) because during previous annual reviews, EPA has
concluded that there are not sufficient data available to determine whether wastewater discharges
from the Ore Mining and Dressing Category warrant a detailed study. In addition to the Steam
Electric Power Generating Category this list includes the following point source categories:
• Fertilizer Manufacturing;
• Inorganic Chemicals Manufacturing;
• Nonferrous Metals Manufacturing;
• Ore Mining and Dressing;
• Organic Chemicals, Plastics, and Synthetic Fibers;
• Petroleum Refining; and
• Pulp, Paper and Paperboard.
EPA recently conducted detailed studies or preliminary reviews of many of the categories listed
above. For each of these categories, because EPA's annual review builds on previous reviews,
EPA primarily looked at the pollutants reported in 2007 and their contribution to their category's
TWPE.
After considering the results of the studies and preliminary category reviews, EPA will
determine whether further study or development or revision of an effluent guideline is
appropriate. Final determinations will be presented in the 2010 Effluent Guidelines Plan.
7.4 Results of 2009 Screening-Level Analysis References
1. Davis, Katherine. 2009. Notes from Telephone Conversation between Elizabeth Sabol,
Eastern Research Group, Inc. and Katherine Davis, Westvaco Texas, L.P. RE: Basis of
TCDD Equivalent Concentrations Reported in 2007. (July 7). EPA-HQ-OW-2008-0517
DCN 06547.
2. Martin, Jason. 2009. Notes from Telephone Conversation between Elizabeth Sabol,
Eastern Research Group, Inc. and Jason Martin, MPM Silicones LLC. RE: Basis of
Copper (Total Recoverable) Concentrations Reported in 2007. (July 1). EPA-HQ-OW-
2008-0517 DCN 06549.
3. McCuutchen, Kate. 2009. Notes from Telephone Conversation between Elizabeth Sabol,
Eastern Research Group, Inc. and Kate McCuutchen, Blue Heron Paper Co. RE: Basis of
Methylmercury Concentration Reported in 2007 in DMR. (July). EPA-HQ-OW-2008-
0517 DCN 06546.
-------
Section 7 - Results of 2009 Screening-Level Analysis
4. Verderese, Jim. 2009. Notes from Telephone Conversation between Elizabeth Sabol,
Eastern Research Group, Inc. and Jim Verderese, General Electric Erie. RE: Basis of
Mercury Concentration Reported in December 2007 in DMR. (July 1). EPA-HQ-OW-
2008-0517 DCN 06548.
5. U. S. EPA. 2009. Technical Support Document for the Preliminary 2010 Effluent
Guidelines Program Plan. EPA-821-R-09-006. Washington, DC. (October). EPA-HQ-
OW-2007-0571 DCN 06703.
7-19
------- |