&EPA
  United States
  Environmental Protection
  Agency
DEVELOPMENT DOCUMENT FOR FINAL
EFFLUENT GUIDELINES AND STANDARDS FOR
THE CONSTRUCTION & DEVELOPMENT
CATEGORY

NOVEMBER 2009

-------
U.S. Environmental Protection Agency
       Office of Water (4303T)
   1200 Pennsylvania Avenue, NW
       Washington, DC 20460

-------
Disclaimer

Neither the United States government nor any of its employees, contractors, subcontractors, or
other employees makes any warranty, expressed or implied, or assumes any legal liability or
responsibility for any third party's use of, or the results of such use of, any information,
apparatus, product, or process discussed in this report, or represents that its use by such a third
party would not infringe on privately owned rights. References to proprietary technologies are
not intended to be an endorsement by the U.S. Environmental Protection Agency.

-------
Contents
Volume I
1.  Overview	1-1
   1.1.   Introduction	1-1
   1.2.   Summary and Scope of the Final Rule	1-1
2.  Background	2-1
   2.1.   Legal Authority	2-1
   2.2.   Clean Water Act	2-1
       2.2.1.  Best Practicable Control Technology Currently Available	2-1
       2.2.2.  Best Conventional Pollutant Control Technology	2-2
       2.2.3.  Best Available Technology Economically Achievable	2-2
       2.2.4.  New  Source Performance Standards	2-3
       2.2.5.  Pretreatment Standards for Existing Sources and Pretreatment Standards for
             New  Sources	2-3
       2.2.6.  Effluent Guidelines Schedule and Previous Actions Related to Construction
             and Development	2-3
       2.2.7.  NPDES Phase I  and II Stormwater Rules	2-4
   2.3.   Pollution Prevent!on Act of 1990	2-6
   2.4.   State Regulations	2-6
3.  Data Collection	3-1
   3.1.   Introduction	3-1
   3.2.   Literature Search	3-1
   3.3.   Data and Information Provided in Response to the 2002 Regulatory Action and
          the 2008  Proposal	3-1
   3.4.   Compilation of State Control Strategies, Criteria, and Standards	3-1
   3.5.   Other Data Sources	3-2
       3.5.1.  Land Use Data	3-2
       3.5.2.  NPDES Permit Notice of Intent Data	3-11
       3.5.3.  Climatic/Rainfall Data	3-11
       3.5.4.  Soils Data	3-17
       3.5.5.  Vendor Data for Active Treatment Systems	3-17
       3.5.6.  Rainfall and Runoff Erosivity Factor	3-18
       3.5.7.  Hydrologic Soil  Groups	3-18
   3.6.   References	3-21
4.  Industry Profile	4-1
   4.1.   Introduction	4-1
   4.2.   Industry Practices and Trends	4-6
       4.2.1.  Overview of Construction Land-Disturbing Activities	4-6
       4.2.2.  Construction Site Size Categories and Estimates of Amount of Disturbed
             Land	4-8
   4.3.   References	4-10
5.  Selection of Pollutants for Regulation	5-1
   5.1.   Introduction	5-1
                                           IV

-------
   5.2.    Pollutants Associated with Construction and Land Development Stormwater
          Runoff	5-1
       5.2.1.  Sediment	5-1
       5.2.2.  Metals	5-3
       5.2.3.  PAHs, and Oil and Grease	5-4
       5.2.4.  Pathogens	5-4
   5.3.    Selection of Pollutants for Regulation	5-4
   5.4.    References	5-5
6.  Limitations and Standards: Data Selection and Calculation	6-1
   6.1.    Introduction	6-1
   6.2.    Turbidity	6-1
   6.3.    Overview of Data Review and Criteria	6-2
   6.4.    Data Selected as Basis for Limitations	6-3
       6.4.1.  Cascade EcoSolutions	6-4
       6.4.2.  Clear Water Compliance Services, Inc	6-5
       6.4.3.  Research  by North Carolina State University	6-5
   6.5.    Systems Excluded as Basis for Limitation	6-5
   6.6.    Application of Criteria to Limitation Data Sets	6-6
   6.7.    Summary of Limitation Data and Data Conventions	6-9
   6.8.    Data Averaging Prior to Limitation Calculations	6-10
   6.9.    Limitation Calculations	6-11
       6.9.1.  Statistical Percentile Basis for Limitations	6-11
       6.9.2.  Long-Term Average	6-12
       6.9.3.  Variability Factor	6-13
       6.9.4.  Calculation of the Limitation	6-15
       6.9.5.  Limitation Includes Autocorrelation Adjustment	6-15
   6.10.  Statistical and Engineering Review of Limitation	6-17
       6.10.1. Performance Data for Model Technology Compared to Limitation	6-17
       6.10.2. Performance of Other Treatment Systems Relative to Model Technology	6-19
       6.10.3. Performance of Model Technology for Gold Placer Mining Wastes	6-21
       6.10.4. Limitation is Consistent with State Action Levels	6-22
   6.11.  Monitoring Considerations	6-22
   6.12.  Compliance	6-23
   6.13.  Summary of Steps Used to Derive the Limitations	6-25
   6.14.  References	6-26
7.  Technology Assessment	7-1
   7.1.    Review of Historical Approaches to Erosion and Sediment Control (ESC)	7-1
   7.2.    Control Techniques	7-3
       7.2.1.  Erosion Control and Prevention	7-3
       7.2.2.  Water Handling Practices	7-26
       7.2.3.  Sediment-Trapping Devices	7-44
       7.2.4.  Other Control Practices	7-68
       7.2.5.  Advanced Treatment and Control Technologies	7-87
   7.3.    References	7-97
8.  BCT Cost-Reasonableness Assessment	8-1
   8.1.    Background on the BCT Cost Test	8-1

-------
   8.2.    Options Evaluated for BCT	8-2
       8.2.1.  Option 1	8-2
       8.2.2.  Option 2	8-2
       8.2.3.  Options	8-2
       8.2.4.  Option 4	8-2
   8.3.    Calculation of the BCT Cost Test	8-2
   8.4.    References	8-4
9.  Estimating Incremental Costs for the Final Regulation	9-1
   9.1.    Overview	9-1
   9.2.    Development of Model Construction Sites and Estimating Treatment Volumes	9-2
       9.2.1.  Model Construction Sites	9-2
       9.2.2.  Estimation of Rainfall Depths and Storage Volumes	9-3
       9.2.3.  Estimation of ATS Treatment Volumes	9-4
   9.3.    Estimation of Costs	9-19
       9.3.1.  Erosion and Sediment Control Costs	9-19
       9.3.2.  Passive Treatment Costs	9-25
       9.3.3.  ATS Costs	9-33
   9.4.    References	9-39
10. Estimating Pollutant Load Reductions	10-1
   10.1.  Overview of Approach	10-1
   10.2.  Model Project Analysis	10-1
   10.3.  Model Parameter and Loads Estimation	10-2
       10.3.l.LS Factor	10-6
       10.3.2. P Factor	10-7
       10.3.3. C Factor	10-8
       10.3.4. Runoff Volume Estimates	10-10
   10.4.  Load Estimation	10-12
   10.5.  Results	10-13
   10.6.  References	10-21
11. Non-Water Quality Environmental Impacts	11-1
   11.1.  Energy Requirements	11-1
       11.1.1. Energy Requirements Attributable to the Regulatory Options	11-1
       11.1.2. Treatment Chemical Production	11-3
       11.1.3. Comparison of Option Energy Requirements to Construction Industry	11-4
   11.2.  Air Emissions Impacts	11-5
   11.3.  Solid Waste Generation	11-6
   11.4.  References	11-6

Volume ll-a
Appendix A: Summary of State Construction and Development Requirements	A-1
Appendix B: Literature Search Annotated Bibliography	B-l
Appendix C: Analysis of Construction Industry Trends Using Notice of Intent Records	C-l
Appendix D: Precipitation Data Representative of Major U.S. Metropolitan Areas	D-l
Appendix E: Determination of Development Rates in U.S. Watersheds	E-l
                                          VI

-------
Volume ll-b
Appendix F: Turbidity Report Tables	F-l
Appendix G: Lognormal Distribution Used for Site-Specific, Long-Term Averages and
   Variability Factors	G-l
Appendix H: Regional Rainfall/Runoff Information Representative of Major U.S.
   Metropolitan Areas	H-l
Appendix I: Model Project Costs for Regulatory Options	1-1


Figures
Figure 3-1. NLCD 1992/2001 land cover change product near Seattle, Washington	3-3
Figure 3-2. EPA Region 1: Percent urban change 1992-2001 by ERF 1_2 watershed	3-5
Figure 3-3. EPA Region 2: Percent urban change 1992-2001 by ERF1_2 watershed	3-6
Figure 3-4. EPA Region 3: Percent urban change 1992-2001 by ERF1_2 watershed	3-6
Figure 3-5. EPA Region 4: Percent urban change 1992-2001 by ERF1_2 watershed	3-7
Figure 3-6. EPA Region 5: Percent urban change 1992-2001 by ERF1_2 watershed	3-7
Figure 3-7. EPA Region 6: Percent urban change 1992-2001 by ERF1_2 watershed	3-8
Figure 3-8. EPA Region 7: Percent urban change 1992-2001 by ERF1_2 Watershed	3-8
Figure 3-9. EPA Region 8: Percent urban change 1992-2001 by ERF1_2 watershed	3-9
Figure 3-10. EPA Region 9: Percent urban change  1992-2001 by ERF1_2 watershed	3-9
Figure 3-11. EPA Region 10: Percent urban change 1992-2001 by ERF 1_2 watershed	3-10
Figure 3-12. Average annual precipitation in the CONUS from PRISM	3-16
Figure 3-13. An example of Washington State RF1 watershed showing urban weighting
      emphasizes MUID: WAI89, while area-weighting would have enhanced MUID
      WA194	3-17
Figure 3-14. Annual R factor values for CONUS	3-19
Figure 7-1. General ATS batch-operating mode	7-88
Figure 7-2. Flow-through ATS operating mode	7-89


Tables
Table 3-1. State and national estimates of urban land from NLCD	3-4
Table 3-2. EPA Region indicator cities	3-12
Table 3-3. Rainfall summary data for indicator cities	3-14
Table 3-4. HSGs by state	3-20
Table 4-1. 2007 NAICS subsectors, industry groups, and industries performing
      construction activities that might disturb land	4-3
Table 4-2. 1987 SIC industry groups performing construction activities that might disturb
      land	4-5
Table 4-3. Model project distribution	4-10
Table 5-1. Studies of uncontrolled soil erosion as TSS from construction sites	5-3
Table 6-1. Data sources and site identification for systems using EPA's model technology
      basis	6-4
Table 6-2. Systems Excluded as Basis for Limitation	6-6
                                         VII

-------
Table 6-3. Summary of reported turbidity measurements (NTU) in effluent (individual
       measurements before daily average calculations)	6-9
Table 6-4. Summary of daily values of turbidity (NTU) in effluent	6-11
Table 6-5. System-specific long-term averages used in limitation calculations	6-12
Table 6-6. System-specific variability factors used in limitation calculations	6-13
Table 6-7. TSS variability factors in recent regulations	6-14
Table 6-8. Effect of autocorrelation adjustments on limitation	6-16
Table 6-9. Daily values greater than daily maximum limitation	6-19
Table 6-10. Daily value summary from systems with less than model technology	6-20
Table 6-11. Turbidity during 1986 Alaskan placer mining study	6-21
Table 7-1. Scheduling considerations for construction activities	7-5
Table 7-2. Conditions where vegetative streambank stabilization is acceptable	7-8
Table 7-3. Maximum permissible velocities for individual site conditions for grass swales.... 7-10
Table 7-4. Typical mulching materials and application rates	7-17
Table 7-5. Measured reductions in soil loss for different mulch treatments	7-18
Table 7-6. Cubic yards of topsoil required for application to various depths	7-25
Table 7-7. Grassed swale pollutant-removal efficiency data	7-32
Table 7-8. Average annual operation and maintenance costs for a grass swale	7-34
Table 7-9. Recommended pipe/tubing sizes for slope drains	7-36
Table 7-10. Slope drain characteristics	7-37
Table 7-11. Maximum slope lengths  for silt fences	7-46
Table 7-12. Typical requirements for silt fence fabric	7-48
Table 7-13. Slope lengths for super silt fences	7-52
Table 7-14. Minimum requirements for super silt fence Geotextile Class F fabric	7-53
Table 7-15. Maximum land slope and distances above a straw bale dike	7-54
Table 7-16. Weir length for sediment traps	7-57
Table 7-17. Range of measured pollutant removal for sediment  detention basins	7-58
Table 7-18. Common concerns associated with sediment traps	7-59
Table 7-19. Studies of TSS in sediment basin effectiveness and effluent from construction
       sites	7-64
Table 7-20. Common concerns associated with sediment basins	7-65
Table 7-21. Riprap sizes and thicknesses	7-71
Table 7-22. Application rates for spray-on adhesives	7-82
Table 7-23. Turbidity reduction values from PAM	7-86
Table 7-24. ATS operating modes	7-88
Table 7-25. ATS state requirements/recommendations	7-90
Table 7-26. Summary of vendor costs	7-92
Table 7-27. Summary of ATS case studies	7-92
Table 7-28. Examples of some commonly available coagulants	7-93
Table 7-29. Chitosan acetate study results	7-94
Table 7-30. DADMAC acetate study results	7-95
Table 7-31. PAM study results	7-95
Table 7-32. Aluminum-based coagulant study results	7-96
Table 8-l.POTW cost test results	8-3
Table 8-2. Cost and pollutant removals for BPT	8-4
Table 8-3. Industry cost-effectiveness test results	8-4
                                           VIM

-------
Table 9-1. Estimated total annual social costs of regulatory options for the C&D industry	9-1
Table 9-4. Model project durations	9-2
Table 9-2. Model project matrix	9-5
Table 9-3. Acreage developed matrix	9-7
Table 9-5. State runoff coefficients for 2-year, 24-hour storm events	9-9
Table 9-6. ATS storage requirements for states	9-11
Table 9-7. Monthly ATS treatment volumes (gallons)	9-13
Table 9-8. ATS system flowrate required (gpm)	9-15
Table 9-9. ATS system flowrate selected for costing (gpm)	9-17
Table 9-10. Sediment basin construction cost data	9-19
Table 9-11. Baseline sediment basin size (cubic feet)	9-20
Table 9-12. Baseline sediment basins costs	9-21
Table 9-13. Temporary cover costs	9-23
Table 9-14. Baseline silt fence and temporary cover assumptions and costs	9-23
Table 9-15. Basin assumptions	9-23
Table 9-16. Surface outlet cost assumptions	9-24
Table 9-17. Costs for surface outlets	9-24
Table 9-18. Passive treatment unit costs: polymer dosing system and labor	9-26
Table 9-19. Passive treatment unit costs: filter berms	9-26
Table 9-20. Monthly treatment volumes for passive treatment (gallons)	9-27
Table 9-21. Monthly chemical cost for passive treatment	9-29
Table 9-22. Total monthly cost for passive treatment	9-31
Table 9-23. ATS costs: 500-gpm system	9-34
Table 9-24. One-time ATS costs model projects	9-35
Table 9-25. Monthly ATS costs for model projects	9-37
Table 10-1. Model project matrix	10-3
Table 10-2. Project matrix for loads analysis	10-5
Table 10-3. Data sources used to obtain RUSLE factors	10-5
Table 10-4. Slope ranges from STATSGO  (percent)	10-6
Table 10-5. P factors for construction site practices	10-8
Table 10-6. P factors used for load estimation	10-8
Table 10-7. C factors for construction site controls	10-9
Table 10-8. C factors used for loads estimation	10-10
Table 10-9. Allocation of states/commonwealths/territories to representative indicator city . 10-10
Table 10-10 Estimated runoff coefficients by HSG for indicator regions	10-12
Table 10-11. Discharged loads—primary analysis case	10-14
Table 10-12. Discharged loads—low sensitivity analysis case	10-15
Table 10-13. Discharged loads—high sensitivity analysis case	10-16
Table 10-14. Sediment removals—primary analysis case	10-17
Table 10-15. Sediment removals—low sensitivity analysis case	10-18
Table 10-16. Sediment removals—high sensitivity analysis case	10-19
Table 10-17. National sediment reductions for regulatory options	10-20
Table 10-18. Total discharge loads and loads modeled in SPARROW	10-21
Table 11-1. Estimated energy consumption by regulatory option	11-2
Table 11-2. Maximum chitosan acetate required under EPA options	11-3
Table 11-3. U.S. acrylamide/PAMs demand	11-4
                                           IX

-------
Table 11-4. 2002 Energy use in NAICS Category 23	11-4
Table 11-5. Estimated incremental energy usage by regulatory option	11-4
Table 11-6. Estimated incremental air emissions by regulatory option (pounds/year)	11-5

-------
Acronyms and Abbreviations
ASCE
ATS
AWWA
BAER
BAT
BCT
BMP
BOD
BPJ
BPT
C&D
CAA
CCI
CFR
CGP
CHIA
CONUS
CSTR
CTAPE
CUD
CWA
DADMAC
DCN
ECRMs
ELG
ENR
EPA
ERF1
ERF1_2
ESC
GIS
gpm
GULD
HDSC
HSG
LEW
MRLC
MS4
MUID
NAICS
American Society of Civil Engineers
Active Treatment System
American Water Works Association
Burned Area Emergency Response
Best Available Technology Economically Achievable
Best Conventional Pollutant Control Technology
best management practice
biochemical oxygen demand
best professional judgment
Best Practicable Control Technology
Construction and Development
Clean Air Act
Construction Cost Index
Code of Federal Regulations
Construction General Permit
Cumulative Hydrologic Impact Analysis
conterminous United States
continuously stirred tank reactor
Chemical Technology Assessment Protocol—Ecology
conditional use designation
Clean Water Act
Diallydimethyl-ammonium chloride
Document Control Number
erosion control and revegetation mats
Effluent Limitations Guidelines
Engineering News-Record
U.S. Environmental Protection Agency
Enhanced River Reach File 1.2
Enhanced River Reach File 2.0
erosion and sediment control
geographic information system
gallons per minute
general use level designation
Hydrometeorological Design Studies Center
hydrologic soil group
Low Erosivity Waiver
Multi-Resolution Land Characteristics Consortium
municipal separate storm sewer system
Map Unit Identifiers
North American Industry Classification System
                                         XI

-------
NEL
NLCD
NOAA
NOT
NOT
NOx
NPDES
NRDC
NSPS
NTU
NWS
OCPSF
PAH
PAM
PFDS
POTW
PPA
PRISM
RCRA
RF1
RUSLE
SC AQMD
scs
SIC
SIP
SMRCA
SPARROW
ssc
STATSGO
SWPPP
SWRPC
TP
TRM
TSS
TSS
USDA
USGS
USLE
VOCs
WDEC
WSDOT
numeric effluent limit
National Land Cover Database
National Oceanic and Atmospheric Administration
Notice of Intent
Notice of Termination
nitrogen oxides
National Pollutant Discharge Elimination System
Natural Resources Defense Council
New Source Performance Standards
nephelometric turbidity units
National Weather Service
organic chemicals, plastics, and synthetic fibers
polycyclic aromatic hydrocarbon
polyacrylamide
Precipitation Frequency Data Server
publicly owned treatment works
Pollution Prevention Act of 1990
Parameter Elevation Regressions on Independent Slopes Model
Resource Conservation and Recovery Act
Reach File Version  1.0
Revised Universal Soil Loss Equation
South Coast Air Quality Management District
Soil Conservation Service
Standard Industrial Classification
state implementation plan
Surface Mining, Reclamation, and Control Act
Spatially Referenced Regressions on Watersheds
suspended sediment concentration
State Soil Geographic Database
stormwater pollution prevention plan
Southeastern Wisconsin Regional Planning Commission
total phosphorus
turf reinforcement mats
total suspended solids
total suspended solids
U.S. Department of Agriculture
U.S. Geological Survey
Universal Soil Loss Equation
volatile organic compounds
Washington Department of Ecology
Washington Department of Transportation
                                         XII

-------
                                                                        Section 1: Overview
1.   OVERVIEW

 1.1.    INTRODUCTION

This document presents technical information to support the U.S. Environmental Protection
Agency's (EPA's) decision and complements the Agency''s Economic Analysis for Final Effluent
Guidelines and Standards for the Construction and Development Category (EPA-821-R-09-
011), and the Environmental Impact and Benefits Assessment for Final Effluent Guidelines and
Standards for the Construction and Development Category (EPA-821-R-09-0012).

A summary of the information contained in the sections of this document is as follows:

       •  Section 2 presents a summary of the legal authority for effluent guidelines and the
          existing EPA storm water program.

       •  Section 3 summarizes the data collection activities and the analytical tools and
          processes followed to support the  final action.

       •  Section 4 summarizes the characteristics of the construction and development
          industry, including major indicators of industry size and annual construction activity.

       •  Section 5 presents a description of pollutants in stormwater runoff known to be the
          most prevalent and of greatest concern to the environment. It also presents the
          selection of pollutants for the final regulation.

       •  Section 6 presents the method  and data used to establish limitations and standards.

       •  Section 7 presents information and data on erosion and sediment control (ESC)
          strategies used by the construction and development industry, including applicability,
          costs, and efficiencies of various technologies.
       •  Section 8 presents a description of the regulatory options EPA considered when
          developing the final rule.

       •  Section 9 presents a description of the approach EPA used in developing costs for the
          regulatory options.

       •  Section 10 summarizes the approach EPA used to estimate the pollutant loads and
          load reductions for the regulatory  options EPA considered.

       •  Section 11 summarizes the non-water quality environmental impacts, including the
          energy requirements, air emissions impacts, and solid waste generation of each
          regulatory option.

 1.2.    SUMMARY AND SCOPE OF THE FINAL RULE

EPA has established effluent limitations guidelines and new source  performance standards for
stormwater discharges from the construction  and development industry.  The guidelines and
standards require discharges from certain  construction sites to meet a numeric turbidity limit.
The guidelines and standards also require all  construction sites that  are now required to obtain a
National Pollutant Discharge Elimination System permit to implement a variety of best
                                          1-1

-------
                                                                         Section 1: Overview
management practices designed to limit erosion and control sediment discharges from
construction sites. EPA evaluated four options in developing the final rule. Those options are
described below:

       •  Option 1 establishes minimum requirements for implementing a variety of ESCs and
          pollution prevention measures on all construction sites that are required to obtain a
          permit.
       •  Option 2 contains the same requirements as Option 1. In addition, construction sites
          of 30 or more disturbed acres would be required to meet a numeric turbidity limit in
          stormwater discharges from the site. The technology basis for the numeric limit is
          active treatment systems (ATS). The numeric turbidity standard would be applicable
          to stormwater discharges for all storm events up to the local 2-year, 24-hour event.

       •  Option 3 contains the same requirements as Option 1. Option 3 also requires all sites
          with 10 or more acres of disturbed land to meet a numeric turbidity standard that is
          based on the application of ATS. The turbidity standard would apply to all
          stormwater discharges for all storm events up to the local 2-year, 24-hour event.

       •  Option 4 contains the same requirements as Option 1. Option 4 also requires all sites
          with 10 or more acres of disturbed land to meet a numeric turbidity standard that is
          based on the application of passive treatment systems. The turbidity standard would
          apply to all stormwater discharges for all storm events up to the local 2-year, 24-hour
          event; although, only certain types of discharges would require monitoring.

The costs and economic impacts of the Options are presented in the Preamble to the Final Rule
and in the Economic Analysis.
                                           1-2

-------
                                                                     Section 2: Background
2.   BACKGROUND

 2.1.     LEGAL AUTHORITY

The U.S. Environmental Protection Agency (EPA) is promulgating Effluent Limitations
Guidelines (ELGs) for discharges associated with construction and development activities under
the authority of the Clean Water Act (CWA) sections 301, 304, 306, 308, 402, and 501 (the
Federal Water Pollution Control Act), Title 33 of the United States Code (U.S.C.) sections 1311,
1314,  1316, 1318, 1342, and 1361. This Background section describes EPA's legal authority for
issuing the regulation, existing state regulations, and other federal regulations associated with
construction and development activities.

 2.2.     CLEAN WATER ACT

Congress adopted the CWA to "restore and maintain the chemical, physical, and biological
integrity of the nation's waters" (section 101(a), 33 U.S.C. 1251(a)). To achieve this goal, the
CWA prohibits the discharge of pollutants into navigable waters except in compliance with the
statute. CWA section 402 requires point source discharges to obtain a permit under the National
Pollutant Discharge Elimination System (NPDES). Those permits are issued by EPA regional
offices or authorized state agencies.

Following enactment of the Federal Water Pollution Control Amendments of 1972 (Pub.L. 92-
500, October  18, 1972), EPA and the states issued NPDES permits to thousands of dischargers,
both industrial (e.g., manufacturing, energy and mining facilities)  and municipal (sewage
treatment plants). As required under Title III of the Act, EPA promulgated ELGs and standards
for many industrial categories, and those requirements are incorporated into the permits.

The Water Quality Act of 1987 (Pub.L. 100-4, February 4, 1987) amended the CWA. The
NPDES program was expanded by defining municipal and industrial stormwater discharges as
point sources. Industrial stormwater dischargers,  municipal separate storm sewer systems (MS4s)
and other stormwater dischargers designated by EPA must obtain NPDES permits pursuant to
section 402(p) (33 U.S.C. 1342(p)).

   2.2.1.    BEST PRACTICABLE CONTROL TECHNOLOGY CURRENTLY
           AVAILABLE

In guidelines for a point source category, EPA may define Best Practicable Control Technology
(BPT) effluent limits for conventional, toxic, and nonconventional pollutants. In specifying BPT,
EPA looks at  a number of factors. EPA first considers the cost of achieving effluent reductions in
relation to the effluent reduction benefits. The Agency also considers the age of the equipment
and facilities, the processes employed and any required process changes, engineering aspects of
the control technologies, non-water quality environmental impacts (including energy
requirements), and such other factors as the Agency  deems appropriate (CWA section
304(b)(l)(B)). Traditionally, EPA establishes BPT effluent limitations on the basis of the
average of the best performance of facilities within the category of various ages, sizes, processes
or other common characteristics.  Where existing  performance is uniformly inadequate, EPA may
                                          2-1

-------
                                                                      Section 2: Background
require higher levels of control than currently in place in a category if the Agency determines
that the technology can be practically applied. See A Legislative History of the Federal Water
Pollution Control Act Amendments of'1972, U.S. Senate Committee of Public Works, Serial No.
93-1, January 1973, p. 1468.

In addition, the Act requires a cost-reasonableness assessment for BPT limitations. In
determining the BPT limits, EPA considers the total cost of treatment technologies in relation to
the effluent reduction benefits achieved. This inquiry does not limit EPA's broad discretion to
adopt BPT limitations that are achievable with available technology unless the required
additional reductions are "wholly out of proportion to the costs of achieving such marginal level
of reduction." See Legislative History, op.  cit, p.  170. Moreover, the inquiry does not require the
Agency to quantify benefits in monetary terms. See, for example, American Iron and Steel
Institute v. EPA, 526 F. 2d 1027 (3rd Cir., 1975).

In balancing costs against the benefits of effluent reduction, EPA considers the volume and
nature of expected discharges after application of BPT, the general environmental effects of
pollutants, and the cost and economic impacts of the required level of pollution control. In past
ELGs and standards, BPT cost-reasonableness removal figures have ranged from $0.21 to $33.71
per pound removed in year 2000 dollars. In developing guidelines, the Act does  not require
consideration of water quality problems attributable to particular point sources, or water quality
improvements in particular bodies of water. See Weyerhaeuser Company v. Costle,  590 F. 2d
1011 (D.C. Cir.  1978).

   2.2.2.   BEST CONVENTIONAL POLLUTANT CONTROL TECHNOLOGY

The 1977 amendments to the CWA require EPA to identify effluent reduction levels for
conventional pollutants associated with Best Conventional Pollutant Control Technology (BCT)
for discharges from existing point sources.  BCT is not an additional limitation but replaces Best
Available Technology Economically Achievable (BAT) for control of conventional pollutants. In
addition to other factors specified in section 304(b)(4)(B), the CWA requires that EPA establish
BCT limitations after consideration of a two-part cost-reasonableness test. EPA explained its
methodology for developing BCT limitations in July  1986 (51 Federal Register  [FR] 24974).

Section 304(a)(4) designates the following as conventional pollutants: biochemical oxygen
demand (BOD5), total suspended solids (TSS), fecal coliform, pH, and any additional pollutants
defined by the Administrator as conventional. The Administrator designated oil  and grease as an
additional conventional pollutant on July 30, 1979 (44 FR 44501). A primary pollutant of
concern at construction sites, sediment,  is commonly measured as TSS.

   2.2.3.   BEST AVAILABLE TECHNOLOGY ECONOMICALLY ACHIEVABLE

In general, BAT effluent guidelines (CWA section 304(b)(2)) represent the best  existing
economically achievable performance of direct discharging plants in the subcategory or category.
The factors considered in assessing BAT include the cost of achieving BAT effluent reductions,
the age of equipment and facilities involved, the processes employed, engineering aspects of the
control technology, potential process changes, non-water quality environmental  impacts
(including energy requirements), and such factors as the Administrator deems appropriate. The
                                          2-2

-------
                                                                    Section 2: Background
Agency retains considerable discretion in assigning the weight to be accorded to these factors.
An additional statutory factor considered in setting BAT is economic achievability. Generally,
EPA determines the economic achievability on the basis of the total cost to the subcategory and
the overall effect of the rule on the industry's financial health. The Agency may base BAT
limitations on effluent reductions attainable through changes in a facility's processes and
operations. As with BPT, where existing performance is uniformly inadequate, EPA may base
BAT on technology transferred from a different subcategory or from another category. In
addition, the Agency may base BAT on manufacturing process changes or internal controls, even
when such technologies are not common industry practice.

   2.2.4.    NEW SOURCE PERFORMANCE STANDARDS

New Source Performance Standards (NSPS) reflect effluent reductions that are achievable on the
basis of the best available demonstrated control technology.  New facilities have the opportunity
to install the best and most efficient production processes and wastewater treatment technologies.
As a result, NSPS  should represent the greatest degree of effluent reduction attainable through
the application of the best available demonstrated control technology for all pollutants (i.e.,
conventional, nonconventional, and priority pollutants). In establishing NSPS, CWA section 306
directs EPA to take into consideration the cost of achieving the effluent reduction and any non-
water quality environmental impacts and energy  requirements.

   2.2.5.    PRETREATMENT STANDARDS FOR EXISTING SOURCES AND
            PRETREATMENT STANDARDS FOR NEW SOURCES

The CWA also defines standards for indirect discharges, i.e., discharges into publicly owned
treatment works. Such standards are Pretreatment Standards for Existing Sources and
Pretreatment Standards for New Sources under section 307(b).

   2.2.6.    EFFLUENT GUIDELINES SCHEDULE AND PREVIOUS ACTIONS
            RELATED TO CONSTRUCTION AND DEVELOPMENT

CWA section 304(m) requires EPA to publish a plan every 2 years that  consists of three
elements. First, under section 304(m)(l)(A), EPA is required to establish a schedule for the
annual review and revision of existing effluent guidelines in accordance with section 304(b).
Section 304(b) applies to ELGs for direct dischargers and requires EPA to revise such
regulations as appropriate. Second, under section 304(m)(l)(B), EPA must identify categories of
sources discharging toxic or nonconventional pollutants for which EPA has not published BAT
ELGs under section 304(b)(2) or NSPS under section 306. Finally, under section 304(m)(l)(C),
EPA must establish a schedule for  the promulgation of BAT and NSPS  for the categories
identified under subparagraph (B) not later than 3 years after being identified in the 304(m) plan.
Section 304(m) does not apply to pretreatment standards for indirect dischargers, which EPA
promulgates pursuant to CWA sections 307(b) and 307(c).

On October 30, 1989, Natural Resources Defense Council, Inc. (NRDC), and Public Citizen,
Inc., filed an action against EPA in which they alleged, among other things, that EPA had failed
to comply with section 304(m). Plaintiffs and EPA agreed to a settlement of that action in a
consent decree entered on January  31, 1992 (Natural Resources Defense Council et al. v
                                         2-3

-------
                                                                      Section 2: Background
Whitman, D.D.C. Civil Action No. 89-2980). The consent decree, which has been modified
several times, established a schedule by which EPA is to propose and take final action for 11
point source categories identified by name in the decree and for 8 other point source categories
identified only as new or revised rules, numbered 5 through 12. EPA selected the Construction
and Development (C&D) category as the subject for new or revised rule #10. The decree, as
modified, calls for the Administrator to sign a proposed ELG for the C&D category no later than
May 15, 2002, and to take final action on that proposal no later than March 31, 2004. A
settlement agreement between the parties, signed on June 28, 2000, requires that EPA develop
regulatory options applicable to discharges from construction, development,  and redevelopment
covering site sizes included in the Phase I and Phase IINPDES stormwater rules  (i.e., 1 acre or
greater). EPA is required to develop options including numeric effluent limitations for
sedimentation and turbidity; control of construction site pollutants other than sedimentation and
turbidity (e.g., discarded building materials, concrete truck washout, trash); best management
practices (BMPs) for controlling post-construction runoff; BMPs for construction sites; and
requirements to design stormwater controls to maintain predevelopment runoff conditions where
practicable.

On June 24, 2002, EPA published a proposed rule for the C&D category that contained several
options for the control of stormwater discharges from construction sites, including ELGs and
NSPS. (67 FR 42644; June 24, 2002). In a final action published on April 26, 2004, EPA
determined that national ELGs would not be the most effective way to control discharges from
construction sites and, instead, chose to rely on the range of existing programs, regulations, and
initiatives that already existed at the federal, state, and local level. (69 FR 22472).

The June 28, 2000, settlement agreement also required EPA to issue guidance to MS4s and other
permittees on maintenance of post-construction BMPs identified in the proposed  ELGs. Because
EPA's proposal or final action does not contain requirements for post-construction BMPs, that
guidance was considered no longer necessary and, therefore, was not fully developed. However,
a draft of the maintenance guidance that was prepared while EPA was considering including
options for post-construction BMPs is in the public docket for the previous rulemaking.

On October 6, 2004, NRDC and Waterkeeper Alliance, as well as New York and Connecticut
filed a motion against EPA alleging that EPA failed to promulgate ELGs and NSPS as required
by the CWA. On December 1, 2006, the district court—in Natural Resources Defense Council,
et al. v U.S. Environmental Protection Agency, et al, 437 F.Supp.2d 1137, 1139 (C.D.
Cal.2006)—held that CWA section 304(m), read together with CWA section 304(b), imposes on
EPA a mandatory duty to promulgate ELGs and NSPS for industrial point source categories
named in a CWA section 304(m) plan. The court ordered EPA to publish proposed regulations in
the FR by December 1, 2008, and to promulgate ELGs and NSPS for the C&D category as soon
as practicable, but no later than December 1, 2009.

   2.2.7.    NPDES PHASE I AND II STORMWATER RULES

As authorized by the CWA, the NPDES permit program was established to control water
pollution by regulating point sources that discharge pollutants into waters of the United States.
Stormwater runoff from construction activities can have a significant effect on water quality. The
NPDES stormwater program requires operators of construction sites to apply for  either a general
                                          2-4

-------
                                                                       Section 2: Background
permit or an individual permit under the NPDES Phase I and II stormwater rules. Phase I of
EPA's stormwater program was promulgated in 1990 under the CWA and addresses, among
other discharges, discharges from construction activities disturbing 5 acres or more of land.
Phase II of the NPDES stormwater program, promulgated in 1999, expands the Phase I Rule by
addressing stormwater discharges from small construction sites disturbing between 1 and 5 acres.
In addition, operators of small construction sites are also required to develop and implement a
stormwater pollution prevention plan (SWPPP), which includes implementing the appropriate
ESC BMPs. The BMP selection and design are at the discretion of permittees (in conformance
with applicable state or local requirements). Moreover, construction activities disturbing less
than 1 acre are also included in Phase II of the NPDES stormwater program if they are part of a
larger, common plan of development or sale with a planned disturbance of equal to or greater
than 1 acre and less than 5 acres, or if they are designated by the NPDES permitting authority.

Most states are authorized to implement the stormwater NPDES permitting program. However,
EPA remains the permitting authority in a few states, territories, and on most Indian country
lands. For construction (and other land disturbing activities) in areas where EPA is the permitting
authority,  operators must meet the requirements of the EPA Construction General Permit (CGP).

The current CGP became  effective on June 30, 2008 (as modified effective September 29, 2008)
and expires on June  30, 2010.  That permit contains substantially the same terms and conditions
as the 2003 CGP.  In response to comments on the proposal, EPA has reorganized the content of
the 2003 permit to better clarify existing requirements. The 2008 CGP applies only to new
discharges. Construction site operators with permit coverage under the 2003 CGP may continue
to operate under the terms of conditions of that permit and need not file a new Notice of Intent
for coverage under the 2008 CGP. Permit coverage is now available for eligible construction
activities in New Hampshire, Oklahoma, Texas, Puerto Rico, federal facilities, and Indian
country lands in Colorado and Montana.

The 2003 permit expanded coverage from the 1998 CGP, which provided coverage for large
construction sites  (i.e., those disturbing greater than 5 acres) to include both small and  large
construction activities (i.e., any project disturbing greater than 1 acre). Small construction
activity was added to the 2003 CGP in response to the promulgation of the NPDES Phase II
Rule.

A major provision required by the CGP is preparation of a SWPPP. The SWPPP focuses on two
major requirements: (1) Providing a site description that identifies sources of pollution to
stormwater discharges associated with industrial activity on-site; and (2) identifying and
implementing appropriate measures to reduce pollutants in stormwater discharges to ensure
compliance with the terms and conditions of the permit. All SWPPPs must be developed in
accordance with sound engineering practices and must be developed specific to the site. For
coverage under the permit, the SWPPP must be prepared before commencement of construction
and then updated as  appropriate. Commencement of construction activities is defined as the
initial disturbance of soils associated with clearing, grading, or excavating activities or other
construction-related activities  (e.g., stockpiling of fill material).

The permit also clarifies that once a definable area of the site has been finally stabilized, no
further SWPPP requirements apply to that portion of the site as long as the SWPPP has been
                                           2-5

-------
                                                                       Section 2: Background
updated accordingly to identify that portion of the site as complete. The SWPPP must be
implemented as written from the beginning of construction activity until final stabilization is
complete. Stabilization practices include seeding of temporary vegetation, seeding of permanent
vegetation, mulching, geotextiles, sod stabilization, vegetative buffer strips, preservation of trees
and mature vegetative buffer strips, and other appropriate measures. For a detailed description of
all permit requirements and conditions, see the CGP.

 2.3.     POLLUTION PREVENTION ACT OF 1990

The Pollution Prevention Act of 1990 (PPA) (42 U.S.C. 13101 etseq., Pub. L. 101-508,
November 5, 1990) makes pollution prevention the national policy of the United States. The PPA
identifies an environmental management hierarchy in which pollution "should be prevented or
reduced whenever feasible; pollution that cannot be prevented should be recycled in an
environmentally safe manner, whenever feasible; pollution that cannot be prevented or recycled
should be treated in an environmentally safe manner whenever feasible; and disposal or release
into the environment should be employed only as a last resort..." (42 U.S.C. 13103). In short,
preventing pollution before it is created is preferable to trying to manage, treat, or dispose of it
after it is created. According to the PPA, source reduction reduces the generation and release of
hazardous substances, pollutants, wastes, contaminants, or residuals at the source, usually within
a process. The term source reduction "...includes equipment or technology modifications,
process or procedure modifications, reformulation or redesign of products, substitution of raw
materials, and  improvements in housekeeping, maintenance, training, or inventory control. The
term source reduction does not include any practice [that] alters the physical, chemical, or
biological characteristics or the volume of a hazardous substance, pollutant, or contaminant
through a process or activity which itself is not integral to or necessary for the production of a
product or the  providing of a service." In effect, source reduction means reducing the amount of
a pollutant that enters a wastestream or that is otherwise released into the environment before
out-of-process recycling, treatment, or disposal.

Although the PPA does not explicitly address stormwater discharges or discharges from
construction sites, the principles of the PPA are implicit in many of the practices used to reduce
pollutant discharges from construction sites. These include controls that minimize the potential
for erosion such as proper phasing of construction, retention of on-site vegetation and
stabilization of disturbed areas as soon as practicable.  Such controls and practices are described
in Section 7 of this document.

 2.4.    STATE REGULATIONS

States and municipalities have been regulating discharges of runoff from the construction and
land development industry to varying degrees for some time. A compilation of state CGPs and
regulations was prepared to help establish the baseline for national and regional levels of control.
Data were collected by reviewing state CGPs, Web sites, summary references, and state ESC or
stormwater management guidance manuals. The state  regulatory data are discussed in Sections
3.4 and 9.2 of this document, and the complete data sheets are included in Appendix A.
                                           2-6

-------
                                                                 Section 3: Data Collection
3.   DATA COLLECTION

 3.1.    INTRODUCTION

As part of the regulatory efforts to develop the proposed Construction and Development (C&D)
regulations in 2002 and the related final action in 2004, the U.S. Environmental Protection
Agency (EPA) gathered and evaluated an extensive amount of technical and economic data from
various sources. EPA used much of the data collected for the previous rulemaking effort in
support of this effort. EPA also collected additional data and information to support the technical
analyses used in developing this final rule. This section summarizes EPA's data collection
efforts.

 3.2.    LITERATURE SEARCH

A literature search was performed to obtain additional information on various erosion and
sediment control (ESC) technologies that pertain to the C&D industry. Journal articles and
professional conference proceedings were reviewed to collect recent data and information related
to ESC design and installation criteria, performance, and related costs. Annotated bibliographies
for the journal articles and professional conference proceedings that EPA reviewed for possible
use in developing this final rule are in Appendix B (costs), Document Control Number (DCN)
44321 (sediment basin performance), and DCN 43114 (passive treatment).

 3.3.    DATA AND INFORMATION PROVIDED IN RESPONSE TO THE 2002
         REGULATORY ACTION AND THE 2008 PROPOSAL

In response to the previous rulemaking efforts for the C&D industry, EPA received numerous
public comments on most aspects of the 2002 proposed rule. EPA considered those comments in
developing this final rule. EPA also considered public comments received on the 2008 proposal.

 3.4.    COMPILATION OF STATE CONTROL STRATEGIES, CRITERIA, AND
         STANDARDS

EPA compiled and evaluated existing state program information for the control of construction
site stormwater. The data were collected by reviewing state construction general permits (CGPs),
Web sites, summary references, state regulations, and ESC design and guidance manuals. A
summary of criteria and standards for construction site stormwater ESC that are implemented by
states is presented in Appendix A. More information on this analysis is in Section 9.2, Analysis
of State Equivalency. Appendix A also includes updated state information that EPA obtained in
early 2007, state-level data sheets and information originally presented in Section 7 and
Appendix D of the 2004 Development Document for Final Action for Effluent Guidelines and
Standards for the Construction and Development Category (EPA-821-B-04-001), and
information originally presented in  Appendix A of the June 2002 Development Document for
Proposed Effluent Guidelines and Standards for the Construction and Development Category
(EPA-821-R-02-007).
                                         3-1

-------
                                                                  Section 3: Data Collection
 3.5.    OTHER DATA SOURCES

    3.5.1.   LAND USE DATA

EPA accessed a number of sources of land cover information at a national scale for use in
estimating the potential number of acres subject to C&D activities.

      3.5.1.1.    National Land Cover Dataset

The National Land Cover Database (NLCD) provides a national source of data on land cover
change. The Multi-Resolution Land Characteristics Consortium (MRLC) has produced the
NLCD data sets that created a 30-meter resolution land cover data layer over the conterminous
United States (CONUS) from Landsat Thematic Mapper satellite imagery. NLCD data are
publicly available for the years 1992 and 2001 (see http://www.epa.gov/mrlc/ and
http://www.mrlc.gov/).

Because new developments in mapping methodology, new sources of input data, and changes in
the mapping legend for the 2001 NLCD confound direct comparison between 2001 NLCD and
the 1992 NLCD (MRLC 1992 and 2001), the U.S. Geological Survey (USGS) prepared and
released the NLCD 1992/2001 Retrofit Land Cover Change Product. The NLCD 1992/2001
Retrofit Land Cover Change Product was developed to offer more accurate direct change
analysis between the two products.

The NLCD 1992/2001 Retrofit Land Cover Change Product uses a specially developed
methodology to provide land cover change information at the Anderson Level I classification
scale, relying on decision tree classification of Landsat imagery from 1992 and 2001. While
NLCD 1992 reported on developed land in the categories of low-residential intensity, high-
residential intensity, commercial/industrial/transportation, and urban/recreational grasses, NLCD
2001 reported categories of developed low, medium,  high, and open space. To compare change
between the two data sets, the developed categories were merged into one overall urban class.
Unchanged pixels between the two dates are coded with the NLCD 2001 Anderson Level I class
code, while changed pixels are labeled with afrom-to land cover change value. Modified
Anderson Level 1 Classifications include the following:
       •  Open water
       •  Urban
       •  Barren
       •  Forest
       •  Grassland/Shrub
       •  Agriculture
       •  Wetlands Ice/Snow

The NLCD 1992/2001 Retrofit Land Cover Change Product was intended to provide a current,
consistent, and seamless data set for the United States at medium spatial resolution for Anderson
Level 1 classes. This land cover change map and all documents pertaining to it are considered
provisional until a formal accuracy assessment can be conducted.
                                         3-2

-------
                                                                     Section 3: Data Collection
EPA used the NLCD to estimate the annual number of acres of land converted to urban land uses
in the United States during the period between 1992 and 2001. At proposal, EPA used the NCLD
results to estimate acres of construction activities subject to the national effluent guidelines
regulations because no national database of the number and size of construction activities exists.
For the final rule analysis, the NLCD data was not used to estimate the amount of construction
activity occurring. EPA used economic data to estimate expected levels of construction activity
(for more information, see the Economic Analysis). EPA used the NCLD to apportion consruction
activity to watersheds as a basis for estimating baseline loadings and loadings reductions of the
regulatory options. Figure 3-1 illustrates an example of the Reach File Version 1.0 (RFl)-level
analysis of the NLCD data. EPA used the RF1 watershed cataloging system (described below)
because the SPARROW model (which is the model EPA used to estimate water quality
improvements) operates at the RF1 scale.
                                                              0'  65 130 Miles'	
                                                               Legend

                                                               |   | RF1 watershed
                                                               1992-2001 Change Data
                                                               ^H Change to Urban

                                                               NLCD Class
                                                               Modified Anderson
                                                               |    | Agriculture
                                                                 ^| Barren
                                                                 ^| Forest
                                                               |    | Grassland/Shrub
                                                               |    | Ice/Snow
                                                               |    | Open Water
                                                                  ] Urban
                                                               I    I Wetlands
Figure 3-1. NLCD 1992/2001 land cover change product near Seattle, Washington.

Table 3-1 summarizes the national- and state-level estimates obtained from the NLCD analysis.
Figures 3-2 through 3-11 graphically present the results of the NLCD land use change analysis at
the RF1 level for each EPA Region. Results are presented as percent urban change between 1992
and 2001. (For an index of the NLCD-related analyses conducted for the rule,  see DCN 43097 in
the Administrative Record.)
                                           3-3

-------
                                                        Section 3: Data Collection
Table 3-1. State and national estimates of urban land from NLCD

Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
1992 Urban
acres
2,066,843
2001 Urban
acres
2,197,496
% of state
that is
developed
(1992)
6.40%
% of state
that is
developed
(2001)
6.80%
Annual rate of
development,
1992-2001
(acres)
14,517
NO DATA
1,285,258
1,836,496
6,278,143
1,609,387
727,078
113,052
4,526,626
3,026,921
1,408,765
1,912,492
6,524,815
1,751,902
736,015
120,720
4,870,084
3,319,772
1 .80%
5.50%
6.20%
2.40%
23.40%
9.40%
13.00%
8.20%
1 .97%
5.73%
6.44%
2.61%
23.69%
10.04%
13.99%
8.99%
13,723
8,444
27,408
15,835
993
852
38,162
32,539
NO DATA
847,520
4,014,480
2,238,170
2,527,225
2,463,194
1,740,669
1,788,423
653,697
698,386
1,174,234
3,746,569
2,648,001
1,721,138
2,845,661
1,187,901
1,699,570
572,706
426,786
1,124,705
799,207
2,682,301
2,816,229
1,667,029
3,549,025
2,387,508
1,552,824
3,006,384
173,764
1,487,194
1,315,111
2,189,700
8,229,892
758,031
304,570
1,818,500
898,118
4,197,711
2,353,388
2,621,239
2,666,459
1,830,327
1,903,893
695,682
754,384
1,203,889
3,946,405
2,731,809
1,827,869
2,967,035
1 ,246,068
1,752,634
646,794
443,382
1,162,613
838,609
2,752,573
2,984,988
1,727,113
3,705,445
2,537,439
1,617,957
3,149,538
177,085
1 ,632,427
1,388,776
2,307,879
8,791,816
831,309
309,628
1,954,409
1 .60%
1 1 .30%
9.90%
7.10%
4.70%
6.80%
6.60%
3.30%
1 1 .30%
23.90%
10.30%
5.20%
5.70%
6.50%
1 .30%
3.50%
0.80%
7.50%
23.70%
1 .00%
8.90%
9.10%
3.90%
13.60%
5.40%
2.50%
10.40%
27.00%
7.70%
2.70%
8.30%
5.00%
1 .40%
5.40%
7.20%
1 .70%
1 1 .82%
10.41%
7.36%
5.09%
7.15%
7.03%
3.51%
12.21%
24.50%
10.85%
5.36%
6.05%
6.78%
1 .36%
3.61%
0.90%
7.79%
24.50%
1 .05%
9.13%
9.65%
4.04%
14.20%
5.74%
2.60%
10.90%
27.52%
8.45%
2.85%
8.75%
5.34%
1 .54%
5.49%
7.74%
5,622
20,359
12,802
10,446
22,585
9,962
12,830
4,665
6,222
3,295
22,204
9,312
11,859
13,486
6,463
5,896
8,232
1,844
4,212
4,378
7,808
18,751
6,676
17,380
16,659
7,237
15,906
369
16,137
8,185
13,131
62,436
8,142
562
15,101
                             3-4

-------
                                                                           Section 3: Data Collection

Washington
West Virginia
Wisconsin
Wyoming
District of Columbia
Nation
1992 Urban
acres
2,286,574
1,016,805
2,345,956
491,168
26,381
96,492,992
2001 Urban
acres
2,402,332
1,049,133
2,411,998
516,818
28,865
101,807,897
% of state
that is
developed
(1992)
5.40%
6.70%
6.70%
0.80%
82.80%

% of state
that is
developed
(2001)
5.67%
6.91%
6.89%
0.84%
90.60%

Annual rate of
development,
1992-2001
(acres)
12,862
3,592
7,338
2,850
276
590,545
                                          % Developed from 1992-2001
                                              0.01% -0.1%
                                              0.11% -0.5%
                                              0.51% -2%
                                              > 2%
                                                    H Miles
Figure 3-2. EPA Region 1: Percent urban change 1992-2001 by ERF1_2 watershed.
                                               3-5

-------
                                                                         Section 3: Data Collection
                % Developed from 1992-2001
                    0.01% -0.1%
                    0.11% -0.5%
                    0.51% -2%
                ^B > 2%
                            L
Figure 3-3. EPA Region 2: Percent urban change 1992-2001 by ERF1_2 watershed.
                                                  % Developed from 1992-2001
                                                     0.01%-0.1%
                                                     0.11%-0.5%
                                                     0.51%-2%
                                                     I > 2%
Figure 3-4. EPA Region 3: Percent urban change 1992-2001 by ERF1_2 watershed.
                                              3-6

-------
                                                                          Section 3: Data Collection
            % Developed from 1992-2001
                0.01% -0.1%
                0.11% -0.5%
                0.51% -2%
            ^H > 2%
Figure 3-5. EPA Region 4: Percent urban change 1992-2001 by ERF1_2 watershed.
                   % Developed from 1992-2001
                       0.01% -0.1%
                       0.11% -0.5%
                       0.51% -2%
                   ^H > 2%
                                                                     ] Miles
Figure 3-6. EPA Region 5: Percent urban change 1992-2001 by ERF1_2 watershed.
                                              3-7

-------
                                                                        Section 3: Data Collection
                                      1          •   «   W    ^
                                       :*>'      "         -"
                                       >-  IT
       % Developed from 1992-2001
           «X01%
           0.01%-0.1%
           0.11%-0.5%
           0.51%-2%
       ^B > 2%
] Miles
Figure 3-7. EPA Region 6: Percent urban change 1992-2001 by ERF1_2 watershed.
                     % Developed from 1992-2001
Figure 3-8. EPA Region 7: Percent urban change 1992-2001 by ERF1_2 Watershed.
                                             3-8

-------
                                                                        Section 3: Data Collection
                                                          % Developed from 1992-2001
                                                             0.01% -0.1%
                                                             0.11% -0.5%
                                                             0.51% -2%
                                                             > 2%
Figure 3-9. EPA Region 8: Percent urban change 1992-2001 by ERF1_2 watershed.
                                             % Developed from 1992-2001
Figure 3-10. EPA Region 9: Percent urban change 1992-2001 by ERF1_2 watershed.
                                             3-9

-------
                                                                    Section 3: Data Collection
                                             % Developed from 1992-2001
                                                 0.01% -0.1%

                                                 0.11% -0.5%

                                                 0.51% -2%
Figure 3-11. EPA Region 10: Percent urban change 1992-2001 by ERF1_2 watershed.

Because NLCD data does not exist for Alaska, Hawaii, or the U.S. territories, EPA's analysis
does not consider pollutant loading reductions for those areas. Detailed definitions and
discussion of the NLCD 1992/2001 Retrofit Product is presented in Appendix E. DCN 43097 in
the Administrative Record provides an index to the NLCD-related analyses conducted for the
rulemaking.

      3.5.1.2.  River Reach File Data

An option for summarizing national land cover change in drainage area units is to use EPA's
RF1  for the CONUS. RF1 is a vector database of approximately 700,000 miles of streams and
open waters in the CONUS. EPA and states use it extensively, and the U.S. Fish and Wildlife
Service and the National Weather Service (NWS) have used it for many years. EPA prepared
RF1  in 1982 from National Oceanic and Atmospheric Administration (NOAA) aeronautical
charts having a scale of 1:500,000. Those charts provided the best nationwide hydrographic
coverage available on a single scale at that time. They include all hydrography shown on USGS
maps having a scale of 1:250,000 with extensive additions, corrections, and improvements in
detail made by NOAA from aerial photography and satellite imagery. In the 1980s, EPA used
RF1  for performing water quality modeling on whole river basins for all the hydrologic regions
in the CONUS. In that role, it was used to provide national assessments and overviews of water
quality and to provide the foundation for a nationwide, stratified, sampling frame for performing
statistical summaries of modeled and measured water quality on all surface waters of the
CONUS.

A consistent, national-scale watershed data set was prepared to enhance the RF1 hydrology data
set. That watershed data set, the Enhanced River Reach File 1.2 (ERF1), was designed to be a
digital database of river reaches capable of supporting regional and national water-quality and
river-flow modeling and transport investigations in the water-resources community. USGS has
used ERF1 to support interpretations of stream water-quality monitoring network data. In such
analyses, the reach network has been used to determine flow pathways between the sources of
                                          3-10

-------
                                                                    Section 3: Data Collection
point and nonpoint pollutants and downstream water-quality monitoring locations in support of
predictive water-quality models of stream nutrient transport.

The Enhanced River Reach File 2.0 (ERF1_2) expands on ERF1 and includes the incremental
and total drainage area founded on the 1-kilometer (km) elevation data for North America (Nolan
et al. 2002). Previous estimates of the water time-of-travel were recomputed for reaches with
water quality monitoring sites that included two reaches. The mean flow and velocity estimates
for the split reaches are based on previous estimation methods (Alexander et al. 1999) and are
unchanged in ERF1_2. Drainage area calculations provide data used to estimate the contribution
of a given nutrient to the outflow. ERF1_2 contains 67,171 watersheds with a minimum size of
247 acres (1  square km [km2]) and an average size of 30,182 acres (122 km2).

EPA used the ERF1_2 as the foundation for summarizing land cover change and in drainage area
units (or watersheds) and for SPARROW (Spatially Referenced Regressions [of nutrient
transport] on Watershed) modeling. Within the context of a geographic information system
(GIS), SPARROW estimates the proportion of watersheds in the CONUS with outflow flux of
several nutrients, including total nitrogen and total phosphorus, (Smith et al. 1997). EPA
modified SPARROW to model changes in sediment flux in  the RF1 network to evaluate
potential benefits of regulatory options. Sediment and nutrient flux were converted to
concentrations using estimates of reach flow for each RF1.

   3.5.2.   NPDES PERMIT NOTICE OF INTENT DATA

EPA used CGP Notice of Intent (NOT) records to characterize construction activity by project
type and project size for subsequent analysis of costs and pollutant loading reductions. Using
NOT data, EPA broadly characterized the construction industry into three land use types
(residential construction, nonresidential construction, and road/highway construction). EPA has
NOT data for approximately 22,000 permit applications, containing data from four states for
construction activities occurring primarily between 2003 and 2009. While the NOT data are
useful for characterizing  construction activities into different project types and project sizes, as
well as for estimating the duration of projects, EPA did not find the NOT data useful as a national
data set to estimate the amount of construction occurring. That is because the NOT data obtained
by EPA are not national in coverage. EPA's analysis of the NOT data are in Appendix C.

   3.5.3.   CLIMATIC/RAINFALL DATA

      3.5.3.1.   NOAA National Weather Service Precipitation Frequency Data
                Server

Variations in rainfall depth and intensity are important factors in determining erosion rates,
sediment discharges, pollutant load reductions,  and control technology costs for construction
sites. EPA used a combination case study approach of 11 indicator cities as well as national data
sets for different components of the analysis. Indicator cities were used for certain components
of the cost analysis, such as  estimating design storm depths. EPA also used indicator cities in the
loadings analysis to develop runoff coefficients. However, national data coverages were used for
other components of the loads analysis, such as estimating average annual precipitation values
forRFls.
                                          3-11

-------
                                                                      Section 3: Data Collection
For the indicator city analysis, EPA selected representative areas in each of the 10 EPA Regions
to be used as a point estimate for the entire region. EPA generally selected the urban area in each
region with the greatest rate of development on the basis of EPA's analysis of land use change
from the NLCD analysis. EPA selected one metropolitan area in each of the 10 EPA Regions,
with the exception of Region  10. In Region 10, EPA selected two indicator cities because the two
areas with the greatest rate of development (Boise City, Idaho, and Seattle, Washington) have
very different rainfall patterns. For each of the 11 indicator cities, EPA obtained detailed rainfall
data and rainfall summaries. EPA also obtained detailed soils data for each of the 11 areas. The
11 indicator cities are identified in Table 3-2.
                            Table 3-2. EPA Region indicator cities
EPA
Region
1
2
3
4
5
6
7
8
9
10
Indicator city
Manchester, New Hampshire
Albany, New York
Washington, DC, Virginia, and Maryland
Atlanta, Georgia
Chicago, Illinois — Indiana
Dallas, Fort Worth, and Arlington, Texas
Kansas City, Missouri and Kansas
Denver and Aurora, Colorado
Las Vegas, Nevada
Boise City, Idaho, and Seattle, Washington
EPA's costing analysis used state-specific design storms for determining stormwater runoff rates
and volumes and for determining storage volumes and treatment system sizing. EPA identified
one major city within each state to serve as an indicator for the entire state. EPA obtained rainfall
summary data for each of these cities for using as a basis for determining expected runoff rates
and rainfall volumes for costing of technologies.

Precipitation data was gathered and analyzed using the NOAA NWS Precipitation Frequency
Data Server (PFDS). The Hydrometeorological Design Studies Center (HDSC) in the Office of
Hydrologic Development of the NWS is in an ongoing process of updating its precipitation
frequency estimates, which are available in NOAA Atlas 14 format. At the time of this writing,
only a portion of the United States had been updated into this format (NWS 2008). Atlas 14
supersedes precipitation frequency estimates contained in previous NWS publications. The
updates are based on more recent and extended data sets, currently accepted statistical
approaches,  and improved spatial interpolation and mapping techniques. A complete list of NWS
publications is at http://www.nws.noaa.gov/ohd/hdsc/currentpf.htm.

NOAA Atlas 14 contains precipitation frequency estimates with associated confidence limits for
the United States for 5-minute through 60-day durations at average recurrence intervals of 1-year
through 1,000-year. The estimates are based on the analysis of annual maximum series and then
converted to partial duration series results. The Atlas 14 rainfall data results used in this study
are shown in Appendix D, Table D-3.
                                           3-12

-------
                                                                     Section 3: Data Collection
For the states not currently updated by NOAA Atlas 14, the rainfall-frequency values for
selected durations were estimated using a series of maps presented in the older NWS
publications. The data for the remainder of the western United States were estimated by using
NOAA Atlas 2, Precipitation Frequency Atlas of the Western United States (NOAA 1973),
which are generalized maps presented for 6- and 24-hour point precipitation for the return
periods of 2, 5,  10, 25, 50, and  100 years. Atlas 2 is published in separate volumes for each of the
states. Similarly, the maps presented in the corresponding technical paper were used for the
remainder of the eastern United States and Hawaii. (Alaska was not included in this study
because EPA lacked sufficient data on the annual amount of construction activities in Alaska).

Precipitation frequency results generated by Atlas 2  or technical paper maps are presented in
Appendix D, Table D-4. The rainfall depths were estimated by identifying the target city on the
Atlas 2 or technical paper map and linearly approximating the rainfall value. For example, if a
city fell between a depth of 4.5 and 5 inches, and the city was approximately 20 percent of the
map distance from the 5-inch line, a rainfall depth of 4.9 inches was estimated. Note that the
maps provide data for depth only. Intensity estimates were calculated by dividing the duration
(e.g., 6- or 24-hour) by the depth. Additionally, Atlas 2 depths were converted from tenths of an
inch to inches.

To analyze the percent of total construction site runoff captured and treated for various
regulatory options, EPA obtained hourly precipitation data for each indicator city. EPA obtained
30 years of hourly  rainfall data from Earthlnfo Version 2.31 (www.earthinfo.com). Earthlnfo
provides National Climatic Data Center meteorological data in an easy-to-use format from which
precipitation data can be extracted. From the 7,000 National Climatic Data Center gages
available, EPA generally used data collected at an airport in or adjacent to each indicator city. In
general, EPA analyzed data for the period between the mid-1970s and mid-2000s. EPA also used
the hourly precipitation data in Earthlnfo to evaluate the number and size of rainfall events that
discharge from construction sites. Appendix H details  this evaluation that focuses on the 11
indicator cities.

Table 3-3 summarizes the state-specific rainfall data EPA used in its analyses.

      3.5.3.2.   Parameter Elevation Regressions on Independent Slopes Model
                (PRISM)

EPA's analyses of the regulatory options used estimates of the average annual precipitation for
each RF1 watershed. Annual precipitation was used to estimate runoff volumes  and baseline
sediment concentrations as well as to evaluate removals under regulatory options that
incorporated a numeric limit. For each RF1 watershed, the average annual precipitation amount
was obtained from the 1-km resolution United States Average Monthly or Annual Precipitation
(1971-2000) PRISM Group raster data coverage (PRISM Group 2006). RF1 watershed
boundaries were used to summarize the PRISM Group average annual rainfall values, and each
RF1 was assigned a value by spatially averaging contributing raster data. Figure 3-12 shows
average annual precipitation for the CONUS from the  PRISM data. Additional information on
the PRISM data is  in Appendix D.
                                          3-13

-------
Table 3-3. Rainfall summary data for indicator cities
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
City
Montgomery
Phoenix
Little Rock
Sacramento
Denver
Hartford
Dover
Tallahassee
Atlanta
Honolulu
Boise
Chicago
Indianapolis
Des Moines
Kansas City
Frankfort
Baton Rouge
Augusta
Baltimore
Boston
Lansing
St. Paul
Jackson
Kansas City
Helena
Lincoln
Las Vegas
Manchester
Hightstown
Santa Fe
Average
annual
precipitation
(inches)
49
8
48
18
13
44
43
62
51
18
11
33
40
32
37
45
59
42
42
42
30
29
52
37
12
28
4
40
47
15
2-year,
24-hour storm
depth
(inches)
4.50
1.40
4.10
2.00
2.00
3.10
3.26
4.75
3.70
4.25
1.20
2.85
2.95
3.25
3.50
3.00
5.25
2.80
3.16
3.10
2.40
2.75
4.45
3.45
1.30
3.00
1.00
2.80
3.31
1.54
10-year,
24-hour storm
depth
(inches)
6.5
2.14
6.05
3
3
4.8
5.08
7.4
5.5
7.8
1.8
4.29
4.13
4.7
5.2
4.34
8.2
4.25
4.85
4.5
3.6
4.2
6.7
5.3
2.1
4.8
1.62
4.3
5.07
2.22
25-year,
24-hour storm
depth
(inches)
7.6
2.59
7
3.5
3.8
5.5
6.36
8.5
6.5
8.9
2.2
5.25
4.83
5.5
6.1
5.23
9.1
4.9
6.08
5.5
4.2
4.7
7.8
6
2.4
5.4
1.96
5
6.3
2.62
10-year, 6
hour storm
depth
(inches)
4.60
1.57
4.35
1.70
2.30
3.25
3.44
5.25
4.20
4.80
1.20
3.30
3.12
3.54
3.90
3.09
5.75
2.90
3.32
3.30
2.70
3.10
4.70
3.85
1.10
3.52
1.29
3.20
3.55
1.77

-------
State
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Puerto Rico
District of Columbia
City
Albany
Charlotte
Bismarck
Columbus
Oklahoma City
Salem
Philadelphia
Providence
Columbia
Pierre
Nashville
Fort Worth
Salt Lake City
Montpelier
Arlington
Seattle
Charleston
Madison
Cheyenne
San Juan
Washington
Average
annual
precipitation
(inches)
37
43
16
38
33
41
42
45
45
16
46
33
15
34
40
35
43
31
15
51
42
2-year,
24-hour storm
depth
(inches)
2.90
3.34
1.90
2.62
3.70
2.50
3.23
3.20
3.62
2.25
3.37
3.90
1.40
2.40
3.11
2.00
2.56
2.80
1.60
4.26
3.16
10-year,
24-hour storm
depth
(inches)
4
4.86
3.25
3.73
5.8
3.5
4.8
4.8
5.28
3.5
4.7
6.3
1.9
3.7
4.78
3
3.55
4.1
2.4
6.76
4.85
25-year,
24-hour storm
depth
(inches)
5.9
5.76
3.75
4.44
6.9
4
5.85
5.7
6.39
4.1
5.53
7.4
2.21
4.25
5.98
3.4
4.16
4.75
2.8
8.29
6.07
10-year, 6
hour storm
depth
(inches)
3.10
3.54
2.50
2.80
4.25
2.90
3.38
3.40
3.85
2.75
3.31
4.55
1.27
2.70
3.29
1.40
2.56
3.15
1.90
4.42
3.32

-------
      Avg Annual Precip (in)
      |    | 1 - 20 ^^61-80
      Q^| 21 -30 ^| 81 -100
      |    | 31 -40 ^m 101 - 150
        ~~|41 -60 ^B 151 -283
Figure 3-12. Average annual precipitation in the CONUS from PRISM.

-------
                                                                   Section 3: Data Collection
   3.5.4.   SOILS DATA

The variation in soil types found in the United States is a significant factor in estimating
sediment discharges. EPA used soil coverage data provided in the State Soil Geographic
Database (STATSGO) (Wolock 1997; USDA 2007) and the CONUS-SOIL data layers (Miller
and White 1998) for the loadings analysis. STATSGO component and layer tables were accessed
through the Pennsylvania State University's active archive
(http://www.soilinfo.psu.edu/index.cgi?soil_data&index.html). EPA extracted data for only the
portions of RF1 watersheds where development change has been documented by NLCD. This
urban masking approach was implemented using a binary raster grid of NLCD 1992/2001
Retrofit Land Cover Change Product representing urbanization to weight RF1 values for
STATSGO data layers. EPA used the mask of urbanization change to create RF1 watershed
parameter values. Essentially, the individual spatial units of the soil coverage—Map Unit
Identifiers (MUIDs)—were summarized by urbanizing area weights into RF1 average values,
instead of using proportional weight based on land area (Figure 3-13). The geographic limits of
the soil coverage evaluated were determined by superimposing indicator city urban area
boundaries—from the U.S. Census Bureau's 2000 Urbanized Areas Cartographic Boundary Files
(U.S. Census Bureau 2000)—on intersecting RF1 watersheds. The resulting list of RF1
watersheds intersecting the rapidly developing indicator city urban areas was used to spatially
identify underlying STATSGO soil coverage MUIDs. Last, soil data associated with the surface
soil layer within the selected MUIDs were extracted from STATSGO to produce the suite of data
evaluated for each indicator city.
   RF
          RF1: 55517
     : 55527

                                     WA205
                                     RF1: 5! 516
                             RF1: 55520
                        WA212
               RF1:55521
                WA194
                                    RF1*55522
                               WA18
    RF1: 55525
    0.9 0.45 0
                                 RF1: 55523
Legend
   | Urban Change
|    | ERF1.2 watershed
     STATSGO MUID
Figure 3-13. An example of Washington State RF1 watershed showing urban weighting
emphasizes MUID: WA189, while area-weighting would have enhanced MUID WA194.


   3.5.5.   VENDOR DATA FOR ACTIVE TREATMENT SYSTEMS

EPA compiled and evaluated information from vendors on treatment technologies that could be
used in setting numeric standard discharge limits for stormwater runoff from construction sites.
EPA conducted an Internet-based search and placed telephone calls to several vendors to gather
                                         3-17

-------
                                                                  Section 3: Data Collection
data on available treatment technologies, costs, and performance (for vendor-specific
information and fact sheets, see DCNs 43000 through 43011 and DCN 43081 in the
Administrative Record). EPA also received unsolicited e-mails with data from vendors. After
publishing the November 2008 proposed rule, EPA received additional data from vendors (see
DCN 43125).

    3.5.6.   RAINFALL AND RUNOFF EROSIVITY FACTOR

EPA used a GIS data layer for the RUSLE R factor to determine average R factors for RF1 as a
component of the loadings analysis. The R factor (USDA 1997) is an indicator of rainfall energy
and intensity and varies seasonally across the United States. EPA uses this data for determining
whether small construction sites can qualify for the Low Erosivity Waiver (LEW) that is in the
NPDES Phase II stormwater regulations. EPA has an online tool that can be used to determine if
sites qualify for the LEW (see http://cfpub.epa.gov/npdes/stormwater/lew/lewcalculator.cfm).
Figure 3-14 shows annual R factor values for the CONUS. Again using an urban masking
approach, EPA derived the R factor values for RF1 watersheds on the basis of averaging values
underlying land undergoing development according to the NLCD 1992/2001 Retrofit Land
Cover Change Product.

    3.5.7.   HYDROLOGIC SOIL GROUPS

EPA used GIS data to determine the percent of each hydrologic soil group (HSG) for each RF1
watershed using the urban masking approach. The per RF1 HSG percentages were then used to
estimate runoff coefficients for each RF1. As described in Section 10.3, EPA first determined the
hydrologic response of indicator cities independently for each soil class, i.e., A soil, B soil, C
soil, and D soil. Next, the effective per RF1 runoff coefficient was determined by prorating the
hydrologic response of the adjacent indicator city using the RF1 HSG percentages. That resulted
in a customized runoff coefficient for each RF1. To provide insight into the variability of the
HSG within CONUS, Table 3-4 shows the percent of each HSG by state.
                                         3-18

-------
CD
                                                     276 - 300
                                                     301 - 350
                                                     351 - 400
                                                     401 - 450
                                                     451 - 500
R factor
|     | 7-9.9
|     | 9.91 - 19.9
|     | 20-29.9
Q^] 30-39.9

H 50-59.9
^| 60-69.9
•• 70 - 79.9
80 - 89.9
90 - 99.9
100- 125
126- 150
151 - 175
176-200 |    | 501 -550
201 - 225 |    | 551 - 600
226-250 |    | 601 -818
251 - 275
                                                                                             500
                                                                                                       J Miles
           Figure 3-14. Annual R factor values for CONUS.

-------
                                                            Table 3-4. HSGs by state
ro
o


Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
HSG
A
8.7%
4.7%
0.6%
10.9%
7.2%
9.1%
20.8%
18.1%
6.6%
4.4%
1 .4%
3.5%
0.9%
3.8%
0.1%
1 .7%
7.7%
10.0%
23.9%
29.0%
8.3%
2.3%
1 .0%
2.9%
B
41 .2%
38.6%
28.3%
32.2%
46.7%
41.1%
30.9%
6.3%
53.1%
46.8%
44.5%
32.6%
66.0%
58.0%
42.7%
14.4%
12.9%
38.6%
16.6%
28.7%
37.4%
32.3%
40.1%
39.5%
C
28.8%
17.2%
35.9%
18.4%
24.6%
35.9%
13.4%
8.6%
16.9%
23.1%
27.0%
41 .8%
1 1 .6%
19.5%
44.9%
28.9%
43.9%
26.4%
34.4%
12.9%
15.4%
38.6%
39.8%
27.2%
D
21 .3%
39.5%
35.1%
38.5%
21 .4%
13.9%
34.9%
67.0%
23.5%
25.7%
27.1%
22.1%
21 .5%
18.7%
12.3%
55.1%
35.5%
25.0%
25.2%
29.4%
38.9%
26.9%
19.0%
30.4%


Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
HSG
A
31 .9%
5.6%
17.1%
12.5%
5.6%
9.6%
7.9%
4.7%
0.6%
6.8%
5.2%
6.0%
15.3%
1 1 .9%
2.9%
0.1%
5.1%
5.3%
4.9%
1 .7%
6.6%
7.3%
14.4%
4.5%
B
53.6%
26.4%
24.8%
32.8%
41 .9%
18.5%
48.8%
56.1%
16.8%
44.5%
32.1%
28.4%
35.7%
41 .8%
45.2%
53.6%
27.2%
36.2%
18.0%
53.7%
53.4%
21 .5%
46.8%
40.5%
C
3.0%
17.7%
41 .4%
25.1%
16.5%
51.1%
16.5%
16.6%
54.6%
22.3%
37.1%
54.2%
32.4%
19.5%
1 1 .5%
30.4%
24.5%
16.2%
54.3%
32.3%
24.2%
54.2%
18.1%
19.5%
D
1 1 .5%
50.3%
16.6%
29.6%
36.0%
20.7%
26.8%
22.6%
28.0%
26.4%
25.6%
1 1 .5%
16.5%
26.8%
40.4%
15.9%
43.2%
42.3%
22.8%
12.3%
15.8%
17.0%
20.7%
35.5%

-------
                                                                   Section 3: Data Collection
 3.6.    REFERENCES

Alexander, R.B., J.W. Brakebill, R.E. Brew, and R.A. Smith. 1999. Enhanced River Reach File
    1.2 (ERF1) (Online database). U.S. Department of the Interior, U.S. Geological Survey,
    Reston, VA. .

Miller, D.A., and R.A. White. 1998. A conterminous United States multilayer soil characteristics
    data set for regional climate and hydrology modeling. Earth Interactions 2(2): 1-26.

MRLC (Multi-Resolution Land Characteristics Consortium). 1992 and 2001. National Land
    Cover Database. U.S. Department of the Interior, U.S. Geological  Survey, Land Cover
    Institute, . Updated March 21, 2008.

Nolan, J.V., J.W. Brakebill, R.B. Alexander, and G.E. Schwarz. 2002. Enhanced River Reach
    File 2.0 (ERF1_2) (Online database). U.S. Department of the Interior, U.S. Geological
    Survey, Reston, VA. .

NOAA (National Oceanic and Atmospheric Administration). 1973. Precipitation Frequency
    Atlas of the Western United States. Vols. 1-3, 5, 9-11. U.S. Department of Commerce,
    National Weather Service. Silver Spring, MD. Precipitation Frequency Data Server (PFDS).
    . Updated October 29, 2008.

NWS (National Weather Service). 2008. Precipitation Frequency Data Server. National Oceanic
    and Atmospheric Administration, National Weather Service, Office of Hydrologic
    Development, Silver Spring, MD. .  Updated March 4,
    2008.

PRISM Group. 2006. United States Average Monthly or Annual Precipitation (1971-2000).
    Oregon State University, Corvallis, OR. . Accessed March
    2008.

Smith, R.A., G.E. Schwarz, and R.B. Alexander. 1997. Regional interpretation of water-quality
    monitoring data. Water Resources Research 33(12):2781-2798.

U.S. Census Bureau. 2000. 2000 Urbanized Areas Cartographic Boundary Files. U.S. Census
    Bureau, Washington, DC. . Accessed
    January 2008.

USDA (U.S. Department of Agriculture).  1997. Predicting Soil Erosion by Water: A Guide to
    Conservation Planning with the Revised Universal Soil Loss Equation (RUSLE). Agriculture
    Handbook Number 703. U.S. Department of Agriculture, Agricultural Research Service.
    January, 1997.

USDA (U.S. Department of Agriculture).  2007. U.S.  General Soil Map (STATSGO) (Database).
    U.S. Department of Agriculture, Natural Resources Conservation Service, Washington, DC.
    .  Updated July 11, 2007.
                                         3-21

-------
                                                                  Section 3: Data Collection
   2004 Development Document for Final Action for Effluent Guidelines and Standards for the
   Construction and Development Category (EPA-821-B-04-001), and information originally
   presented in Appendix A of the June 2002 Development Document for Proposed Effluent
   Guidelines and Standards for the Construction and Development Category (EPA-821-R-02-
   007).

Wolock, D.M. 1997. STATSGO soil characteristics for the conterminous United States:
   U.S. Geological Survey Open-File Report 97-656. U.S. Geological Survey, Reston, VA.
   . Accessed December 2007.
                                         3-22

-------
                                                                      Section 4: Industry Profile
4.   INDUSTRY PROFILE

 4.1.    INTRODUCTION

The construction industry is divided into three major subsectors: general building contractors,
heavy construction contractors, and special trade contractors. General contractors build
residential, industrial, commercial, and other buildings. Heavy construction contractors build
sewers, roads, highways, bridges, and tunnels. Special trade contractors typically provide
carpentry, painting, plumbing, and electrical services.

Because the U.S. Environmental Protection Agency (EPA) is developing regulations to address
water quality issues, this document focuses on the construction subsectors most closely
associated with land-disturbing activities. General contractors and heavy construction
establishments are, by definition, the most likely to conduct activities that could affect water
resources. Note, however, that for individual projects, responsibility for land-disturbing activities
and potential effects on water quality might not be obvious because general contractors often
subcontract all or some of the actual construction work. Hence, the following subsections
describe the subsector categories most likely to be responsible for land-disturbing activities at the
national level.

The construction and development (C&D) industry is classified in the 2007 North American
Industry Classification System (NAICS) under Sector 23, Construction (U.S. Census Bureau
2008a). NAICS is the system used for classifying industry establishments by type of economic
activity, which replaced the U.S. Standard Industrial Classification (SIC) system.

The construction sector comprises establishments primarily engaged  in constructing buildings or
engineering projects (e.g., highways and utility systems). Establishments primarily  engaged in
preparing sites for new construction and establishments primarily engaged in subdividing land
for sale as building sites also are included in this sector (U.S. Census Bureau 2008a).

Construction work done can include new work, additions, alterations, or maintenance and
repairs. Activities of these establishments generally are managed at a fixed place of business, but
they usually perform construction activities at multiple project sites. Establishments identified as
construction-management firms are also included in the sector. The construction sector is divided
into three types of activities or subsectors as described below (from U.S. Census Bureau 2008a):

       •    Subsector 236—Construction of Buildings
           This subsector comprises establishments of the general contractor type and operative
           builders involved in constructing buildings. The work performed can include new
           work, additions, alterations, or maintenance and repairs. The on-site assembly of
           precut, panelized, and prefabricated buildings and construction of temporary
           buildings are included in this subsector. Part of or all the production work for which
           the establishments in this subsector have responsibility can be subcontracted to other
           construction establishments—usually specialty trade contractors. Establishments in
           this subsector are classified on the basis of the types of buildings they construct. This
           classification reflects variations in the requirements of the underlying production
           processes.
                                           4-1

-------
                                                                      Section 4: Industry Profile
       •  Subsector 237—Heavy and Civil Engineering Construction
          This sub sector comprises establishments whose primary activity is constructing entire
          engineering projects (e.g., highways and dams), and specialty trade contractors,
          whose primary activity is producing a specific component for such projects. Specialty
          trade contractors in Heavy and Civil Engineering Construction generally are
          performing activities that are specific to heavy and civil engineering construction
          projects and are not normally performed on buildings. The work performed can
          include new work, additions, alterations, or maintenance and repairs.
          Specialty trade activities are classified in this subsector if the skills and equipment
          present are specific to heavy or civil engineering construction projects. For example,
          specialized equipment is needed to paint lines on highways. That equipment is not
          normally used in building applications, so the activity is classified in this subsector.
          Traffic signal installation, while specific to highways, uses much of the same skills
          and equipment that are needed for electrical work in building projects  and is therefore
          classified in Subsector 238, Specialty Trade Contractors. Establishments in this
          subsector are classified on the basis of the types of structures that they construct. This
          classification reflects variations in the requirements of the underlying production
          processes.

       •  Subsector 238—Special Trade Contractors
          This subsector comprises establishments whose primary activity is performing
          specific activities (e.g., pouring concrete, site preparation, plumbing, painting,  and
          electrical work) involved in building construction or other activities that are similar
          for all  types of construction, but that are not responsible for the  entire project. The
          work performed can include new work, additions, alterations, maintenance, and
          repairs. The production work performed by establishments in this subsector is usually
          subcontracted from establishments of the general contractor type or operative
          builders, but especially in remodeling and repair construction. Work also can be done
          directly for the owner of the property. Specialty trade contractors usually perform
          most of their work at the construction site, although they might  have shops where
          they perform prefabrication and other work. Establishments primarily  engaged in
          preparing sites for new construction are also included in this subsector. There are
          substantial differences in types of equipment, work force skills, and other inputs
          required by specialty trade contractors. Establishments in this subsector are classified
          on the basis of the underlying production function for the specialty trade in which
          they specialize.

Table 4-1 provides a list of the 3-digit subsectors, 4-digit industry groups and 5-digitNAICS
industries in the construction sector.
                                            4-2

-------
                                                                     Section 4: Industry Profile
Table 4-1. 2007 NAICS subsectors, industry groups, and industries performing construction
                            activities that might disturb land
2007 NAICS Sector 23 - Construction
236
2361
23611
236115
236116
236117
236118
2362
23621
236210
23622
236220
237
2371
23711
237110
23712
237120
23713
237130
2372
23721
237210
2373
23731
237310
2379
23799
237990
238
2381
23811
238110
23812
238120
23813
238130
23814
238140
23815
238150
23816
238160
23817
238170
23819
238190
Construction of Buildings
Residential Building Construction
Residential Building Construction
New Single-Family Housing Construction
New Multifamily Housing Construction
New Housing Operative Builders
Residential Remodelers
Nonresidential Building Construction
Industrial Building Construction
Industrial Building Construction
Commercial and Institutional Building Construction
Commercial and Institutional Building Construction
Heavy and Civil Engineering Construction
Utility System Construction
Water and Sewer Line and Related Structures Construction
Water and Sewer Line and Related Structures Construction
Oil and Gas Pipeline and Related Structures Construction
Oil and Gas Pipeline and Related Structures Construction
Power and Communication Line and Related Structures Construction
Power and Communication Line and Related Structures Construction
Land Subdivision
Land Subdivision
Land Subdivision
Highway, Street, and Bridge Construction
Highway, Street, and Bridge Construction
Highway, Street, and Bridge Construction
Other Heavy and Civil Engineering Construction
Other Heavy and Civil Engineering Construction
Other Heavy and Civil Engineering Construction
Specialty Trade Contractors
Foundation, Structure, and Building Exterior Contractors
Poured Concrete Foundation and Structure Contractors
Poured Concrete Foundation and Structure Contractors
Structural Steel and Precast Concrete Contractors
Structural Steel and Precast Concrete Contractors
Framing Contractors
Framing Contractors
Masonry Contractors
Masonry Contractors
Glass and Glazing Contractors
Glass and Glazing Contractors
Roofing Contractors
Roofing Contractors
Siding Contractors
Siding Contractors
Other Foundation, Structure, and Building Exterior Contractors
Other Foundation, Structure, and Building Exterior Contractors
                                          4-3

-------
                                                                      Section 4: Industry Profile
2007 NAICS Sector 23 - Construction
2382
23821
238210
23822
238220
23829
238290
2383
23831
238310
23832
238320
23833
238330
23834
238340
23835
238350
23839
238390
2389
23891
238910
23899
238990
Building Equipment Contractors
Electrical Contractors and Other Wiring Installation Contractors
Electrical Contractors and Other Wiring Installation Contractors
Plumbing, Heating, and Air-Conditioning Contractors
Plumbing, Heating, and Air-Conditioning Contractors
Other Building Equipment Contractors
Other Building Equipment Contractors
Building Finishing Contractors
Drywall and Insulation Contractors
Drywall and Insulation Contractors
Painting and Wall Covering Contractors
Painting and Wall Covering Contractors
Flooring Contractors
Flooring Contractors
Tile and Terrazzo Contractors
Tile and Terrazzo Contractors
Finish Carpentry Contractors
Finish Carpentry Contractors
Other Building Finishing Contractors
Other Building Finishing Contractors
Other Specialty Trade Contractors
Site Preparation Contractors
Site Preparation Contractors
All Other Specialty Trade Contractors
All Other Specialty Trade Contractors
Source: U.S. Census Bureau 2008a

Before NAICS was created, C&D industries were classified using the SIC system. Any data
collected before January 1997 might still be classified under that system. SIC classifications are
relevant to the effluent guidelines because certain U.S. Census Bureau data for the construction
industry were collected until 1994 and, therefore, were classified under the SIC system rather
than the NAICS. Under the SIC system, industries that might perform land-disturbing activities
were classified under Division C-Construction, and Division H-Finance, Insurance, and Real
Estate. Those divisions include the following SIC major groups (from U.S. Census Bureau
2008b):

       •  SIC Major Group 15-Building Construction General Contractors and Operative
          Builders
          This group includes general contractors and operative builders primarily engaged in
          constructing residential, farm, commercial, or other buildings. General building
          contractors who combine a special trade with their contracting are also included.
       •  SIC Major Group 16-Heavy Construction other than Building Construction
          Contractors
          This group includes general contractors primarily  engaged in heavy construction
          other than building construction, such as highways and streets, bridges, sewers,
          railroads, irrigation projects, flood control projects, and marine construction, as well
          as special trade contractors primarily engaged in activities of a type clearly
          specialized in such heavy construction and not normally performed on buildings or
          building-related projects.
                                           4-4

-------
                                                                      Section 4: Industry Profile
       •  SIC Major Group 17-Construction Special Trade Contractors
          This group includes special trade contractors who undertake activities of a type that
          are specialized either in building construction or in both building and non-building
          projects.

       •  SIC Major Group 65-Real Estate
          This group includes real estate operators and the owners and lessors of real property,
          as well as buyers, sellers, developers, agents,  and brokers.

Major groups 15 and 16 are further defined by the type of construction performed. Table 4-2
provides a list of the more specific industry groups and industries that might perform land-
disturbing activities.

The focus of the regulation is on construction activities carried out by firms covered by NAICS
codes 233 and 234 or SIC codes 15 and 16. (As discussed in the preamble of the final rule,
Special  Trade Contractors, NAICS 238 or SIC 17, are typically subcontractors and not identified
as National Pollutant Discharge Elimination System permittees.) Furthermore, the residential,
nonresidential, and heavy construction subsectors receive the greatest emphasis because they
account for the vast majority of construction projects and are responsible for most of the land
disturbance in the United States.

  Table 4-2.1987 SIC industry groups performing construction activities that might disturb land
SIC Major Group 15
Industry Group 152: General Building Contractors - Residential
1521
1522
General Contractors - Single-family Houses
General Contractors - Residential Buildings, Other Than Single-family
Industry Group 153: Operative Builders
1531
Operative Builders
Industry Group 154: General Building Contractors - Nonresidential
1541
1542
General Contractors - Industrial Buildings and Warehouses
General Contractors - Nonresidential Buildings, Other Than Industrial
SIC Major Group 16
Industry Group 161 : Highway and Street Construction, Except Elevated Highways
1611
Highway and Street Construction, Except Elevated Highways
Industry Group 162: Heavy Construction, Except Highway and Street
1622
1623
1629
Bridge, Tunnel, and Elevated Highway Construction
Water, Sewer, Pipeline, and Communications and Power Line
Heavy Construction Not Elsewhere Classified
SIC Major Group 17
Industry Group 179: Miscellaneous Special Trade Contractors
1771
1794
Concrete Work
Excavation Work
SIC Major Group 65
Industry Group 655: Land Subdividers and Developers
6552
Land Subdividers and Developers, Except Cemeteries
    Source: U.S. Census Bureau 2008b
                                            4-5

-------
                                                                    Section 4: Industry Profile
 4.2.    INDUSTRY PRACTICES AND TRENDS

This section first provides a description of the types of C&D activities that result in the
disturbance of land and are responsible for the potential discharge of pollutants of concern to
surface waters. Then national estimates of the amount of disturbed acreage are provided.
Additional information including detailed descriptions of industry size and revenues is in EPA's
Economic Analysis for Final Effluent Guidelines and Standards for the Construction and
Development Category (USEPA 2009a).

   4.2.1.   OVERVIEW OF CONSTRUCTION LAND-DISTURBING ACTIVITIES

Constructing a building or facility involves a variety of activities, including the use of equipment
that alters the site's environmental conditions. Such changes include vegetation and top soil
removal, regrading, and drainage pattern alteration. The following provides a brief description of
typical land-disturbing activities at construction sites and the types of equipment employed.

      4.2.1.1.   Construction Site Preparation

Construction activities generally begin with the planning and engineering of the site and site
preparation. During this stage, mobile offices, which are usually housed in trailers, are
established on the construction site. The construction company uses such temporary structures to
handle vital  activities such as preparing and submitting applicable permits, hiring employees and
subcontractors, and ensuring that proper environmental requirements are met. The entire
construction yard is delineated with erosion and sediment controls (ESCs) installed and security
measures established. The latter includes installing fences and signs to warn against trespassing
and to mark dangerous areas. After the site is secured, equipment is brought to the site (and is
stored there throughout the construction period).

      4.2.1.2.   Clearing, Excavating, and Grading

Construction on any size parcel of land almost always calls for a remodeling of the earth (Lynch
and Hack 1984). Therefore, actual site construction begins with site clearing and grading.
Organic material—in particular,  roots—cannot support the weight of buildings and must be
removed from the top layer of ground. (Some developers stockpile the organic material for use
during the landscaping phase of construction rather than paying for it to be hauled from the site.)
Construction contractors must ensure that earthwork activities meet local, state, and federal
regulations for soil and erosion control, runoff, and other environmental controls. The size of the
site, extent of water present, soil types, topography, and weather determine the kinds of
equipment used in site clearing and grading (Peurifoy and Oberlender  1989). Material that will
not be used on the site must be hauled away by tractor-pulled wagons, dump trucks, or
articulated trucks (Peurifoy and Oberlender 1989).

Equipment used for lifting excavated and cleared materials include aerial-work platforms,
forwarders, cranes, rough-terrain forklifts, and truck-mounted cranes. In addition, track loaders
are used for digging and dumping earth (Caterpillar 2000; Reed Business  Information 2000;
Lynch and Hack 1984; Peurifoy  and Oberlender 1989).
                                           4-6

-------
                                                                     Section 4: Industry Profile
Excavation and grading are performed by several different types of machines. Those tasks can
also be done by hand, but that is generally more expensive (Lynch and Hack 1984). When
grading a site, builders typically ensure that new grades are as close to the original as possible, to
avoid erosion and stormwater runoff (Lynch and Hack 1984). Proper grading also ensures a flat
surface for development and drains water away from constructed buildings.

Excavation and grading equipment includes backhoes, bulldozers (including the versatile tracked
bulldozer), loaders, directional drilling rigs, hydraulic excavators, motor graders, scrapers, skid-
steer loaders, soil stabilizers, tool carriers, trenchers, wheel loaders, and pipeliners. Equipment
selection depends on functions to be performed and specific site conditions (Caterpillar 2000;
Reed Business Information 2000; Lynch and Hack 1984; Peurifoy and Oberlender 1989).
Therefore, multiple types of equipment are used throughout the clearing and grading process.

Self-transporting trenching machines, wheel-type trenching machines, and ladder-type trenching
machines are also used during site excavation. Self-transporting trenching machines are used to
create shallow trenches, such as for underground wire and  cables. This type of machine has  a
bulldozer blade attached to the front, is highly maneuverable, and can be used to dig narrow,
shallow trenches. Wheel-type trenching machines also dig  narrow trenches, most often for water
mains and gas and oil pipelines. Ladder-type trenching machines are used to dig  deep trenches,
such as for sewer pipes. These machines might have a boom mounted at the rear. Along the
boom are cutter teeth and buckets that are attached to chains. As the machine moves, it digs dirt
and moves it to the sides of the newly formed trench (Peurifoy and Oberlender 1989).

Power shovels can also be used for excavating soils. They  are used on all classes of earth that
have not been loosened. For solid rock, prior loosening is required. As materials  are excavated,
they are immediately loaded  onto trucks or tractor-pulled wagons and hauled from the site
(Peurifoy and Oberlender 1989). Hydraulic excavators, with either a front or a back shovel,  are
also used to dig into the earth and to load a hauling vehicle. There are several categories of
hydraulic excavators, including backhoes, back shovels, hoes, and pull shovels. Hydraulic
excavators are one of the most widely used types of excavating equipment because of their ease
of use and their ability to remove the earth that caves as it is moved. They are effective
excavating machines, and they are easy to use in terms of loading excavated soil  onto a hauling
vehicle (Peurifoy and Oberlender 1989).

Draglines, used to dig ditches or build levees, can transport soil within casting limits, thus
eliminating the need for hauling equipment (Peurifoy and Oberlender 1989). Draglines have a
bucket that hangs from a cable. The bucket is brought through the dirt and toward the operator
(Lynch and Hack 1984). Draglines can be used on both wet and dry ground and can dig earth out
of pits that contain water (Peurifoy and Oberlender 1989).  They are most useful for making large
cuts and channels below the level of the machine and for making valleys, mounds, slopes, and
banks (Lynch and Hack 1984). Draglines have a lower output than power shovels and do not
excavate rock as well as  power shovels (Peurifoy and Oberlender 1989).

Draglines can be converted to clamshells by replacing the dragline bucket with a clamshell
bucket. A clamshell is typically used for handling sand, gravel, crushed stone, sandy loam, and
other loose materials; it is not efficient in handling compacted earth, clay, or other dense
                                           4-7

-------
                                                                     Section 4: Industry Profile
materials. A clamshell is lowered into a material, and the bucket closes on the material. It is then
raised over a hauling vehicle and the materials are deposited (Peurifoy and Oberlender 1989).

Scrapers, either self-powered or drawn by tractors, dig and compact materials by taking up earth
from its underside with toothed scoops and loading it into hauling vehicles. Scrapers are useful in
removing earth and weak or broken rock and for excavating hills and rock faces. Some scrapers
are designed for long hauls; others with good traction are used on steep slopes (Lynch and Hack
1984).

A crawler tractor, which pulls a rubber-tired self-loading scraper, is often used for short-haul
distances. The crawler tractor uses a drawbar pull to load the scraper. It has good traction and
can operate on muddy  roads. It is, however, a slower vehicle and thus is more appropriate for
shorter hauls.

Wheel-type, tractor-pulled scrapers—which come in two- and four-wheel drive tractors—are
used for longer hauling distances. Unlike the crawler tractor-pulled scrapers,  the wheel-type,
tractor-pulled scrapers do not maintain good traction. Under such conditions, a helper tractor,
such as  a bulldozer, might be used (Peurifoy and Oberlender 1989).

All these machines shape and compact the earth, a crucial site preparation step. In addition,
earthwork activities might require that fill be brought in. In such cases, the fill must be spread in
uniform, thick layers and compacted to a specified density with an optimum moisture content.
Graders and bulldozers are the most common earth-spreading machines. Machines that compact
include  tractor-pulled sheep's foot rollers, smooth-wheel rollers, pneumatic rollers, and vibrating
rollers, among other equipment (Peurifoy and Oberlender 1989). Rollers and scarifiers are used
either to compact or to break up the ground (Lynch and Hack 1984).

To remove rock, it must first be loosened and broken up—usually through drilling or blasting.
Drilling equipment includes jackhammers, wagon drills, drifters, churn rills, and rotary drills;
each is designed to work on a specific size and type of rock. Dynamite and other explosives are
used to loosen rock (Peurifoy and Oberlender 1989).

After the materials have been excavated and removed and the ground cleared and graded, the site
is ready for construction.

   4.2.2.   CONSTRUCTION SITE SIZE CATEGORIES AND ESTIMATES  OF
            AMOUNT OF DISTURBED LAND

The regulatory options that EPA evaluated apply to construction sites of all types (i.e.,
residential, commercial, and industrial). Because the costs for ESC are largely driven by site size,
EPA must estimate the distribution of construction sites by size category, land use type, and
geographic region to estimate the total cost of the options. In addition, estimating distribution of
sites by  type allows EPA to estimate the cost to  each construction sector.
                                           4-8

-------
                                                                   Section 4: Industry Profile
      4.2.2.1.   National Estimates of New Development

EPA used the National Land Cover Dataset (NLCD) to estimate the amount of new developed
land occurring annually in the conterminous United States (CONUS) between 1992 and 2001
(see Table 3-1 and Appendix E). EPA's comparison of the 1992 and 2001 NLCD resulted in an
estimated annual rate of development of approximately 590,000 acres per year. By overlaying
geographic information system (GIS) layers of states and watersheds with the NLCD data, EPA
was able to estimate the annual number of acres of new development at both the state and
watershed level between 1992 and 2001 (For state-level annual estimates of new development,
see Table 4-6). EPA used the Reach File Version 1.0 (RF1) stream reach network and associated
watershed boundaries for the watershed-level estimates. EPA estimated annual development
rates between 1992 and 2001 for approximately 44,000 RF1 watersheds where a net increase in
urban land cover was identified. Approximately 7,800  additional RF1 watersheds showed either
no change or a minor decrease in urban land uses between 1992 and 2001. EPA scaled the
amount of development in each RF1 to the  year 2008 using historical construction spending data.
For additional details, see Economic Analysis for Final Effluent Guidelines and Standards for the
Construction and Development Category (USEPA 2009a). RF1 watersheds and stream reaches
are employed by the USGS SPARROW water quality model—the model EPA has selected to
assess potential environmental benefits of additional regulation of the industry (USGS 2008).
See the Environmental Impact and Benefits Assessment for Final Effluent Guidelines and
Standards for the Construction and Development Category (EPA 2009b) for additional details
and results of the water quality assessment  performed by EPA.

Because NLCD  data does not exist for Alaska, Hawaii, and the U.S. territories, EPA's analysis
does not include pollutant loading reduction or environmental benefits estimates for those areas.
However, the amount of development in those areas is expected to be low compared to the rest of
the United States; therefore, any errors in EPA's estimates are expected to be minor. EPA did
estimate costs for Alaska and Hawaii using economic data as an indicator of the amount of
construction activity occurring.

      4.2.2.2.   Model Project Distribution

EPA broadly characterized the acreage constructed annually into various future land uses and
construction project sizes. That characterization provides a basis for developing and then using
mathematical models that represent broad sectors of the industry to estimate compliance costs
and pollutant  loading reductions. EPA divided Notices of Intent (NOIs) into 36 groups based on
12 site size categories and three major land-use types (residential,  nonresidential, and
transportation). Projects were further subdivided into 12 categories of different durations.

The distribution of model projects into these categories was  developed by reviewing NOIs
submitted by permittees (see Appendix C, Analysis of Construction Industry Trends using
Notice of Intent Records). The NOI records were individually characterized on the basis of land
use, and the NOI records used provided site construction acreage and project duration
information. Individually, the site project models each represent large fractions of the
construction projects developed annually and cover the major project types in the C&D industry.
EPA used this model project matrix as a basis for estimating costs and pollutant removals for the
                                          4-9

-------
                                                                   Section 4: Industry Profile
industry. Table 4-3 shows the model project matrix developed. Table 9-2 shows the complete
model project matrix that includes the breakout by project sizes.

                           Table 4-3. Model project distribution

Size
category
(acres)
1-2.99
3-4.99
5-7.49
7.5-9.99
10-14.99
15-19.99
20-29.99
30-39.99
40-59.99
60-79.99
80-99.99
100 <
Median
size
(acres)
1.9
3.8
6.0
8.5
12.0
17.0
23.0
34.0
46.0
69.0
85.1
145.0
Total
Residential
Projects
4,914
3,693
1,992
1,680
2,421
1,556
1,810
984
921
373
242
344
20,930
Acres
9,337
14,033
1 1 ,952
14,280
29,052
26,452
41,630
33,456
42,366
25,737
20,594
49,880
318,769
Non residential
Projects
23,237
12,410
6,709
3,579
4,084
2,102
2,078
865
847
366
213
356
56,846
Acres
44,150
47,158
40,254
30,422
49,008
35,734
47,794
29,410
38,962
25,254
18,126
51,620
457,892
Transportation
Projects
2,417
1,298
770
494
548
272
363
128
180
52
56
118
6,696
Acres
4,592
4,932
4,620
4,199
6,576
4,624
8,349
4,352
8,280
3,588
4,766
17,110
75,988
National
Projects
30,568
17,401
9,471
5,753
7,053
3,930
4,251
1,977
1,948
791
511
818
84,472
Acres
58,079
66,124
56,826
48,901
84,636
66,810
97,773
67,218
89,608
54,579
43,486
118,610
852,650
 4.3.    REFERENCES

Caterpillar. 2000. Caterpillar, Inc., Peoria, IL. .

Lynch K., and G. Hack. 1984. Site Planning. 3rd ed. The MIT Press, Cambridge, MA.

Peurifoy, R.L., and G.D. Oberlender. 1989. Estimating Construction Costs. 4th ed. McGraw Hill
   Book Company, New York.

Reed Business Information. 2000. Construction Equipment. Oak Brook, IL, .

U.S. Census Bureau. 2008a. 2007 North American Industry Classification System (NAICS).
   U.S. Department of Commerce, U.S. Census Bureau, Washington, DC.
   . Updated March 28, 2008; accessed April
    15,2008.

U.S. Census Bureau. 2008b. 2007North American Industry Classification System (NAICS).
   Concordances: 2007 NAICS to 2002 NAICS.xls and 2002 NAICS to 1987 SIC.xls.
   U.S. Department of Commerce, U.S. Census Bureau, Washington, DC.
   . Accessed
   November 7, 2008.
                                         4-10

-------
                                                                  Section 4: Industry Profile
USEPA (U.S. Environmental Protection Agency). 2009a. Economic Analysis for Final Effluent
   Guidelines and Standards for the Construction and Development Category (EPA-821-R-09-
   011). U.S. Environmental Protection Agency, Office of Water, Washington, DC.

USEPA (U.S. Environmental Protection Agency). 2009b. Environmental Impact and Benefits
   Assessment for Final Effluent Guidelines and Standards for the Construction and
   Development Category (EPA-821-R-09-012). U.S. Environmental Protection Agency, Office
   of Water, Washington, DC.

USGS (U.S. Geological Survey). 2008. The SPARROW Surface Water-Quality Model: Theory,
   Application and User Documentation. U.S. Geological Survey Techniques and Methods,
   Book 6, Section B, Chapter 3. U.S. Geological  Survey, Reston, VA.
                                         4-11

-------
                                                   Section 5: Selection of Pollutants for Regulation
5.   SELECTION OF POLLUTANTS FOR REGULATION

 5.1.    INTRODUCTION

Construction and development (C&D) activities can generate a broad range of environmental
impacts by introducing new sources of contamination and by altering the physical characteristics
of the affected land area. In particular, those activities can result in both short- and long-term
adverse effects on surface water quality in streams, rivers, and lakes in the affected watershed by
increasing the loads of various pollutants in receiving waterbodies, including sediments, metals,
organic compounds, pathogens, and nutrients. Ground water also can be adversely affected
through diminished recharge capacity. Other potential effects include the physical alteration of
existing streams and rivers due to the excessive flow and velocity of stormwater runoff.

Construction activities typically involve excavating and clearing existing vegetation. During the
construction period, the affected land is usually stripped and the soil compacted, leading to the
potential for increased stormwater runoff and high rates of erosion. If the denuded and exposed
areas contain hazardous contaminants or pollutants (either naturally  occurring or from previous
land uses), they can be carried at increased rates to surrounding waterbodies by stormwater
runoff. Although the denuded construction site is only a temporary state (usually lasting less than
6 months), the landscape is permanently altered even after the land has been restored by
replanting vegetation.

Pollutants associated with C&D stormwater discharges can adversely affect the environment in a
number of ways.  Potential effects include impairment of water quality, destruction of aquatic life
habitats, and enlargement of floodplains. The Environmental Impact andBenefits Assessment for
Final Effluent Guidelines and Standards for the Construction and Development Category (EPA
2009b) discusses the potential affects of C&D stormwater runoff on the environment. The
discussion in the  remainder of this section focuses on those pollutants generated at a site during
active construction.

 5.2.    POLLUTANTS ASSOCIATED WITH CONSTRUCTION AND LAND
         DEVELOPMENT STORMWATER RUNOFF

A number of pollutants are associated with C&D stormwater runoff. The descriptions of
pollutants in this  subsection do not represent the complete suite of contaminants that can be
found in the runoff but focus instead on those that are known to be the most prevalent and of
greatest concern to the environment. Those pollutants include sediment, metals, polycyclic
aromatic hydrocarbons (PAHs), oil and grease, and pathogens. A more thorough discussion of
pollutants and pollutant sources are in the Environmental Assessment document.

    5.2.1.  SEDIMENT

Sediment is an important and ubiquitous constituent in urban stormwater runoff. Surface runoff
and raindrops detach soil from the land surface, resulting in sediment transport into streams and
rivers. Sediment and turbidity can affect habitat, water quality, temperature, pollutant transport,
and can cause sedimentation in downstream receiving waters. The effects of excess sediment in
                                          5-1

-------
                                                     Section 5: Selection of Pollutants for Regulation
the water include direct physical effects such as reducing visibility and light in the water column,
physical abrasion of plant surfaces, clogging gill openings, and entombing eggs and fry in redds.
Effects can also be indirect, as in changes to the chemical composition (e.g., pH, hardness) of the
water, light penetration or turbidity, and temperature profile, which in turn affect primary
productivity with repercussions in terms offish behavior, and overall community profiles and
trophic structure.

Sediment level measurement can be divided into several distinct subgroups:

       •   Total suspended solids (TSS) are a dry-weight measure of the suspended parti culate
          material in water. Measuring TSS in urban stormwater allows for estimation of
          sediment transport, which can have significant effects locally and in downstream
          receiving waters. TSS is typically measured in milligrams per liter (mg/L).

       •   Turbidity is a measure of the amount of solids and other materials in the water.
          Turbidity readings are somewhat dependent on particle size, shape, and color.
          Turbidity is typically measured in nephelometric turbidity units (NTUs). Turbidity
          can exhibit control over biological functions, such as the ability of submerged aquatic
          vegetation to receive light.

       •   Total dissolved solids are a measure of the dissolved constituents in water and are a
          primary indication of the purity of drinking water.

       •   Settleable solids, expressed as milliliters per liter (mL/L), are a measure of the solids
          that will settle to the bottom of a cone-shaped container (called an Imhoff cone) in a
          60-minute period. Settleable solids are primarily a measure of particles that can be
          removed by sedimentation.

       •   Suspended sediment concentration (SSC) is a measure similar to TSS; however, there
          are differences in the two analytical methods. SSC is determined by measuring the
          dry weight of all sediment from a known volume of sample. TSS is measured by
          filtering a subsample and measuring the weight of the dried solids. SSC and TSS
          values from the same  sample can vary greatly, especially as the fraction of sand-sized
          particles in a sample increases. This is primarily because of the subsampling
          procedure involved in TSS calculations where typically  a pipette is used to withdraw
          a subsample from the sample container. That procedure  might not capture a
          representative fraction of larger particles in the subsample. The U.S. Geological
          Survey  has analyzed differences attributable to the two methods and determined that
          SSC is a more appropriate measure of the mass of solids in natural-water  samples
          (Gray et al. 2000). That might also apply to stormwater discharges, especially if a
          significant fraction of sand-sized particles are present.

Erosion from construction sites can be a significant  source of sediment pollution to nearby
streams. A number of studies have  shown high concentrations of TSS in uncontrolled runoff
from construction sites, and results from the studies are summarized in Table 5-1. One study,
conducted in 1986, calculated that construction sites are responsible for an estimated export of
80 million tons of sediment into receiving waters each year (Goldman et al.  1986). On a unit area
basis, construction sites can export sediment at 20 to 1,000 times the rate of other land uses
(Schueler 1997).
                                           5-2

-------
                                                     Section 5: Selection of Pollutants for Regulation
          Table 5-1. Studies of uncontrolled soil erosion as TSS from construction sites
Site
Seattle, Washington
SR204
Mercer Island
RT1
RT2
SB1
SB2
SB2
SB4
Pennsylvania Test Basin
Georgia Model
Maryland Model
Uncontrolled Construction Site Runoff
(Maryland)
Hamilton County, Ohio
Mean TSS (mg/L)
Mean inflow TSS
concentration
(mg/L)
17,500
3,502
1,087
359
4,623
625
415
476
2,670
9,700
3,000
3,000
4,200
2,950
3,860
Source
Horneretal. 1990
Horner et al. 1990
Horner et al. 1990
Schueler and Lugbill 1990
Schueler and Lugbill 1990
Schueler and Lugbill 1990
Schueler and Lugbill 1990
Schueler and Lugbill 1990
Schueler and Lugbill 1990
Jarrett 1 996
Sturm and Kirby 1991
Barfield and Clar 1985
York and Herb 1978
Islam etal. 1998
N/A
    N/A - Not Applicable

For summaries of studies with monitoring or modeling data and annotated bibliographies for the
journal articles and professional conference proceedings that the U.S. Environmental Protection
Agency (EPA) reviewed, see Document Control Numbers (DCNs) 44321 and 43114.

    5.2.2.   METALS

Many toxic metals can be found in urban stormwater, although typically only metals such as
zinc, copper, lead, cadmium, and chromium have been identified in the literature as being of
primary concern because of their prevalence in urban stormwater runoff and their potential for
environmental harm. Those metals are generated by motor vehicle exhaust, weathering of
buildings, burning fossil fuels, atmospheric deposition, and  other common urban activities.

Metals can bioaccumulate in stream environments, resulting in plant growth inhibition and
adverse health effects on bottom-dwelling organisms (Masterson and Bannerman 1994).
Generally the concentrations found in urban stormwater are not high enough for acute toxicity
(Field and Pitt 1990). Rather,  it is the cumulative effect of the concentration of the metals over
time and the buildup in the sediment and animal tissue that are of greater concern.

Construction sites are not thought to be important sources of metals contamination. Runoff from
such sites could have high metals contents if the soil is already contaminated, or if metals are
naturally present in site soils.  Imported fill can also be a source of contamination. Construction
activities alone do not usually result in significant metals contamination, although there is little
data available  on this subject.
                                           5-3

-------
                                                     Section 5: Selection of Pollutants for Regulation
    5.2.3.   PAHS, AND OIL AND GREASE

Petroleum-based substances such as oil and grease and PAHs are found frequently in urban
stormwater runoff. Many constituents of PAHs and oil and grease, such as pyrene and
benzo[b]fluoranthene, are carcinogens and toxic to downstream biota (Menzie-Cura &
Associates 1995). Oil and grease and PAHs normally travel attached to sediment and organic
carbon. Downstream accumulation of these pollutants in the sediments of receiving waters such
as streams, lakes, and estuaries is of concern.

Construction activities during site development are not believed to be major contributors of these
contaminants to stormwater runoff. Improper operation and maintenance of construction
equipment at construction sites, as well as poor housekeeping practices (e.g., improper storage of
oil and gasoline products and construction materials),  could lead to leakage or spillage of
products that contain hydrocarbons.

    5.2.4.   PATHOGENS

Microbes are commonly found in urban stormwater. Although not all microbes are harmful,
several species such as the pathogens Cryptosporidium and Giardia can directly cause diseases
in humans. The presence of bacteria such as fecal coliform bacteria, fecal streptococci, and
Escherichia coli  (i.e., E. coli) indicates a potential health risk. High levels of these bacteria can
result in beach closings, restrictions on shellfish harvest, and increased treatment for drinking
water to decrease the risk of human health problems.

Construction site activities are not believed to be major contributors to pathogen contamination
of surface waters. The only potential known source of pathogens from  construction sites are
portable septic tanks used by construction workers. Those systems, however, are typically  self-
contained; although leaks or spills could result in releases.

 5.3.    SELECTION OF POLLUTANTS FOR REGULATION

When determining which pollutants to consider for regulation, EPA applied the following
priorities for discharges from the C&D industry:

       •  Focus on pollutants directly attributable to the industry, using indicator pollutants
          where necessary

       •  Focus on pollutants most commonly encountered under most  settings, (i.e., not  to
          preconstruction site contamination issues or accidental discharges)

       •  Focus on pollutants that are most manageable given the current suite of available
          technologies

In support of the 2002 and 2004 regulatory efforts, EPA conducted an extensive evaluation of the
literature to identify pollutants present in stormwater discharges from C&D sites. While the
literature contains extensive information on pollutants present in stormwater discharges from
urban areas, there were little data available on pollutants present in stormwater discharges from
construction sites during the active phase of construction other than for sediment, TSS, and
turbidity. That is not surprising, because construction  site stormwater management is primarily
                                           5-4

-------
                                                     Section 5: Selection of Pollutants for Regulation
concerned with controlling solids from exposed soil areas. There is the potential for other
pollutants to be discharged from construction sites depending on factors such as prior land uses.
For example, if the prior land use was agriculture, the potential exists for discharge of pollutants
such as nutrients and pesticides. Likewise, areas of redevelopment that occur on sites where
previous land uses included industry could discharge pollutants such as organics and metals. In
addition, pollutants such as metals and nutrients can be present in native site soils and could be
discharged from construction sites. Also, high pH can result from stormwater being exposed to
freshly placed concrete. However, EPA was not able to identify sufficient data in the literature to
warrant developing controls specific to pollutants other than sediment, TSS, and turbidity in
stormwater discharges from active construction sites. Although EPA identified other pollutants
of concern for the industry, EPA did not develop regulatory options specifically targeted at
controlling each of these individual pollutants. The Environmental Assessment contains a more
thorough discussion of pollutants found in stormwater discharges from construction sites.

Instead, EPA chose to develop regulatory options using an indicator pollutant, turbidity. While
turbidity might not correlate well with TSS,  designing management systems for controlling
turbidity will likely result in control of other pollutants such as TSS, nutrients, and metals that
are present in the solid-phase (attached to sediments). In addition, turbidity, unlike TSS, can be
measured with relative ease in the field using hand-held turbidity meters or automated, in-line
turbidity meters. An in-line turbidity meter,  coupled with a data logger, can offer real-time data
on turbidity levels in stormwater discharges.

Particles that contribute to turbidity can be of such a fine grain that they will not be removed by
the mechanisms whereby most best management practices operate, mainly settling and filtration.
Hence, the options developed for the final rule focus on passive and active treatment of
stormwater runoff using polymers to remove turbidity, as well as TSS and other pollutants.
Section 7 discusses technologies designed to reduce and remove such fine colloidal particles.

 5.4.    REFERENCES

Barfield, B.J., and M. Clar. 1985. Development of New Design Criteria for Sediment Traps and
   Basins. Prepared for the Maryland Resource Administration, Annapolis, MD.

Goldman, S.J., K. Jackson, and T.A. Bursztynsky.  1986. Erosion and Sediment Control
   Handbook. McGraw-Hill, New York.

Gray, J.R., D.  Glysson, L.M. Turcios, and G.E. Schwarz. 2000.  Comparability of Suspended-
   Sediment Concentration and Total Suspended Solids Data. Water-Resources Investigation
   Report 00-4191. U.S. Geological Survey, Reston, VA.

Field. R., and R.E. Pitt. 1990. Urban Storm-Induced Discharge Impacts: U.S. Environmental
   Protection Agency Research Program Review. Water Science and Technology 22(10/11): 1-7.

Horner, R.R., J. Guedry, and M.H. Kortenhof. 1990. Improving the Cost Effectiveness of
   Highway Construction Site Erosion and Pollution Control. Washington State Transportation
   Center and the Federal Highway Administration.  Seattle, WA.
                                           5-5

-------
                                                   Section 5: Selection of Pollutants for Regulation
Islam, M.M., D. Taphorn, andH. Utrata-Halcomb. 1998. Current Performance of Sediment
   Basins & Sediment Yield Measurement of Construction Sites in Unincorporated Hamilton
   County, Ohio. Prepared for Hamilton County Soil and Water Conservation District,
   Cincinnati, OH.

Jarrett, A. 1996. Sediment Basin Evaluation and Design Improvements. Prepared for Orange
   County Board of Commissioners, Hillsborough, NC, by Pennsylvania State University, State
   College, PA.

Masterson, J.P., and R.T. Bannerman. 1994. Impacts of Stormwater Runoff on Urban Streams in
   Milwaukee County, Wisconsin. In Proceedings of American Water Resources Association,
   National Symposium on Water Quality, Nov. 6-10,  1994, Chicago, IL, pp. 123-133.

Menzie-Cura & Associates. 1995. Measurements and Loadings ofPolycyclic Aromatic
   Hydrocarbons (PAH) in Stormwater, Combined Sewer Overflow. MBP-95-06. Massachusetts
   Bay Program, MA.

Schueler, T. 1997. Impact of Suspended and Deposited Sediment. Article 14 in The Practice of
   Watershed Protection, eds. T.R. Schueler and H.K. Holland, pp. 64-65. Center for
   Watershed Protection, Ellicott City, MD.

Schueler, T., and J.  Lugbill. 1990. Performance of Current Sediment Control Measures at
   Maryland Construction Sites. Metropolitan Washington Council of Governments,
   Washington, DC.

Sturm, T.W., and R.E. Kirby. 1991. Sediment Reduction in Urban Stormwater Runoff from
   Construction Sites. Georgia Institute of Technology, Atlanta, GA.

USEPA (U.S. Environmental Protection Agency). 2009b. Environmental Impact and Benefits
   Assessment for Final Effluent Guidelines and Standards for the Construction and
   Development Category (EPA-821-R-09-012). U.S. Environmental Protection Agency, Office
   of Water, Washington, DC.

York T.H., and WJ. Herb. 1978. Effects of Urbanization and Stream/low on Sediment Transport
   in the Rock Creek andAnacostia River Basins. Montgomery County, MD, 1972-1974.
   U.S. Geological Survey Professional Paper No. 1003. U.S. Geological Survey, Reston, VA.
                                          5-6

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
6.   LIMITATIONS AND STANDARDS: DATA SELECTION AND
     CALCULATION

 6.1.    INTRODUCTION

This section describes the data selection and statistical methodology used by the
U.S. Environmental Protection Agency (EPA) in calculating the limitations and standards for the
Construction and Development (C&D) point source category. As is the case for most effluent
limitation guidelines and standards, the effluent limitations and standards are based on long-term
average effluent values and variability factors that account for variation in treatment performance
within a particular treatment technology over time. For simplicity, the following discussion
refers only to effluent limitations guidelines; however, the discussion also applies to new source
standards.

EPA is promulgating a daily maximum limitation for turbidity,  and Section 6.2 briefly describes
the pollutant parameter.  Section 6.3 provides an overview of EPA's criteria typically used to
select data sets used as the basis for limitations. Section 6.4 describes the available discharge
data sets that met the  criteria. Section 6.5 describes the data sets that were excluded as a result of
applying the criteria.  Section 6.6 verifies that the individual values within the retained data sets
also are appropriate as the basis of the limitation. Sections 6.7 and 6.8 provide summaries of the
data before and after averaging to obtain daily values. Section 6.9 provides an overview of the
limitations, percentile basis, and calculations. Section 6.10 describes the engineering review of
the limitations. Sections 6.11 and 6.12 discuss issues related to monitoring and compliance with
the limitation. Section 6.13 summarizes the steps used to calculate the limitations. Section 6.14
provides references.

In the proposed rule, EPA also was considering a limitation on pH to protect against extreme
acidity or alkalinity. EPA has not promulgated a pH limitation for the final rule.

 6.2.    TURBIDITY

As described in Section 5 and in more detail in the Environmental Assessment, there are a
number of pollutants  associated with discharges from C&D sites. EPA is promulgating effluent
limitations for turbidity, as an indicator of the presence of those pollutants being discharged from
the C&D site. Turbidity is a simple measurement that requires only the use of a turbidimeter and
can be conducted in the field. Readings are made in nephelometric turbidity units or NTUs.
Turbidity measurement does not require any sample preparation, other than shaking the sample
bottle well before analysis. The sample is simply poured into a glass tube and placed inside the
calibrated instrument. The result is read directly from the instrument. There are also a variety of
digital turbidity probes, which can be coupled with a microprocessor controlled data logger and
combination meter/data loggers available that can be used to automatically read and log turbidity
values in-situ.
                                           6-1

-------
                                     Section 6: Limitations and Standards: Data Selection and Calculation
 6.3.    OVERVIEW OF DATA REVIEW AND CRITERIA

To develop a limitation, EPA generally seeks to obtain as much monitoring data as possible on
the effectiveness of the different treatment options it evaluates, through solicitation of
information from the public and industry. Here, EPA received data from a number of treatment
technology vendors, but no data from the regulated industry. As described in Sections 6.4, 6.5,
and 6.6, EPA qualitatively reviewed all the data before selecting a large subset to calculate the
limitations. In selecting the data, EPA applied the following criteria in determining if the data
were appropriate to use as the basis for the final rule. In its rulemakings for other industries, EPA
has used the same or similar criteria to develop the limitations and standards.

One criterion requires that the influents and effluents from the treatment components represent
typical wastewater (or in the  case of the C&D industry, stormwater) from the industry, with no
significant incompatible wastewater from other sources (e.g., sanitary wastes). Application of
this criterion results in EPA selecting only those sites where the commingled wastewaters did not
result in substantial dilution,  more concentrated wastewaters, or wastewaters with different types
of pollutants than those generated by C&D wastewater.

A second criterion ensures that the pollutants were present in the influent at sufficient
concentrations to evaluate treatment effectiveness. By verifying that influent includes
measureable solid content, EPA ensures that its limitations resulted from treatment and not
simply the absence of turbidity in the wastestream.

A third criterion generally requires that the system demonstrate good operation of the model
treatment technology. EPA determines whether a system meets this criterion on the basis of
documentation about the system installed at the site, discussions with site management,
evaluation of site diagrams, and comparison to the performance of treatment systems at other
sites. In addition, because control of turbidity at construction sites is a function of up-slope
erosion and sediment controls, as well as proper application and sizing of passive treatment
controls, EPA also evaluated whether the overall site controls (if information was available) were
representative of BAT, and whether controls were adequately sized, operated and maintained and
whether a particular site would represent typical site conditions. In general, EPA reviewed this
information to determine if the system was adequately sized; properly operated and maintained;
and whether the resulting data represent typical site conditions. As a result of these
communications and reviews, EPA determined that some data were representative of normal
operating conditions for the facility and that the level of treatment was adequate, and excluded
data that reflected a technology application that did not coincide with the selected BAT and/or
where the overall site controls were deemed to be inadequate.

A fourth criterion typically requires that the data cannot represent periods of treatment upsets or
shut-down periods. This criterion sometimes results in the exclusion of periods when the site first
starts operating the equipment (start-up). Although this criterion is more applicable to
wastewater treatment than stormwater management practices at construction sites, there are some
similarities in this case (for example, non-optimized dosage rates). As result of this criterion,
EPA could exclude certain time periods and other outliers in the data from an otherwise well-
operated site.
                                           6-2

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
EPA has not included the size of the site as a criterion, because the site size and associated runoff
volumes determine the design and size of the management practices, rather than its performance.

 6.4.    DATA SELECTED AS  BASIS FOR LIMITATIONS

As a consequence of applying the four criteria, EPA selected only data that were representative
of the model technology, which in this case is polymer-aided settling. All of the sites used by
EPA as the basis for the limitation employed either polymer-aided settling in ponds (using either
chitosan acetate, chitosan lactate or PAM) or polymer-aided settling/filtration using check dams.
EPA is confident that the resulting database fully characterizes the performance of the model
technology for all C&D sites subject to the limitation for several reasons:
       1.  Theoretically, there is no reason that the technology cannot be applied everywhere. It
          is a simple technology that only requires appropriate sizing of the ponds and
          conveyances, applying the correct polymer, using the appropriate dosing schedule,
          and conducting needed maintenance activities.
       2.  Different soil types and rainfall amounts are managed by sizing the ponds and
          conveyances properly, providing adequate detention time for settling, applying the
          correct polymer, and using the  appropriate dosing schedule. For example, if the soil
          (e.g., clay) does not readily infiltrate rainfall, then the storage volume would need to
          be larger than one at a site with more porous soil. In another example, if the site has a
          steeper slope, then the site might contain more check dams per channel than a more
          level site.
       3.  Size of the site is managed by placing the appropriate number of systems on the site
          and/or sizing the ponds and conveyances properly.
       4.  The model technology, polymer aided settling, has  widespread use across a wide
          range of industries (e.g., POTWs, drinking water treatment, industrial wastewater
          treatment) for solids control. In the industries which have used this technology for
          decades, it has been successful. It is less prevalent in the C&D industry because
          managing turbidity in stormwater discharges has not been required until recently.

EPA's database of performance data includes more than 29,000 turbidity measurements that
were used as the basis for the limitation. The data were from 25 treatment  systems at 9 sites in
three states and covers both the Eastern and Western Unites States and a range of construction
types. The data were provided by two commercial firms (Cascade EcoSolutions and Clear Water
Compliance Services) and researchers at North Carolina  State  University. Table 6-1 identifies
the data sources, the site name or location, and the abbreviations used to identify the systems at
those sites. These abbreviations are used throughout Chapter 6 and the data listings in
Appendix F.
                                           6-3

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
 Table 6-1. Data sources and site identification for systems using EPA's model technology basis
Source
Cascade
EcoSolutions
Clear Water
Compliance Services,
Inc.
NCSU Research
(McLaughlin, et al)
Site name or location
Beacon Hill Reservoir Burying Project,
Seattle, WA
Brightwater Waste Water Treatment Plant in
King County, WA
Sea-Tac Airport in King County, WA
Sound Transit Central Link Light Rail
Tacoma/Seattle, WA
Beacon Hill Reservoir Burying Project,
Seattle, WA
Sea-Tac Airport in King County, WA
Springville, NY highway widening
Redmond, WA residential project
North Carolina mountain roadway project in
2006-7
North Carolina mountain roadway project in
2008
Abbreviation
BHRBP
BWWTP
SEAAIR
STCLLR
BHRBP2 *
KC-variations
NY
Red. East,
Red .West
NCR.1
NCR.2
NC.Road
Construction
type
Other
Commercial
Transportation
Linear
Linear
Transportation
Transportation
Residential
Transportation
Transportation
' Because BHRBP2 is a more complete data set, it was used to calculate the long-term average and variability factor for this site.
   6.4.1.   CASCADE ECOSOLUTIONS

Cascade EcoSolutions is a vendor that provides the chitosan-based flocculants used by service
providers to treat turbid stormwater. For the proposed rule, Cascade EcoSolutions provided
influent and effluent data for the advanced treatment system (ATS) at six sites in Washington
State. After the proposal, EPA contacted the vendor for more details about the influent data. The
vendor confirmed that the influent to ATS for four of the sites was effluent from passive
treatment systems, that is, the model technology for the final rule. As a result, EPA concluded
that the data from the four sites were appropriate to use in developing the final limitation that is
based on passive treatment systems. The four sites are all in the Seattle area in Washington:
       •  Beacon Hill Reservoir Burying Project (BHRBP)

       •  Brightwater Wastewater Treatment Plant (BWWTP)

       •  Sea-Tac Airport (SEAAIR)

       •  Sound Transit Central Link Light Rail (STCLLR)
For three sites, BHRBP, BWWTP, and STCLLR., the vendor provided turbidity measurements
via a ChitoVan Performance Review Data Set (Cascade EcoSolutions 2008). For SEAAIR, the
vendor provided detailed information about the treatment and sites in an engineering report
(Minton 2006) and a separate Engineering Report Data file (Cascade EcoSolutions 2008). The
report identifies the SEAAIR project as supporting the construction of Sea-Tac's third runway.
                                          6-4

-------
                                   Section 6: Limitations and Standards: Data Selection and Calculation
   6.4.2.   CLEAR WATER COMPLIANCE SERVICES, INC.

Clear Water Compliance Services Inc. provides comprehensive water treatment services
including the design, installation, and monitoring of treatment systems for stormwater and
construction runoff. The company provided data for the following 20 systems using the model
technology:

       •   16 systems at the third runway at Sea-Tac airport. These data sets are identified as KC
          for King County followed by the Site number (or pond), and then system number. For
          example, KC1.1 is the first system at site 1.

       •   One system at a highway widening site in Springville, New York. This data set is
          identified as "NY."

       •   Two systems at an 18-acre residential development project in Redmond, Washington.
          They installed a 250 gallons per minute (gpm) treatment system at the West Basin
          and a 500 gpm treatment system at the East Basin, both of which operated over a 2
          year period. The data sets are identified as Red.West and Red.East.

       •   One system at a large infrastructure project that is part of the City of Seattle's
          Reservoir Burying Program. The project involved demolition of an existing reservoir
          and re-building of a large vault-reservoir that was then buried below the surface and
          had a park built over the top. The site area for this site was approximately 22 acres.
          The data set is identified as BHRBP2 to distinguish it from the Cascade EcoSolutions
          data for this site.

   6.4.3.   RESEARCH BY NORTH CAROLINA STATE UNIVERSITY

Dr. Richard A. McLaughlin and others at North Carolina State University studied stormwater
runoff from three systems for erosion and sediment control on two roadway projects in the North
Carolina mountains. EPA determined that three of the systems were consistent with its model
technology:

       •   NC.Road is described in a paper Target Turbidity Limits for Passive Treatment
          Systems (McLaughlin No Date) that provides data for from September 2008 to
          January 2009.

       •   NCR. 1 and NCR.2 are described in the Journal of Soil and Water Conservation
          (McLaughlin 2009). The systems were installed as part of a university research
          project. The measurements were collected from June 2006 to October 2006 at road
          widening and paving projects in the North Carolina mountains.

 6.5.    SYSTEMS EXCLUDED AS BASIS FOR LIMITATION

EPA excluded data from all systems that did not meet the criteria described in Section 6.3. Table
6-2 summarizes the system exclusions and EPA's rationale for each. In all cases, the data
represented a different technology, such ATS, than the model technology basis for the regulation.
                                         6-5

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
                     Table 6-2. Systems Excluded as Basis for Limitation
Source
Clear Creek Systems,
Inc. (proposal data)
Cascade EcoSolutions
(proposal data)
Oregon Department of
Environmental Quality
(proposal data, Jurries
no date)
Resource Planning
Associates (41 1 07) and
Minton(41108)
Clear Water
Compliance Services
Site name or location
California, Oregon, and
Washington
WSDOT SR-522 Road
Improvement Project (Elliot
Road)
Lakeside
West Linn Corporate Park
Hoodview Estates
Six systems at a commercial
site in Redmond, CA
Morrisville, NC, was a
commercial site with a runoff
area of 82 acres and two
treatment systems.
Abbreviation
2
3
4
6
8
11
BZR08
SC05
SC08
ELLRD
LSIDE
WLCPO
HEO
RED.1-RED.6
not incorporated into
EPA's databases
Reason for exclusion
The effluent data are from
ATS. Influent data is
influent into the
pretreatment pond. The
vendor did not collect
data after the
pretreatment pond but
before ATS filtration, and
thus, the influent does not
represent EPA's model
technology.
The effluent data
represent ATS which is
not EPA's model
technology.
The effluent data
represent ATS. The
influent data were not
pretreated with EPA's
model technology.
The effluent data
represent ATS. The
influent data were not
pretreated with EPA's
model technology.
These sites used batch
treatment in cells, which
is more extensive than
EPA's model technology.
Data were provided only
as minimum and
maximum values which
could not be used to
calculate daily averages.
 6.6.    APPLICATION OF CRITERIA TO LIMITATION DATA SETS

After excluding data sets from systems that were not representative of the model technology for
C&D wastewaters, EPA performed a final review of the individual data points from the systems
that it had selected as the basis of the limitations. For this review, EPA returned to the criteria
identified in Section 6.3 and applied each one to the data sets identified in Table 6-1.

The first criterion ensures that the wastewater contains primarily stormwater associated with
C&D operations. EPA considered two aspects, post-paving conditions and recirculation, in
evaluating the data for this criterion:

       •  Post-paving conditions: The journal article describing NCR.l and NCR.2 included
          data after the pavement was complete. EPA determined that post-paving conditions
                                          6-6

-------
                                     Section 6: Limitations and Standards: Data Selection and Calculation
          were not representative of C&D discharges, and excluded them as the basis of the
          limitation.

       •  Recirculation: EPA has not excluded any data from systems that recycled effluent.
          Although some of the sites recycled water after ATS filtration to the pretreatment
          ponds, EPA determined that this practice would have little overall effect on turbidity
          within the basin because the recycle flow rate is small in comparison to the storage
          volumes contained in the ponds. In addition, other inputs into the pond from surface
          runoff would be expected. Before determining that the data of such systems should be
          included as the basis of the limitation, EPA considered the effect of recirculation on
          treatment performance. If recirculation had an impact, it would be expected to dilute
          the resulting effluent to lower concentration levels.

To evaluate whether this was the case, EPA evaluated the SEAAIR data set, which was one data
set that indicated when recirculation occurred. EPA evaluated the individual measurements and
the daily averages derived from them.

       •  Individual measurements: Of the 31 recirculation events (some had multiple events
          per day), 16 had increased turbidity in the subsequent reading, and 14 had less (one
          reading was the final of the day and therefore had no subsequent reading). EPA then
          considered effluent concentration reported for the reading during recirculation and the
          following measurement. The effect is more pronounced  with increased turbidity for
          30 measurements and less turbidity for 18. Both evaluations indicate that the effluent
          concentrations tended to increase  after circulation at this location.

       •  Daily averages: On the five days when recirculation occurred, the observed daily
          averages were greater on four days and lower on one when including the recirculation
          events. In each case, the change was less than 6 percent.

Because the effect of recirculation was the opposite from what would be expected if dilution
were the only influence (i.e., the effluent turbidity values were generally more concentrated (i.e.,
higher) not dilute), EPA concluded that recirculation did not appear to have a significant  effect
on effluent concentrations. Instead, the higher turbidity values might be from sediment
resuspension in the pretreatment pond due to the turbulence caused by the recycled water. In
addition, additional inputs to the pond from surface runoff, groundwater flows, and activities
such as dewatering operations on the site could have contributed additional volume to the ponds,
which would also affect the effluent turbidity. Because the recirculation did not appear to
significantly influence effluent turbidity, EPA determined that it was appropriate to retain all of
the data, including data from time periods where recirculation was  occurring, in its limitation
data set.

The second criterion ensures that the effluent levels are the result of treatment rather than dilute
influent. Although most of the effluent data were not paired with the corresponding influent,
based upon its review of the existing data  and other sources that describe C&D stormwater  (see
Section 5), EPA is confident that turbidity would have been present at any site, and thus,  has not
used this criterion to exclude any data set. Influent into the passive  treatment system is affected
by many parameters. EPA examined the data in several ways.
                                           6-7

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
       •  EPA considered whether the influents were likely to be more concentrated than the
          effluents. McLaughlin (2009) provides turbidity concentrations from a section of the
          roadway construction project with standard best management practices (BMP) for the
          NCR.l site. Both sections of the project would be expected to have similar soil types,
          rainfall, and other characteristics. By comparing data from the two sections, EPA was
          able to obtain a lower bound of the difference between untreated stormwater and
          PAM-treated stormwater. That is, EPA would expect even more of a difference than
          the results using BMP data. Considering only the days when measurements were
          made at both sections, the BMP site removed turbidity to a level of 3669 NTU, while
          the model technology (fiber check dams with PAM) removed turbidity to a level of
          26.8 NTU for a removal of 99 percent. EPA performed a similar comparison for the
          NCR.2 data and found a removal of 97 percent. This finding demonstrates that
          influent would be substantially greater than effluent from the model technology.

       •  EPA examined the literature to determine turbidity  in stormwater generated at
          construction sites. EPA evaluated a variety of literature sources, and summarized
          numerous studies evaluating the effectiveness of sediment basins, which are among
          the most common management practices used at construction sites. The literature
          indicates that stormwater discharges into sediment basins had turbidity values ranging
          from tens of NTUs to tens of thousands of NTUs (see DCN 44321). Therefore,
          turbidity and sediment are clearly present in construction site stormwater.

The third criterion requires that the system demonstrate good operation of the model treatment
technology. EPA had limited information about whether the treatment technology was well
operated. However, because this technology is relatively simple, EPA chose to  assume that most
effluent data from the model technology represented good performance. In evaluating this
criterion, EPA identified several areas that needed additional engineering investigation:

       •  Red.East and Red.West are two systems at the same residential  site. The two systems
          generally demonstrate larger turbidity values than the other systems. Because these
          basins were pretreatment basins before ATS filtration, the general goal at this site (as
          well as other ATS sites used in the calculation of the limitation) was to reduce the
          turbidity to a level that is acceptable for filtration, which is usually less than 500
          NTUs. Therefore, the dosage rate at this site was likely not optimized to reduce
          turbidity to low levels  (because that was not the goal of the pretreatment basin),  but
          the basin did remove significant turbidity. Because  the range of data values were
          generally within the range observed by the other systems in the data set (although
          more  often in the high end), EPA retained the data for both systems  as the basis for
          the limitation.

       •  NCR. 1: The discharge of 335 NTU on 7/25/06 was much greater than the other
          values observed at this site. In the engineering review, EPA contacted the author and
          EPA learned that a utility company had buried a line in the center of the treatment
          ditch  after moving the wattles out of the way. Although the wattles were returned to
          the ditch, they  were not stapled in which is essential to proper operation.
          Consequently,  the system performed poorly even during relatively little rain (9 mm).
          Because the system was not installed properly on that day, EPA excluded the value as
          a basis for the limitation.
                                          6-8

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
       •  NCR.2: The discharge of 533 NTU on 9/24/08 was much greater than the other
          values observed at this site. EPA contacted the author to obtain additional information
          on the data. Although the author was not able to provide any specific information on
          site activities that may have contributed to the much higher value on that day, EPA
          determined that because this data point was not consistent with other data from this
          site and from NCR.2 that this data point was likely not representative of normal
          operation. Therefore, this data point was excluded from calculation of the limitation.

The fourth criterion  ensures that the data represent normal operations. EPA's application of the
criterion excluded all turbidity measurements with zero values and/or associated with no flow.

 6.7.    SUMMARY OF LIMITATION DATA AND DATA CONVENTIONS

In developing the limitation, EPA focused its review on the performance and operating
conditions of sites that used systems that were consistent with EPA's model technology. Table
6-3 provides a summary of the reported turbidity measurements from the 25 systems with the
model technology (Section 6.4). EPA received more than 29,000 measurements of turbidity from
systems that met the requirements for EPA's model  technology.  These data are provided in
Listing 1 of Appendix F.  (DCN 42107 provides the data in an electronic spreadsheet file.) This
section describes EPA's review of the data, identifies data issues, and explains the rationale for
excluding certain data points from the limitations calculations.

EPA excluded the data for BHRBP (from Cascade EcoSolutions), because the data for BHRBP2
(from Clean Water) contained many of the same measurements taken at the same monitoring
point at the same time. Although there were a few differences, overall, measurements were the
same for sample dates summarized in both files. Because BHRBP2 provided data for a longer
period of time, EPA retained this data set and excluded the BHRBP from its calculations and
data listings.
           Table 6-3. Summary of reported turbidity measurements (NTU) in effluent
                 (individual measurements before daily average calculations)
System
BHRBP2
BWWTP
KC1.1
KC1.2
KC1.3
KC1.4
KC1.5
KC2.1
KC2.2
KC2.3
KC2.4
KC2.5
Number of
values
3,260
104
1,374
1,723
610
127
822
1,616
1,476
1,928
1,034
1,178
Average
(NTU)
73.964
113.579
59.256
61.272
56.131
42.649
42.677
90.528
91.243
70.314
74.662
71.082
Standard
deviation
92.456
57.366
36.703
40.033
20.084
11.79
18.16
49.272
47.118
41.326
48.28
45.01
Minimum
3.1
2.7
6
6.7
6.6
17.3
5.2
8
2.5
4.4
10.4
1.8
Median
47.2
135.25
51.3
50.7
51.9
45.3
39.6
81.6
82.35
63.1
69
62.45
Maximum
989.7
284
365.2
388.2
213.9
114.3
193
637.5
644.8
574
695.2
695.7
                                          6-9

-------
                                      Section 6: Limitations and Standards: Data Selection and Calculation
System
KC3.1
KC3.2
KC3.3
KC3.4
KCS.Pond
NC.Road*
NCR.1**
NCR.2**
NY
Red. East
Red. West
SEAAIR
STCLLR
OVERALL3"
Number of
values
594
620
721
622
110

105
9
7,089
3,467
762
366
196
> 29,91 3
Average
(NTU)
55.245
55.265
48.263
52.792
48.136

46.12
61.22
101.466
256.524
132.949
108.406
66.009

Standard
deviation
23.042
34.903
24.176
21.799
18.453



82.395
201.905
104.745
33.327
46.555

Minimum
12.1
8.7
10.1
0.4
10.4



1.6
1.6
1.4
24.51
4.7
0.4
Median
58.1
56.75
41.4
54.6
45.45



90.4
202.7
117.45
108.545
57.3

Maximum
320
486.6
195.9
224.6
130.9
339


1,000.9
1,000.9
614.1
209.22
293.9
1,000.9
a. The data for NC.Road included the mean, standard deviation and maximum value for each of 7 days. Because the number of
values used to calculate these statistics is not known, the only value that can be reported on this table for NC.Road is the maximum.
b. The data from NCR.1 and NCR.2 included the mean and the number of observations for each day. Samples collected at this site
were composites for the entire storm event. Therefore, the only values that can be reported on this table for NCR.1 and NCR.2 are
the number of values and the mean.
 6.8.     DATA AVERAGING PRIOR TO LIMITATION CALCULATIONS

The limitations for turbidity, as presented in today's notice, are provided as the maximum daily
discharge limitation. Definitions provided in 40 CFR 122.2 state that the maximum daily
discharge limitation is the highest allowable daily discharge. The definitions also state that
"[djaily discharge means the discharge of a pollutant measured during a calendar day or any
24-hour period that reasonably represents the calendar day for purposes of sampling."

In calculating the limitations, EPA analyzed the data from each treatment system separately,
even if the systems were located at the same site. (This is consistent with EPA's practice for
other industrial categories.) To be consistent with the daily discharge definition, EPA
arithmetically averaged all measurements recorded for each day from each treatment system
before calculating the limitations. EPA refers to this averaged value as the daily value.

Listing 2 of Appendix F identifies the 914 daily values obtained from arithmetically averaging
the values summarized in Table 6-3. Table 6-4 provides a summary of the daily values. From the
25 treatment systems, EPA observed a minimum daily value of 2.5 (NY) to a maximum of 672
NTU (Red.East).
                                            6-10

-------
                                     Section 6: Limitations and Standards: Data Selection and Calculation
                Table 6-4. Summary of daily values of turbidity (NTU) in effluent
System
BHRBP2
BWWTP
KC1.1
KC1.2
KC1.3
KC1.4
KC1.5
KC2.1
KC2.2
KC2.3
KC2.4
KC2.5
KC3.1
KC3.2
KC3.3
KC3.4
KCS.Pond
NC.Road
NCR.1
NCR.2
NY
Red. East
Red. West
SEAAIR
STCLLR
Total
Number of
daily values
116
8
28
32
10
3
16
33
30
40
23
19
13
13
15
15
7
7
12
3
220
169
56
9
17
914
Arithmetic
average
69.02
68.43
54.66
54.61
54.63
44.86
41.61
79.53
82.00
61.56
60.00
65.86
48.27
48.17
40.53
42.98
43.93
55.14
37.75
49.67
96.50
209.73
107.96
103.39
61.05
100.49
Standard
deviation
57.57
60.00
18.72
20.95
12.52
7.06
10.37
33.05
37.70
26.77
33.55
23.60
18.58
20.81
21.57
20.26
13.78
50.81
28.98
37.82
52.89
143.99
81.53
35.61
34.97
93.08
Minimum
12.00
12.70
21.85
13.90
41.76
38.63
32.13
23.32
21.12
21.65
23.52
34.08
18.30
18.02
14.78
15.58
25.13
11.00
9.00
15.00
2.50
5.35
11.92
34.84
11.22
2.50
Median
53.05
34.88
51.01
50.39
51.26
43.42
38.77
73.98
76.14
58.82
51.49
64.23
44.34
47.14
35.19
44.73
42.09
40.00
31.00
44.00
95.64
195.03
91.86
105.75
56.68
71.43
Maximum
527.75
151.30
103.39
117.17
85.66
52.53
75.88
154.77
192.14
130.94
152.17
117.74
79.02
79.96
79.01
73.54
63.99
167.00
109.00
90.00
549.39
672.65
341.21
155.92
161.76
672.65
 6.9.    LIMITATION CALCULATIONS

The limitations for turbidity, as presented in today's notice, are provided as the maximum daily
discharge limitation. This section describes the statistical percentile basis of the limitation
(Section 6.9.1), the concepts and calculations for the long-term average, the variability factor,
and the limitation (Sections 6.9.2, 6.9.3, and 6.9.4). Section 6.9.5 describes autocorrelation and
its effect on the value of the limitation.

   6.9.1.   STATISTICAL PERCENTILE BASIS FOR LIMITATIONS

The daily maximum limitation is an estimate  of the 99th percentile of the distribution of the daily
measurements. EPA calculates the daily maximum limitation on the basis of a percentile chosen
with the intention, on one hand,  to accommodate reasonably anticipated variability within the
control of the site and, on the other hand, to reflect a level of performance consistent with the
                                          6-11

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
Clean Water Act requirement that the BAT effluent limitation and NSPS be based on well-
operated and maintained facilities. The percentile for the daily maximum limitation is estimated
using the product of the long-term average and the variability factor derived from data that
represent the performance of the model technology under all conditions when properly operated
and controlled. For the rule, EPA estimated the long-term average and variability factor using a
statistical model based on the lognormal distribution as described in Appendix G.

   6.9.2.   LONG-TERM AVERAGE

In the first of two steps in estimating the different types of limitations, EPA determines an
average performance level (the long-term average) that systems representing well-designed and
operated model technologies (which reflect the appropriate level of control) are capable of
achieving. This long-term average is calculated from the data from the sites using the model
technology. The long-term average of 64.13 NTU is the median value of 25 long-term averages
collected from the 25 treatment systems. The long-term averages ranged from a minimum of 37
NTU (NCR.l) to a maximum of 251 NTU (Red.East).  The median is the midpoint of the 25
values, and is the 13th largest value, which is associated with KC2.4. As a consequence of using
the median, 12 of the system-specific averages are above the long-term average and 12 are
below, as shown in Table 6-5. EPA expects that all sites subject to the limitations will design and
operate their treatment systems to achieve the long-term average performance level on a
consistent basis because  sites with well-designed and operated model technologies have
demonstrated that this can be done.
          Table 6-5. System-specific long-term averages used in limitation calculations
System
BHRBP2
BWWTP
KC1.1
KC1.2
KC1.3
KC1.4
KC1.5
KC2.1
KC2.2
KC2.3
KC2.4
KC2.5
KC3.1
KC3.2
KC3.3
KC3.4
KCS.Pond
NC.Road
NCR.1
NCR.2
Number of
daily values
116
8
28
32
10
3
16
33
30
40
23
19
13
13
15
15
7
7
12
3
Long-term
average (NTU)
68.96
132.41
55.81
56.08
55.77
46.66
42.09
82.67
85.96
63.52
64.13
68.46
52.53
53.09
44.88
47.84
47.59
88.19
47.11
194.31
Rank
(smallest=1)
15
23
10
11
9
3
1
17
18
12
13 (median)
14
7
8
2
6
5
19
4
24
                                          6-12

-------
                                     Section 6: Limitations and Standards: Data Selection and Calculation
System
NY
Red. East
Red. West
SEAAIR
STCLLR
Number of
daily values
220
169
56
9
17
Median LTA
Long-term
average (NTU)
105.20
251.48
122.70
117.79
70.40
Rank
(smallest=1)
20
25
22
21
16
64.13
   6.9.3.   VARIABILITY FACTOR

EPA acknowledges that variability around the long-term average results from normal operations.
This variability means that occasionally sites can discharge at a level that is greater than the
long-term average. This variability also means that sites can occasionally discharge at a level that
is considerably lower than the long-term average. Consequently, in the second step of developing
a limitation, EPA determines an allowance for the variation in pollutant concentrations when
processed through well-designed and operated treatment systems.  This allowance for variance
incorporates all components of variability including process and wastewater generation, sample
collection, shipping,  storage, and analytical variability. This allowance is incorporated into the
limitations through the use of the variability factors, which are calculated from the data from the
sites using the model technology. The variability factor of 4.322 is the arithmetic average (or
mean) of 25 variability factors collected from the 25 systems also  used as the basis of the long-
term average. Table 6-6 provides the 25 system-specific variability factors. The variability
factors ranged from a minimum of 1.775 (KC1.5) to a maximum of 10.203 (BWWTP), and were
calculated as shown in Appendix G.
           Table 6-6. System-specific variability factors used in limitation calculations
System
BHRBP2
BWWTP
KC1.1
KC1.2
KC1.3
KC1.4
KC1.5
KC2.1
KC2.2
KC2.3
KC2.4
KC2.5
KC3.1
KC3.2
KC3.3
KC3.4
Number of daily
values
116
8
28
32
10
3
16
33
30
40
23
19
13
13
15
15
Daily variability factor
(VF1)
3.642
10.203
2.283
2.508
1.867
1.953
1.775
2.890
3.104
2.845
3.646
2.576
3.219
3.503
3.878
3.819
                                           6-13

-------
                                        Section 6: Limitations and Standards: Data Selection and Calculation
System
KCS.Pond
NC.Road
NCR.1
NCR.2
NY
Red. East
Red. West
SEAAIR
STCLLR
Number of daily
values
7
7
12
3
220
169
56
9
17
Mean VF1
Daily variability factor
(VF1)
2.809
8.135
5.859
12.968
4.140
6.391
5.934
3.586
4.513
4.322
In its evaluation of the daily variability factor, because it has not regulated turbidity for other
industries, EPA compared the daily variability factor developed from the C&D data with
variability factors developed for TSS effluent limitations guidelines and standards promulgated
during the past 12 years for various industrial categories. Because turbidity and TSS are treated
similarly by treatment systems, EPA would expect TSS levels and turbidity to exhibit similar
daily variability factors. While turbidity represents the appropriate parameter to regulate for
pollutant control in C&D discharges rather than TSS,1 EPA looked at TSS variability factors as a
check on its calculated turbidity daily variability factor. As shown in Table 6-7, the values for the
variability factors are relatively close in value, ranging from 2.9 to 5.4, with an arithmetic
average of 4.1 which is close to the variability factor of 4322 calculated for the C&D turbidity
limitation. EPA concluded that the value of 4322 ensures a level of control that EPA considers
achievable for discharges from C&D sites.
                      Table 6-7. TSS variability factors in recent regulations
Category
Centralized Waste Treatment
(USEPA2000)
Waste Combustors (USEPA
1999b)
Iron and Steel (USEPA 2002)
Subcategory
Organics
Oils
Metals
Commercial Hazardous Waste
Combustors
Coke By-Products
Other
Option
4
9
3
4

BAT1
DRI_BPT
FORGING
Value
4.8
2.9
3.2
3.6
4.2
4.6
3.5
4.4
1 As previously explained, turbidity is the BAT regulated pollutant parameter EPA selected as a pollutant itself and
as an indicator of toxic and non conventional pollutants discharged from C&D sites. While the discharge of
pollutants may be identified by a number of measures including measurement of, among others, turbidity,  suspended
solids and total suspended solids, EPA has selected turbidity as the better measure for sediment discharge  rather than
TSS for several reasons. These include the fact that discharges from sites with appreciable clay soils may exhibit
low TSS concentration but still have high turbidity level indicative of higher pollutant content.
                                              6-14

-------
                                     Section 6: Limitations and Standards: Data Selection and Calculation
Category
Landfills (USEPA1999a)
Pulp, Paper, and
Paperboard, Cluster Rule
(USEPA1997)
Transportation Equipment
Cleaning (Science
Applications International
Corporation 2000)
Subcategory
1) Hazardous and
2) Non-Hazardous*
bleached papergrade kraft and soda
Barge/Chemical & Petroleum
Food Direct
Option


1
2
Value
4.4
3.11
4.7
5.4
1 The variability factors for both subcategories were based on the same data.
    6.9.4.   CALCULATION OF THE LIMITATION

Using its standard approach for effluent guidelines, EPA calculated the value of the daily
maximum limitation (280 NTU) using the product of the long-term average (64.13 NTU) and
daily variability factor (4.322):
       Daily Maximum Limitation   =     Long-Term Average x Variability Factor
                                         (64.13 MIT) x (4.322)
                                         277.17 NTU
EPA rounded the value of the limitation to two significant digits (i.e., 280 NTU).

As a consequence of using the long-term average and variability factor as the basis of the
limitation, sites that are designed and operated to achieve long-term average levels should be
capable of compliance with the limitations, which incorporate variability, at all times.

    6.9.5.   LIMITATION INCLUDES AUTOCORRELATION ADJUSTMENT

The limitation calculations include an adjustment for possible bias due to statistical
autocorrelation. When data are said to be positively autocorrelated, it means that measurements
taken at specific time intervals (such as 1 day  or 2 days apart) are related. For example, positive
autocorrelation would be present in the data if the effluent concentration was relatively high one
day and was likely to remain at similar high values the next and possibly succeeding days.
Because the values tend to be similar from day to day, the variance estimate may be dampened
even within a relatively large time period. By  accounting for autocorrelation, the adjusted
variance then better reflects the underlying variability that would be present if the data were
collected over an even longer period. To evaluate autocorrelation, generally, the statistical
analysis requires at least a 50-day period with measurements from every day during the period.2
C&D discharge data generally do not have measurements for every day. Instead, they are
generally associated with days with precipitation, and no discharge (or zero) for the other days.
After determining that the data were not suitable for the statistical analysis, EPA then considered,
from an engineering aspect, whether it was likely that treatment from one day to the next would
2 Box and Jenkins (1976), a classic textbook on time series analyses, states, "It is normally supposed that successive
values of the time series under consideration ... are available for analysis. If possible, at least 50 and preferably 100
observations should be used." (page 18)
                                           6-15

-------
                                     Section 6: Limitations and Standards: Data Selection and Calculation
be similar. On one hand, conditions can vary considerably from day to day which would lead
EPA to conclude that treatment would be relatively unaffected by the previous day's treatment.
On the other hand, if stormwater is detained in a pond for a sufficient amount of time, the settling
process might indicate that discharges one day apart might be similar.  Because EPA's
engineering review was inconclusive, EPA determined that, as a conservative measure to ensure
that the limitation was achievable, it was appropriate to incorporate a statistical adjustment for
autocorrelation.

As explained in Section 6.9.3, turbidity has not been regulated in other industries, and thus, EPA
again investigated whether it could transfer autocorrelation adjustments developed for TSS
limitations in other industries. Of the choices listed in Table 6-7  and for which the statistical
documentation is readily available, EPA only adjusted for autocorrelation in the TSS NSPS for
the pulp, paper, and paperboard industry. EPA determined that it was appropriate to transfer the
autocorrelation adjustment because: 1) EPA expects turbidity and TSS to be treated similarly by
the model technology; and 2) the model technologies for the two industries similarly detain the
wastestreams prior to discharge. EPA also notes that the adjustment for the 1998 regulation was
relatively large, and thus, it is unlikely that the C&D discharges would require an even larger
adjustment. Appendix G describes the application of the  1998 adjustment to the variance of the
C&D data. The consequence of applying this adjustment to the C&D discharges was an increase
in the value of the limitation from 189 to 280 NTU. Table 6-8 provides the values of the system-
specific long-term averages and variability factors with and without the autocorrelation adjustment.
                 Table 6-8. Effect of autocorrelation adjustments on limitation
System
BHRBP2
BWWTP
KC1.1
KC1.2
KC1.3
KC1.4
KC1.5
KC2.1
KC2.2
KC2.3
KC2.4
KC2.5
KC3.1
KC3.2
KC3.3
KC3.4
KCS.Pond
NC.Road
NCR.1
Number of
daily values
116.00
8.00
28.00
32.00
10.00
3.00
16.00
33.00
30.00
40.00
23.00
19.00
13.00
13.00
15.00
15.00
7.00
7.00
12.00
WITHOUT autocorrelation
adjustment
Long-term
average (NTU)
68.01
74.32
54.80
54.93
54.66
45.04
41.56
80.40
83.00
62.09
60.43
66.22
48.99
48.91
41.00
43.82
44.30
57.15
38.43
Daily variability
factor (VF1)
3.510
6.007
2.067
2.277
1.573
1.417
1.585
2.599
2.746
2.606
3.068
2.204
2.502
2.682
3.005
2.966
2.008
4.609
4.071
WITH autocorrelation adjustment
(used as basis of limitation)
Long-term
average (NTU)
68.96
132.41
55.81
56.08
55.77
46.66
42.09
82.67
85.96
63.52
64.13
68.46
52.53
53.09
44.88
47.84
47.59
88.19
47.11
Daily variability
factor (VF1)
3.642
10.203
2.283
2.508
1.867
1.953
1.775
2.890
3.104
2.845
3.646
2.576
3.219
3.503
3.878
3.819
2.809
8.135
5.859
                                           6-16

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
System
NCR.2
NY
Red. East
Red. West
SEAAIR
STCLLR
Number of
daily values
3.00
220.00
169.00
56.00
9.00
17.00
Median LTA
Mean VF1
99th Percentile
WITHOUT autocorrelation
adjustment
Long-term
average (NTU)
58.60
104.22
245.48
115.34
106.05
63.20
Daily variability
factor (VF1)
5.426
4.054
6.194
5.411
2.531
3.515
58.60
3.225
189.00
WITH autocorrelation adjustment
(used as basis of limitation)
Long-term
average (NTU)
194.31
105.20
251.48
122.70
117.79
70.40
Daily variability
factor (VF1)
12.968
4.140
6.391
5.934
3.586
4.513
64.13
4.322
277.17
 6.10.   STATISTICAL AND ENGINEERING REVIEW OF LIMITATION

In conjunction with the statistical methods, EPA performs an engineering review to verify that
the limitations are reasonable based on the design and expected operation of the model
technologies and the site conditions. Data from some sites demonstrate the best available
technology. Data from other sites could demonstrate the same technology but not the best
demonstrated design and operating conditions for that technology. EPA recognizes that, as a
result of the limitation, some dischargers might need to improve treatment systems, erosion and
sediment controls, and/or treatment system operations to consistently meet the effluent
limitation. EPA determined that this consequence is consistent with the Clean Water Act
statutory framework, which requires that discharge limitations reflect the best available
technology (BAT) or best available demonstrated technology (BADT).

The following sections describe several aspects of the engineering review. Section 6.10.1
compares the value of the limitation to the performance data used as the basis of the limitation.
Section 6.10.2 compares the performance data for other technologies to the value of the
limitation. Section 6.10.3 evaluates the performance data for the C&D model technology
collected during the rulemaking for gold placer mining discharges to confirm that the data are
consistent with the limitation. Section 6.10.4 compares the limitation to benchmarks established
by several states. Section 6.10.5 considers other factors and performance.

   6.10.1.  PERFORMANCE DATA FOR MODEL TECHNOLOGY COMPARED TO
            LIMITATION

To evaluate the  value of the limitation, EPA compared the value of the limitation to the daily
values used to calculate the limitation. Because of the statistical models used to derive the
limitation from their data, EPA would expect about one percent of the values to be greater than
the limitation. From an engineering perspective, in most instances where the daily values were
greater than the  turbidity limitation, the system was not optimized appropriately and/or there was
some other factor responsible for the higher values. Table 6-9 summarizes the results of this joint
statistical and engineering review of the data and the performance at each system.
                                         6-17

-------
                              Section 6: Limitations and Standards: Data Selection and Calculation
•  21 of the 25 systems had all the daily values less than the limitation. Because the
   database contained fewer than 100 daily values for each system, EPA would expect
   only one, if any, daily values at each system would be greater than the limitation. For
   these 19 systems, EPA concluded that the finding that none of the values were greater
   than the limitation is consistent with what is expected from the 99th percentile basis of
   the statistical methodology. The finding was also consistent from an engineering
   perspective because properly operated and controlled systems are expected to operate
   below the level of the limitation.
•  BHRBP2 had 1 of 116 (or one percent)  daily values greater than the limitation, which
   is consistent with what is expected from the 99th percentile basis of the statistical
   methodology. The engineering investigation revealed that the daily value of 528 NTU
   was observed on December 3, 2007 during an extreme storm event in Seattle, WA
   (http://www.climate.washington.edu/events/dec2007floods/). The Office of the
   Washington State Climatologist estimated that 6-hour and 24-hour precipitation
   amounts were near 100-year rain frequency levels. This event does not represent
   normal  conditions and the rule exempts such events from complying with the
   limitation. EPA notes that data from 13  other systems in the Seattle area is also
   available during this storm event and the average turbidities for the day were  all
   below the value of the limitation. Thus,  even during the extreme storm event, it was
   possible to achieve turbidity control.

•  New York had two of 220 (one percent) daily values greater than the limitation,
   which also is consistent with what is statistically expected from the 99th percentile
   basis of the statistical  methodology. The two values, 550 and 284 NTU, were
   observed on two consecutive days, October 2 and 3 in 2006, which were the first two
   days reported by the vendor. The engineering review concluded that they were likely
   the result of insufficient flocculant dosing during start-up operations. EPA has
   determined that they do not represent normal operations because optimization, even
   during startup operations, is relatively simple and easy to achieve as demonstrated by
   the other systems in the limitation data set. In addition, as described earlier, this pond
   was pretreatment to an ATS system, and therefore the targeted turbidity in the
   effluent was likely 500 NTUs or less.

•  Red.East had 50 of its 169 values greater than the limitation with a maximum value of
   673 NTU. Based upon its engineering review of the data, EPA concluded that it was
   likely not optimizing its system because it only needed to meet a target level of 500
   NTU (although several values were above this level). Proper dosing is necessary for
   adequate turbidity removal, and it is likely that the dosage rate of flocculant was not
   optimized for obtaining lower turbidity  because the other systems evaluated and
   described here produced consistently lower turbidity values.

•  Red.West had 2 of its 56 values greater  than the limitation which is slightly more than
   what is  statistically expected. These values were observed on November 6 (334 NTU)
   and December 14, 2006 (341 NTU). The engineering review concluded that this
   system was likely not optimized, because it was targeting the model technology to a
   relatively high level of 500 NTU.
                                   6-18

-------
                                  Section 6: Limitations and Standards: Data Selection and Calculation
                Table 6-9. Daily values greater than daily maximum limitation
System
BHRBP2
BWWTP
KC1.1
KC1.2
KC1.3
KC1.4
KC1.5
KC2.1
KC2.2
KC2.3
KC2.4
KC2.5
KC3.1
KC3.2
KC3.3
KC3.4
KCS.Pond
NC.Road
NCR.1
NCR.2
NY
Red. East
Red. West
SEAAIR
STCLLR
Number of
Daily Values
116
8
28
32
10
3
16
33
30
40
23
19
13
13
15
15
7
7
12
3
220
169
56
9
17
Daily values greater than daily
maximum limitation of 280 NTU
Number of
values
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
50
2
0
0
Percent of total
number
0.9%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.9%
29.6%
3.6%
0.0%
0.0%
   6.10.2. PERFORMANCE OF OTHER TREATMENT SYSTEMS RELATIVE TO
           MODEL TECHNOLOGY

EPA compared the limitation to data from other treatment systems to determine if the
performance data behaved as expected. For systems, such as ATS, that are more complex than
the model technology, EPA expects the performance data to have lower levels of turbidity than
the limitation data. For less sophisticated systems, such as best management practices, EPA
expects the performance data to have higher levels of turbidity. As described below, these
comparisons to the limitation generally confirmed EPA's expectations.

EPA considered treatment data from systems that were expected to perform better than the model
technology. Of the approximately 24,000 effluent measurements from 38 systems using ATS, the
largest measured value was 71.6 NTU. As EPA expected, in all cases, the advanced systems
demonstrated lower levels of turbidity than  required by the limitation. In addition, EPA
                                        6-19

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
considered the influent data for the system for the WSDOT SR-522 road improvement project at
Elliot Road (ELLRD). Although EPA did not use this data in calculating the limitation, its
average value was 42.10 NTU, which was lower than the 64 NTU long-term average basis of the
limitation. In addition, the maximum value obtained for ELLRD was 182 NTU which is well
below the limitation of 280 NTU. Thus, the ELLRD system performance is better than EPA's
model technology.

EPA also considered data from other studies that did not use the model technology (polymer-
aided settling) but that used other conventional BMPs. See Chapter 5 and DCN 44321 for a
summary of studies evaluating the performance of sediment basins. Two key studies of
conventional BMPs are Warner and Collins-Camargo (2001) and Horner, Guedry and Kortenhof
(1990). EPA also evaluated other studies that evaluated passive treatment, but were not used as a
basis for calculating the limitation. See DCN 43114 for a summary of various other passive
treatment approaches. Two key studies evaluating polymer-aided settling were prepared for the
Auckland Regional Council (2004 and 2008). These studies evaluated TSS, not turbidity, and
hence were not used for calculation of the limitation, but provide useful information on polymer-
aided settling in sediment basins.

EPA also considered influent data that it had collected for the proposed rule to determine if the
levels appeared to be greater than the turbidity limitation. In this comparison, EPA was verifying
that it was establishing a limitation that would require treatment. The 1008 influent
measurements that reflect either uncontrolled wastewater and/or runoff managed by upslope
BMPs (i.e., relatively minor treatment) prior to the pond. As shown in Table 6-10, the average
turbidity levels ranged from 105 to 2581 NTU, and thus, all are greater than the long-term
average basis (64 NTU) for the limitation. In addition, the maximum values generally are larger
than any value observed from effluent from the model technology. Thus, EPA concluded that
treatment is likely to be necessary for sites to comply with the limitation.

        Table 6-10. Daily value summary from systems with less than model technology
               mi	•_____*     n _-j_i	_j_-_      f**_	_•	i        •••__•	       R/l^wirvti i
Site
11
2
3
4
6
8
BZR08
HEO
LSIDE
SC05
SC08
WLCPO
Number of
daily values
32
23
16
18
8
14
9
4
76
7
14
10
Arithmetic
average (NTU)
148.63
2,581.29
105.39
398.86
552.59
610.56
323.63
604.00
242.45
255.50
848.08
194.50
Standard
deviation
187.25
1,255.08
96.40
283.98
201.28
329.19
83.90
99.37
120.62
47.34
143.65
123.68
Minimum
(NTU)
1.08
853.00
27.40
10.20
209.00
204.00
149.00
466.00
52.25
210.00
625.73
78.10
Maximum
(NTU)
1,020.00
4,816.00
380.20
985.00
951.00
1,000.00
420.00
696.00
662.89
331.00
1,080.00
472.00
                                          6-20

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
    6.10.3.  PERFORMANCE OF MODEL TECHNOLOGY FOR GOLD PLACER
            MINING WASTES

As an additional step in the engineering review, EPA reviewed its records for its 1988
rulemaking for the gold placer mining subcategory of the ore mining and dressing point source
category. The placer mining regulation established a limitation for settleable solids based on
simple settling, which is not equivalent to the model technology for C&D. However, during
development of the placer mining regulation EPA conducted treatability studies at placer mining
sites to evaluate the  performance of simple settling and chemically aided settling in reducing
settleable solids, TSS and turbidity. Although the mining wastes evaluated had solids content
generally higher than expected for C&D, and that these were jar tests as opposed to basin
influent/effluent samples, EPA found comparable performance in the treatability tests. The 1986
Alaskan Placer Mining Study Field Testing Program Report (USEP A 1987) performed
chemically assisted tests that determined the effect of poly electrolytes and polyethylene oxide
(PEO) on turbidity levels. Results, presented in Table 6-11, indicate the PEO, with and without
polyelectrolyte, can  reduce turbidity to low levels. Moreover, although the initial turbidity levels
are generally substantially greater than the levels in C&D wastewater, the model technology was
able to drop below 280 NTU within one hour. The only exception (mine 4998) dropped to 220
NTU during the second hour. After 6 hours (EPA expects longer settling periods for C&D
wastewater in ponds, which are generally designed to provide 24 to 72 hours of detention time),
EPA would expect to see about half of the values to be greater than (and half less than) the long-
term average basis of the limitation. The result, with only four values greater than the long-term
average, is consistent, and perhaps even better, performance than EPA expects from the model
technology. Thus, EPA concludes that the gold placer mining study demonstrates that EPA's
model technology can achieve the low levels required by the limitation, even in the presence of
extremely high levels of turbidity in the influent.
                Table 6-11. Turbidity during 1986 Alaskan placer mining study
Chemically aided settling (PEO tests)
turbidity (NTU)
Mine No.
4922
4998
4999
5000
5001
5002
5003
5004
Test No.
21
4
8
12
15
17
24
29
Average: All Mines (PEO)
Initial
39,500
39,500
11,000
6,500
2,450
22,500
1,235
23,250
18,242
30 min
35,000
350
188
45
54
90
144
134
4,501
1 hr
31
275
148
44
41
83
95
93
101
2hr
21
220
144
42
40
82
77
83
89
3hr
20
210
140

38
82
68
71
90
4hr
12
175
136

35
80
60
69
81
5hr
11
140
128


80
53
59
79
6hr
10
140
122


75
51
52
75
                                          6-21

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
Chemically aided settling (polyelectrolyte)
turbid ity(NTU)
Mine No.
4922

4998

4999

5000

5001
5003

5004

Test No.
19
20
2
3
6
7
10
11
14
23
25
27
28
Average: All Mines
(Polyelectrolyte)
Initial
40,000
42,500
18,600
8,500
16,250
22,000
1,680
2,050
2,000
9,000
1 1 ,320
23,150
23,500
16,965
30 min
24,250
21,250
58
570
134
84
37
40
32
14
41
41
45
3,584
1 hr
52
63
37
54
97
70
34
30
29
13
36
20
23
43
2hr
26
50
30
54
90
67
23
25
27
12
37
10
18
36
3hr

36
28

90

22

26
12


11
32
4hr

33
28

89

18

25
11


9
30
5hr

27
28

89

15


11


6
29
6hr

27
27

78

15


10


5
27
   6.10.4.  LIMITATION IS CONSISTENT WITH STATE ACTION LEVELS

EPA also compared its long-term average basis of the limitation with action levels established by
four states. EPA is not aware of any statistical studies that were the basis of the action levels in
these four states; however, EPA assumes that there was a technology basis or BPJ basis that
established that the levels were reasonable. While an action level is not an enforceable limit, it
does indicate that the state determined that discharges greater than the action level may require
additional optimization or treatment technology. To comply with the limitation, EPA
recommends that operators target the system performance to the long-term average basis of the
limitation and control discharges above that level. Because the state action levels and the long-
term average both indicate a potential need for operator action, EPA compared the state action
levels to 64 NTU, the long-term average basis of the limitation, EPA found that its long-term
average basis is consistent with or more stringent than the state action levels:  50 NTU in
Vermont (DCN 42108); 120 NTU in Oregon (DCN 42109) and Washington (DCN 42110); and
250 NTU in California (DCN 42104). In addition, California has established an enforceable
NTU limit of 500 in its current general permit for certain sites. The data from Washington
demonstrate that the action levels have generally been effective in controlling the discharges
with 75 percent of the daily values less than the action level of 120 NTU.

 6.11.   MONITORING CONSIDERATIONS

Effluent guidelines act as a primary mechanism to control the discharge of pollutants to waters of
the United States. The C&D regulations will be applied to C&D sites through incorporation in
individual National Pollutant Discharge Elimination System (NPDES) permits or a general
permit issued by EPA or authorized states or tribes under section 402 of the Clean Water Act. In
                                          6-22

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
complying with the final rule, the number of measurements required each day would be
determined by the permit authority. While the actual monitoring requirements will be determined
by the permitting authority, in developing the limitation, the Agency has assumed that sites will
report one value for every day that the discharge occurs to be consistent with permit definitions
provided in 40 CFR 122.2 that define the maximum daily discharge limitation as the "highest
allowable daily discharge."

EPA recognizes that some  sites regularly monitor multiple times throughout the day to ensure
that the treatment system is operating properly. In addition, because turbidity can be measured
real-time, it is possible to use an automated turbidity meter in conjunction with a data logger to
obtain data during the entire period of discharge. While EPA agrees that such monitoring is
appropriate, EPA would, however, discourage the practice of allowing the number of monitoring
samples to vary arbitrarily  merely to allow a site to achieve a desired average concentration, i.e.,
a value below the limitation that day. EPA expects that enforcement authorities would prefer, or
even require,  monitoring samples at some regular, predetermined frequency. As explained
below, if a site has difficulty complying with the limitation on an ongoing basis, the site should
improve its equipment, operations, and/or maintenance.

 6.12.   COMPLIANCE

EPA promulgates limitations that sites are capable of complying with at all times by properly
operating and maintaining their processes and treatment technologies. However, the issue of
exceedances or excursions (values that exceed the limitations) is often raised. Comments often
suggest that EPA include a provision that a facility is in compliance with permit limitations if its
discharge does not exceed the specified limitations, with the exception that the discharge may
exceed the monthly average limitations 1 month out of 20 and the daily average limitations 1 day
out of 100. This issue was, in fact, raised in other rules, including EPA's final Organic
Chemicals, Plastics, and Synthetic Fibers (OCPSF) rulemaking. EPA's general approach in that
case for developing limitations based on percentiles was the same as this rule and was upheld in
Chemical Manufacturers Association v. U.S. Environmental Protection Agency, 870 F.2d 177,
230 (5th Cir.  1989). The Court determined the following:

       EPA reasonably concluded that the data points exceeding the 99th and 95th
       percentiles represent either quality-control problems or upsets because there can
       be no  other explanation for these isolated and extremely high discharges.  If these
       data points result from quality-control problems, the exceedances they represent
       are within the control of the plant. If, however, the data points represent
       exceedances beyond the control of the industry, the upset defense is available.

Additionally, this issue was raised in EPA's Phase I rule for the pulp and paper industry. In that
rulemaking, EPA used the  same general approach for developing limitations based  on percentiles
that it had used for the OCPSF rulemaking and for the proposed CAAP rule. This approach for
the monthly average limitation was upheld in National Wildlife Federation et al. v.
Environmental Protection Agency, 286 F.3d 554, 573 (D.C. Cir. 2002). The Court determined
that:
                                          6-23

-------
                                     Section 6: Limitations and Standards: Data Selection and Calculation
       EPA's approach to developing monthly limitations was reasonable. It established
       limitations based on percentiles achieved by facilities using well-operated and
       controlled processes and treatment systems. It is therefore reasonable for EPA to
       conclude that measurements above the limitations are due to either upset
       conditions or deficiencies in process and treatment system maintenance and
       operation. EPA has included an affirmative defense that is available to mills that
       exceed limitations due to an unforeseen event. EPA reasonably concluded that
       other exceedances would be the result of design or operational deficiencies. EPA
       rejected Industry Petitioners' claim that facilities are expected to operate
       processes and treatment systems so as to violate the limitations at some pre-set
       rate. EPA explained that the statistical methodology was used as a framework to
       establish the limitations based on percentiles. These limitations were never
       intended to have the rigid probabilistic interpretation that Industry Petitioners
       have adopted. Therefore, we reject Industry Petitioners' challenge to the effluent
       limitations.

As that Court recognized, EPA's allowance for reasonably anticipated variability in its effluent
limitations, coupled with the availability of the upset defense, reasonably accommodates
acceptable excursions. Any further excursion allowances would go beyond the reasonable
accommodation of variability and would jeopardize the effective control of pollutant discharges
on a consistent basis and/or bog down administrative and enforcement proceedings in detailed
fact-finding exercises, contrary to congressional intent. See, for example, Rep. No. 92-414, 92d
Congress, 2d Sess. 64, reprinted in^4 Legislative History of the Water Pollution Control Act
Amendments of 1972 (at 1482); Legislative History of the Clean Water Act of 1977 (at 464-65).

More recently, for EPA's rule for the  iron and steel industry, EPA's selection of percentiles was
upheld in American Coke and Coal Chemicals Institute v. Environmental Protection Agency,
452 F.3d 930, 945 (D.C. Cir. 2006). The Court determined that

       The court will not second-guess EPA's expertise with regard to what the
       maximum effluent limits represent. See Nat'l Wildlife, 286  F.3d at 571-73. As
       EPA explains in the Final Development Document, the daily and monthly average
       effluent limitations are not promulgated with the expectation that a plant will
       operate with an eye toward barely achieving the limitations. Final Development
       Document at § 14.6.2. Should a plant do so, it could be expected to exceed these
       limits frequently because of the foreseeable variation in treatment effectiveness.
       Rather, the effluent limitations are promulgated with the expectation that plants
       will be operated with an eye towards achieving the equivalent of the LTA for the
       BAT-1 model technology. Id.  However, even operated with the goal of achieving
       the BAT-1 LTA, a plant's actual results will vary. EPA's maximum daily
       limitations are designed to be forgiving enough to cover the operations of a well-
       operated model facility 99% of the time, while its maximum monthly average
       limitations are designed to be forgiving enough to accommodate the operations of
       a well-operated model facility 95% of the time.  See id. EPA's choice of percentile
       distribution represented by its maximum effluent limitation  under the CWA
       represents an expert policy judgment that is not arbitrary or capricious.
                                          6-24

-------
                                    Section 6: Limitations and Standards: Data Selection and Calculation
EPA expects that sites will comply with promulgated limitations at all times. If the exceedance is
caused by an upset condition, the site would have an affirmative defense to an enforcement
action if the requirements of 40 CFR 122.41(n) are met. If the exceedance is caused by a design
or operational deficiency, EPA has determined that the site's performance does not represent the
appropriate level of control (best available technology for existing sources; best available
demonstrated control technology for new sources). For other promulgated limitations and
standards, EPA has determined that such exceedances can be controlled by diligent process and
wastewater treatment system operational practices such as frequent inspection and repair of
equipment, use of backup systems, and operator training and performance evaluations. For this
effluent guideline, EPA has provided reasonable relief from the effects of large storm events by
requiring that the limitation only apply to discharges on days with precipitation less than the
local 2-year, 24-hour storm event.

 6.13.   SUMMARY OF STEPS USED TO DERIVE THE LIMITATIONS

This section summarizes the steps used to derive the limitations for turbidity:

Step 1   EPA calculated daily averages from the individual measurements for each treatment
        system.

Step 2   EPA calculated the system-specific long-term averages and daily variability factors for
        each of the 25 systems  that had the model technology.

Step 3   EPA calculated the long-term average of 64.13 NTU as the median of the site long-
        term averages. (See Table 6-5.) EPA expects that all sites subject to the limitations will
        design and operate their treatment systems to achieve the long-term average
        performance level on a consistent basis.

Step 4   EPA calculated the variability factor of 4322 as the mean of the system-specific
        variability factors. (See Table 6-6.) If a site operates its treatment system to meet the
        relevant long-term average, EPA expects the site to be able to meet the limitations. The
        variability factor assures that normal fluctuations in a site's treatment are accounted for
        in the limitation. By accounting for these reasonable excursions above the long-term
        average, EPA's use of variability factors results in limitations that are generally well
        above the actual long-term averages.

Step 5   EPA calculated the daily maximum limitation of 280 NTU using the product of the
        long-term average (64.13 NTU) and the daily variability factor (4.322).

Step 6   EPA compared the daily maximum limitations to the site daily values used to develop
        the limitations. (See Table 6-9.) EPA usually performs this comparison to determine
        whether it used appropriate distributional assumptions for the data used to develop the
        limitations. This comparison considers whether the curves EPA used provide a
        reasonable^ to the actual effluent data or if there was an engineering or process reason
        for an unusual discharge. Although the fact that the Agency performs such an analysis
        before promulgating limitations might give the impression that EPA expects occasional
        exceedances of the limitations, this conclusion is incorrect. EPA promulgates
                                          6-25

-------
                                   Section 6: Limitations and Standards: Data Selection and Calculation
        limitations that facilities are capable of complying with at all times by properly
        operating and maintaining their treatment technologies. After performing an
        engineering evaluation of the larger values, EPA concluded that the limitation was
        reasonable.

 6.14.   REFERENCES

Auckland Regional Council. 2004. The Use of Flocculants and Coagulants to Aid the Settlement
   of Suspended Sediment in Earthworks: Trials, Methodology and Design. Technical
   Publication 227. Auckland Regional Council. (DCN 41112).

Auckland Regional Council. 2008. Performance of a Sediment Retention Pond Receiving
   Chemical Treatment. Auckland Regional Council. (DCN 42102).

Battelle. 2009a. Memorandum from Laura Aume to Maria Smith. C&D Analysis of ATS Data.
   September  14, 2009 (DCN 42112).

Battelle. 2009b. Memorandum from Laura Aume to Maria Smith. C&D: Final Limitations.
   November 20, 2009 (DCN 42128).

Box, G., and G. Jenkins. 1976. Time Series Analysis forecasting and control. Revised Edition.
   Holden-Day, Inc., CA.

Cascade EcoSolutions. 2008. Personal communications transmitting information and files on
   February 14, April 27, April 30, May 8, and May 13, 2008. Data sets included Engineering
   Report Data and ChitoVan Performance Review Data Set. (DCN 43002).

Clager, Tyrone. 2009. Personal communication transmitting files and other information. (DCN
   42137).

Clear Creek Systems,  Inc. 2008. Personal communications transmitting files and other
   information. Data  files included Data from Some Sites 2005-06 Season and Chitosan
   Treatment Results 2005.  (DCN 43003).

Clear Creek Systems,  Inc. 2009. Data files for King County, Morrisville, Redmond, Seattle and
   Springville sites. (DCN 43125).

Gannon, Joe. 2009. Personal communication transmitting information about ATS sites (DCN
   42130).

Holloway, Nathan. 2009. Personal communications transmitting files and other information.
   Data files include 8LST01 History.xls andSHUDOl FIELD DATA.xlsx. (DCNs 42133,
   42134, 42135 and 42136).

Horner, R., J. Guedry, and M. Kortenhof. 1990. Improving the Cost Effectiveness of Highway
   Construction Site Erosion and Pollution Control. Report No. WA-RD 200.1. University of
   Washington, Washington State Transportation Center, Seattle, WA. (DCN 01350).
                                         6-26

-------
                                   Section 6: Limitations and Standards: Data Selection and Calculation
lurries, D. No date. Flocculation of Construction Site Runoff in Oregon. Oregon Department of
   Environmental Quality. .
   Accessed October 24, 2008. (DCN 43010.3).

McLaughlin, R. No Date. Target Turbidity Limits for Passive Treatment Systems.

McLaughlin, R. 2009. Personal communications transmitting information about passive
   treatment sites (DCNs 42131 and 42132).

McLaughlin, R., S. King, and G. Jennings. 2009. Improving Construction Site Runoff Quality
   with Fiber Check Dams and Polyacrylamide. Journal of Soil and Water Conservation
   64(2): 144-154.

Minton, G. 2006. Technical Engineering Evaluation Report (TEER) for the Chitosan-Enhanced
   Sand Filtration Technology for Flow-Through Operations. Prepared for Natural Site
   Solutions, LLC, Redmond, WA. (DCN 43002 with supporting data file at DCN 43002:
   Engineering Report Data.xls.) by Minton, Resource Planning Associates, Seattle, WA.

Minton, G., A.  Benedict, C. 1999.  Polymer-Assisted Clarification ofStormwaterfrom
   Construction Sites: Experience in the City of Redmond, Washington. Prepared for the City of
   Redmond, Washington, by Resource Planning Associates, Seattle, WA, and HoweConsult.
   (DCN  41107).

Minton, G., and A. Benedict. 1999. Use of Polymers to Treat Construction Site Stormwater.
   Proceedings of the International Erosion Control Association (IECA), Conference 30, Pp.
   177-188. IECA, Steamboat Springs, CO. (DCN 41108).

Office of the Washington State Climatologist. 2007 Record Flooding.
   . Accessed November 22, 2009.

Science Applications International Corporation. 2000. Statistical Support Document for Effluent
   Limitations Guidelines and Standards for the Transportation Equipment Cleaning Category.
   (DCN  45053).

USEPA (U.S. Environmental Protection Agency). 1987. Final Draft 1986 Alaskan Placer
   Mining Study Field Testing Program Report. (DCN 42013).

USEPA (U.S. Environmental Protection Agency). 1997. Statistical Support Document for the
   Pulp and Paper Industry: SubpartB. (DCN 45050).

USEPA (U.S. Environmental Protection Agency). 1999a. Statistical Support Document for
   Effluent Limitations Guidelines and Standards for the Landfills Category. EPA-821-B-99-
   007. (DCN 45052).

USEPA (U.S. Environmental Protection Agency). 1999b. Statistical Support Document for Final
   Effluent Limitations Guidelines and Standards for Commercial Hazardous Waste
   Combustors. Volume I, Volume II. EPA 821-B-99-010. (DCN 45051).
                                         6-27

-------
                                   Section 6: Limitations and Standards: Data Selection and Calculation
USEPA (U.S. Environmental Protection Agency). 2000. Development Document for Effluent
   Limitations Guidelines and Standards for the Centralized Waste Treatment Point Source
   Category. Volume I, Volume II. EPA 440/1-87/009.
   .

USEPA (U.S. Environmental Protection Agency). 2002. Development Document for Effluent
   Limitations Guidelines and Standards for the Iron and Steel Manufacturing Point Source
   Category. EPA-821 -R-02-004. .

USEPA (U.S. Environmental Protection Agency). 2009a. Passive Data Summary Spreadsheet.
   November 23, 2009 (DCN 42125).

USEPA (U.S. Environmental Protection Agency). 2009b. Passive Zero Excluded Data
   Spreadsheet. November 23, 2009 (DCN 42126).

USEPA (U.S. Environmental Protection Agency). 2009c. ATS Data Spreadsheet. November 23,
   2009 (DCN 42127).

USEPA (U.S. Environmental Protection Agency). 2009d. Email from Jesse Pritts to Maria
   Smith: ATS Data Transmittal. August 12, 2009 (DCN 42129).

Warner, R. and Collins-Camargo, F. Erosion Prevention and Sediment Control Computer
   Modeling Project. Surface Mining Institute, Lexington, KY. 2001. (DCN 430741).
                                        6-28

-------
                                                            Section 7: Technology Assessment
7.   TECHNOLOGY ASSESSMENT

This technology assessment is intended to determine the amount and quality of data available to
describe the performance of site runoff control practices, the ability of each practice to
effectively control the effects of runoff, and the design criteria or standards used to size each
practice to ensure effective control of runoff.

 7.1.     REVIEW OF HISTORICAL APPROACHES TO EROSION AND SEDIMENT
         CONTROL (ESC)

Most early sediment control was related to agriculture and was installed as a way to maintain our
natural resource base. On-site control was the primary emphasis, attempting to prevent erosion
rather than trap sediment. Strategies were developed to minimize exposure of bare soil to the
erosive power of rainfall and  stormwater, using aboveground cover management, residue
management, strip cropping, and terracing to limit the length of overland flow. Effects on
receiving streams and downstream areas  had not yet been identified as an issue. In the 1960s
concern began to be expressed about the  quantities of sediment in streams and reservoirs, and
sediment was first identified as a pollutant. Initially, the major focus of sediment control was on
the surface mining industry with the passage of the Clean Water Act and then the Surface
Mining, Reclamation, and Control Act (SMRCA) (PL 95-87). The first approach taken to
sediment control was a design standard, requiring a sediment detention basin  with a 24-hour
detention time; total suspended solids (TSS) standards of 35 milligrams per liter (mg/L) average
and 70 mg/L peak were also promulgated but were not typically  enforced. The U.S.
Environmental Protection Agency (EPA) later evaluated the TSS standard and moved to a
settleable solids standard of 0.5 mL/L because a modeling effort showed that it was  not possible
to trap fine sediments, but that a 0.5 mL/L settleable solids standard could be met with a
reasonably sized sediment basin (Ettinger and Lichty 1979).

In the late 1960s and early 1970s,  sediment in streams and waterways originating from urban
construction sites became an issue, which was then addressed in the Clean Water Act. EPA
developed a list of best management practices (BMPs) and standards for their construction
(USEPA 1971). In general, those standards were adopted from those of other agencies and were
not based on studies related to urban runoff.

In 1987 the Clean Water Act was amended to include stormwater discharges  from urban areas.
The Phase I National Pollutant Discharge Elimination System (NPDES) stormwater regulations
were published in  1990, requiring all municipalities with municipal separate storm sewer
systems serving populations greater than 100,000, construction sites 5  acres and larger, and
certain industrial sites to obtain a permit. The permit required the development of a stormwater
pollution prevention plan (SWPPP) that typically included a stormwater and sediment control
plan. In 1999 the Phase II NPDES stormwater regulations were published, extending permit
coverage to construction sites of 1 acre or larger and municipalities with populations greater than
50,000 (or populations greater than 10,000 where population density is more  than 1,000 people
per square mile). The regulations allow use of general permits in lieu of individual site or facility
permits. The degree of oversight of construction varies widely among the states.
                                          7-1

-------
                                                             Section 7: Technology Assessment
In the past two decades, increased concern at the local level has been focused on sediment
pollution of streams and waterways, particularly originating from construction, while less
concern has been focused on the effects of increased construction on stormwater and chemical
production. Much of the government concern originated from the Phase I and Phase IINPDES
stormwater regulations. A number of states and their local agencies have developed standards
and BMPs for sediment control, most of which do not have a scientific basis but were adopted
from other agencies. Some states, however, did conduct studies that gave their standards some
scientific basis. For example, Maryland evaluated its BMP standards in the 1980s by using
modeling techniques, and the state changed its sediment basin standards to account for the
effects of surface area on the trapping efficiency in sediment ponds. On the basis of typical soils
in the region and modeling studies, the state adopted a surface area to peak discharge ratio of
0.01 cubic feet per second (cfs) per acre as a criterion (Barfield and Clar 1985; McBurnie et al.
1990). Maryland was thus the first state to use a design criterion that was related to the overflow
rate. Other states also used some of Maryland's results (Smolen et al. 1988).

Recent efforts have moved closer to an effluent standard approach. South Carolina conducted a
detailed analysis and published regulations that required a trapping efficiency or settleable solids
standard (SCDHEC 1995). In addition, results from a detailed model were used to develop
simplified design aids (Hayes and Barfield 1995;  Holbrook et al. 1998). Some municipalities are
following suit to develop scientifically based standards of their own. For example, in 1998
Louisville, Kentucky (Hayes et al. 2001) developed standards and design aids for stormwater and
sediment control, following the example of South Carolina.

There are no examples in which an integrated approach to stormwater and sediment control has
been used on construction sites. The closest analog is the Cumulative Hydrologic Impact
Analysis (CHIA) required in surface mining by the SMRCA. SMRCA requires each applicant
for a surface mining permit to conduct a hydrologic impact analysis. Subsequently, the
regulatory authority is required to conduct a CHIA for the entire watershed. Note that although a
CHIA is required, it is seldom undertaken on a scale that is useful.

Many of the advances in sediment control have been based on the  capability to predict, a priori,
the ability of a given design to meet a standard. For example, when the settleable solids standard
was developed for surface mining, most regulatory authorities adopted it with the requirement
that permit applicants would demonstrate through the use of widely accepted computer models
that the proposed design would meet the settleable solids standard.

Most of the early work in  modeling sediment production stemmed from efforts in the 1950s to
develop a soil loss equation that would apply to the entire nation and allow evaluation of
alternative erosion  control practices. That led to the relationship known as the Universal Soil
Loss Equation (USLE) (Wischmeier and Smith 1965) and its subsequent derivative, the Revised
USLE (RUSLE) (Renard et al. 1994). Those efforts focus on erosion control; thus, the
relationships do not predict sediment yield. A flurry of efforts in the late 1970s and early 1980s
lead to the development of sediment yield relationships such as the Modified USLE (MUSLE)
(Williams, No Date), the CREAMS model (Knisel 1980), SEDCAD (Warner et al. 1999), and
SEDIMOT II (Wilson et al. 1982) and its derivatives. The MUSLE and CREAMS models did
not include methods to evaluate the impact of sediment trapping structures, but SEDIMOT II
contained relationships developed at the University of Kentucky to predict the effect of
                                          7-2

-------
                                                            Section 7: Technology Assessment
reservoirs (Ward et al. 1977; Wilson and Barfield 1984), check dams (Hirschi 1981), and
vegetative filter strips (Hayes et al. 1984). The MUSLE, SEDCAD, and SEDIMOT II models
were based on single storms, while the CREAMS model was based on continuous simulation
modeling. Details on those models are in Haan et al. (1994).

More recently, modeling has improved, resulting in several new relationships. The WEPP
watershed model is one example of a continuous simulation approach. It includes computational
procedures for a wide variety of sediment control structures (Lindley et al. 1998). Another
example of a single storm-based model is SEDIMOT III (Barfield et al.  1996), which modifies
the earlier SEDIMOT II model to include channel erosion routines and a wide variety of
sediment control techniques. A significant drawback in the SEDIMOT III and WEPP models is
that they do not have a good technique for predicting the impact of filter fence, which is the most
common technique used today for sediment control. The authors of SEDCAD have attempted to
provide algorithms to represent (silt) filter fence removals, although work remains before global
acceptance in the literature.

Concerns for changes in geomorphology resulting from flow alterations have resulted in several
modeling approaches. Early efforts were focused on what is known as the regime theory, in
which changes in channel property are linked, qualitatively, to changes in flow. Examples
include models of Lane (1955) and Schumm (1977). In addition, some statistically based models
were developed, but they are not universally applicable (Blench 1970; Simons and Albertson
1960).  More recently, models have been developed using physically based concepts to predict
changes in geomorphology as related to changes in flow. The models of Chang (1988) are good
examples. It is possible to predict, to a limited extent, the change in channel properties as
affected by changes in flow.

The impact of changes in flow and geomorphology on habitat is one major area in which
information is lacking. Although this deficiency can be addressed  qualitatively, it is not possible
to predict quantitatively how a given change in geomorphology will affect habitat. Additional
information is needed to develop a strategy on the basis  of the integrated assessment approach.

 7.2.    CONTROL TECHNIQUES

The following section presents a discussion of the commonly used ESC  practices.  Information on
applicability, design and installation criteria, maintenance considerations and effectiveness are
presented, when available. This section does not discuss proprietary and vendor-supplied BMPs,
many of which are variations of conventional BMPs such  as sediment barriers, filters, and
erosion control and prevention practices.

   7.2.1.  EROSION CONTROL AND PREVENTION

     7.2.1.1.  Planning, Staging, Scheduling

General Description
A construction sequence schedule is a specified work schedule that coordinates the timing of
land-disturbing activities and installing ESC measures. The goal of a construction  sequence
                                          7-3

-------
                                                               Section 7: Technology Assessment
schedule is to reduce on-site erosion and off-site sedimentation by performing land-disturbing
activities and installing ESC practices in accordance with a planned schedule (Smolen et al.
1988).

Construction site phasing involves disturbing only part of a site at a time to prevent erosion from
dormant parts (Claytor 1997). Grading activities and construction are completed and soils are
effectively stabilized on one part of the site before grading and construction begin at another
part. This differs from the more traditional practice of construction site sequencing, in which
construction occurs at only one part of the site at the time, but site grading and other site-
disturbing activities typically occur simultaneously, leaving portions of the disturbed site
vulnerable to erosion. Construction site phasing must be incorporated into the overall site plan
early on. Elements to consider when phasing construction activities include the following
(Claytor 1997):

       •  Managing runoff separately in each phase
       •  Determining whether water and sewer connections and extensions can be
          accommodated

       •  Determining the fate of already completed downhill phases

       •  Providing separate construction and residential accesses to prevent conflicts between
          residents living in completed stages of the site and construction equipment working
          on later stages (USEPA 2000)

Applicability
Construction sequencing can be used to plan earthwork and ESC activities at sites where land
disturbances might affect water quality in a receiving waterbody.

Design and Installation Criteria
Construction sequencing schedules should, at a minimum, include the following (NCDNR 1988;
MDE 1994):

       •  The ESC practices that are to be installed

       •  The principal development activities

       •  The measures that should be installed before other activities are started

       •  The compatibility with the general contract construction schedule

Table 7-1 summarizes other important scheduling considerations in addition to those listed
above.
                                            7-4

-------
                                                              Section 7: Technology Assessment
                Table 7-1. Scheduling considerations for construction activities
Construction activity
Construction survey stakeout
Preconstruction meeting with owner,
contractor, and regulatory agency
Construction access — entrance to
site, construction routes, areas
designated for equipment parking
Clearing and grading required for
installing controls
Sediment traps and barriers — basin
traps, silt fences, outlet protection
Runoff control — diversions,
perimeter dikes, water bars, outlet
protection
Runoff conveyance system —
stabilize streambanks, storm drains,
channels, inlet and outlet protection,
slope drains
Land clearing and grading — site
preparation (cutting, filling, and
grading; sediment traps; barriers;
diversions; drains; surface
roughening)
Surface stabilization — temporary
and permanent seeding, mulching,
sodding, riprap
Building construction — buildings,
utilities, paving
Landscaping and final stabilization —
adding topsoil, trees, and shrubs;
permanent seeding; mulching;
sodding; riprap
Schedule consideration
Before initiating any construction activity, a construction survey
stakeout should be conducted. The stakeout should identify the limits
of disturbance and location of control structures, especially perimeter
controls.
The meeting should take place before any construction activity begins
at the site. The survey stakeout is reviewed, especially the limits of
disturbance and location of controls.
This is the first land-disturbing activity. As soon as construction takes
place, any bare areas should be stabilized with gravel and temporary
vegetation.
In conjunction with the construction access, the clearing and grading
required for installing ESCs should take place.
After the construction site has been accessed, principal basins should
be installed, with the addition of more traps and barriers as needed
during grading.
Install key practices after installing principal sediment traps and before
land grading. Additional runoff control measures can be installed
during grading.
If necessary, stabilize streambanks as soon as possible, and install
the principal runoff conveyance system with runoff control measures.
The remainder of the systems can be installed after grading.
Implement major clearing and grading after installing principal
sediment and key runoff control measures, and install additional
control measures as grading continues. Clear borrow and disposal
areas as needed and mark trees and buffer areas for preservation.
Immediately apply temporary or permanent stabilizing measures to
any disturbed areas where work has been either completed or
delayed.
During construction, install any ESC measures that are needed.
This is the last construction phase. Stabilize all open areas, including
borrow and spoil areas, and remove and stabilize all temporary control
measures.
Effectiveness
Construction sequencing can be an effective tool for ESC because it ensures that management
practices are installed where necessary and when appropriate. A comparison of sediment loss
from a typical development and from a comparable phased project shows a 42 percent reduction
in sediment export in the phased project (Claytor 1997).
Limitations
Weather and other unpredictable variables can affect construction sequence schedules. The
proposed schedule and a protocol for making changes resulting from unforeseen problems should
be plainly stated in an applicable ESC plan.
                                           7-5

-------
                                                               Section 7: Technology Assessment
Maintenance
The construction sequence should be followed throughout the project, and the written ESC plan
should be modified before any changes in construction activities are executed. The plan can be
updated if a site inspection indicates the need for additional ESC as determined by contractors,
engineers, or developers.

Cost
Construction sequencing is a low-cost BMP because it requires a limited amount of a
contractor's time to provide  a written plan for coordinating construction activities and
management practices. Additional time might be needed to update the sequencing plan if the
current plan is not providing sufficient ESC.

Although little research has been done to assess the costs of phasing versus conventional
construction costs, it is known that it will be possible to implement successful phasing for a
larger project (Claytor 1997).

      7.2.1.2.   Vegetative Stabilization

Vegetation can be used during construction to stabilize and protect soil exposed to the erosive
forces of water, as well as post-construction to provide a filtration  mechanism for stormwater
pollutants. The following discussion refers to vegetative stabilization as a construction BMP that
stabilizes and protects soil from erosion.

General Description

Vegetative stabilization measures employ plant material to protect soil exposed to the erosive
forces of water and wind. Selected vegetation can reduce erosion by more than 90 percent
(Fifield 1999). Natural plant communities that are adapted to the site provide a self-maintaining
cover that is less expensive than structural alternatives. Plants provide erosion protection to
vulnerable surfaces by the following (Heyer, No Date):

       •  Protecting soil surface from the impact of raindrops
       •  Holding soil particles in place
       •  Maintaining the soil's capacity to absorb water
       •  Using living root systems to hold soil in place, increasing overall bank stability
       •  Directing flow velocity away from the streambank
       •  Acting as a buffer against abrasive transported materials
       •  Causing sediment deposition, which  reduces sediment load and reestablishes the
          streambank

The designer should be aware of and respond to local conditions that could influence the
development of vegetative stabilization measures. As with any planting design, climate,
maintenance practices, the availability of plant material (including native species), and many
other factors will influence such considerations as plant or seed mix selection, installation
methods, and project scheduling.
                                            7-6

-------
                                                               Section 7: Technology Assessment
Slope Stabilization. On slopes, the goal of vegetative stabilization is not only to reduce surface
erosion, but also to prevent slope failure. Vegetation should provide dense coverage to protect
soils from the direct effects of precipitation and help intercept runoff. A variety of plants should
be used to provide root systems that are distributed throughout all levels of the soil, increasing
slope shear strength and giving plants a greater ability to remove soil moisture. Uniform mats of
shallow rooting plants should be avoided because, while such plants might increase runoff
infiltration, they cannot remove soil moisture beyond the surface level, leaving slopes potentially
saturated and prone to slippage. Shallow, interlocking root systems could  also increase the size
of a soil slippage by holding together and pulling down a larger area of slope after a small
section has given way. Large trees that have become unstable can also pull down slopes and
should be removed. Using plants with low water requirements can reduce the potential for soil
saturation from irrigation.

Swale Stabilization. On swales, the goal of vegetative stabilization is to prevent erosion within
the swale, where runoff is concentrated and flows at higher velocities. If natural stream channels
are involved, vegetation with deep root systems should be preserved, or if absent, planted above
the channel to help maintain the channel banks. More information is provided in the subsequent
section dealing with grass-lined swales.

Surface Stabilization. On large, flat areas, the goal of vegetative stabilization is to reduce the
loss of surface soil from sheet erosion. Vegetation should provide complete coverage to reduce
the force of precipitation, which can shift soil particles to seal openings in the soil, reducing
infiltration and increasing runoff. Vegetation should also provide many stem penetrations to slow
runoff and increase infiltration. Deep rooting plants are less critical for erosion control in flat
areas than on slopes because soils are not subject to the same forces that can cause slippage on a
slope. However, trees and shrubs can increase infiltration, lessening the buildup of runoff, and
transpire large volumes of water, reducing soil saturation.

In areas susceptible to wind erosion, the goal of vegetative stabilization is to establish direct
protection of the soil. Vegetation should provide dense and continuous surface cover. Binding
the soil deeply is generally not a requirement. The ideal vegetation for this purpose is grass,
which forms a mat of protection. In areas where the vegetation is developed, the grass generally
has high maintenance requirements. In less developed, open areas, unmown grass, including
perennial native species, can be used to provide protection. Trees and shrubs also can provide
protection from the wind.

Shoreline Stabilization. In lakes and ponds, the goal of vegetative stabilization is to prevent
erosion of the shoreline. Wetland plants anchor the bottom of the lake or pond adjacent to the
shore and help dissipate the erosive energy of waves. An important consideration in planting
along shorelines is the need to establish favorable conditions for plant establishment and growth.
These include the proper grading of side slopes and the control of upland erosion to prevent the
buildup of silt and associated pollutants in the water. Designers should maintain awareness of
regulatory requirements that might influence vegetation projects in a wetland environment
(USAF 1998).

Vegetation used for shoreline stabilization work should be native material selected because of
strength, resiliency, vigor, and ability to withstand periodic inundation. Woody vegetation with
                                            7-7

-------
                                                               Section 7: Technology Assessment
short, dense, flexible tops and large root systems works well. Other important factors include
rapid initial growth, ability to reproduce, and resistance to disease and insects.

According to Heyer (No Date), most streambank stabilization plantings have used various
willows, including black willow (Salix nigra), sandbar willow (S. interior), meadow willow (S.
petiolaris\ heartleaf willow (S. rigida), and Ward willow (S. caroliniana). The size used
depends on the severity of the erosion and the type of bank to be stabilized.

Most tree revetment projects use either eastern red cedar (Juniperus virginiana) or hardwoods
such as northern pin oak (Quercus ellipsoidalis). Important suggestions include the following:

       •   Choose trees with many limbs and branches to trap as much sediment as possible.
       •   Select decay-resistant trees.
       •   Use recently cut trees—dead trees are more brittle and likely to break apart.
       •   The tree size-diameter of the tree crown should be about two-thirds of the height of
           the eroding bank.
       •   Cut off any trunk without limbs.
       •   Place the tree revetments overlapping, butt end pointing upstream.
       •   Begin and end revetments at  stable points along the bank.
       •   Choose an anchoring system according to the bank material to be stabilized and the
           weight of the object to be anchored.

Vegetative measures for streambank stabilization offer an alternative to structural measures and
are becoming well known as bioengineering techniques for streambanks. Using vegetative
material for streambank stabilization could be the first step in reestablishing the riparian forest,
which is essential for long-term stability of the streamside and floodplain areas. Each site must
be evaluated separately as to the feasibility of using natural material (Heyer, No Date).

Vegetative streambank stabilization, with the goal of protecting streambanks from the erosive
forces of flowing water, is generally applicable where bankfull flow velocity does not exceed 6
ft/sec and soils are erosion resistant (Smolen et al.  1988). Table 7-2 includes general guidelines
for maximum allowable velocities in streams to be protected by vegetation.

          Table 7-2. Conditions where vegetative streambank stabilization is acceptable
Frequency of bankfull flow
> 4 times/yr
1 to 4 times/yr
< 1 time/yr
Maximum allowable velocity for
highly erodible soil
4 ft/sec
5 ft/sec
6 ft/sec
Maximum allowable velocity for
erosion-resistant soil
5 ft/sec
6 ft/sec
6 ft/sec
Source: Smolen et al. 1988
                                            7-8

-------
                                                              Section 7: Technology Assessment
Temporary Vegetative Stabilization. Temporary vegetative cover such as rapidly growing
annuals and legumes can be used to establish a temporary vegetative cover. Such covers are
recommended for areas that (Fifield 1999)

       •  Will not be brought to final grade within 30 days or are likely to be redisturbed

       •  Require seeding of cut and fill slopes under construction

       •  Require stabilization of soil storage areas and stockpiles

       •  Require stabilization of temporary dikes, dams, and sediment containment systems

       •  Require development of cover or nursery crops to assist with establishing perennial
          grasses

Examples of temporary vegetation include wheat, oats, barley, millet, and sudan grass.
Temporary seeding might not be effective in arid or semi-arid regions where seasonal lack of
moisture prevents germination. It might be necessary to use a mixture of warm and cool season
grasses to ensure germination. Mulching and geotextiles can be used to help provide temporary
stabilization with vegetation, particularly in  situations where establishing cover could be
difficult.

Permanent Vegetative Stabilization. Permanent vegetative cover such as a perennial grass or a
legume cover can be used to establish a permanent vegetative cover. Permanent vegetation is
recommended for the following (Fifield 1999):

       •  Final graded or cleared areas where permanent vegetative cover is needed to stabilize
          the soil

       •  Slopes designated to be  treated with erosion control blankets

       •  Grass-lined channels or waterways designed to be protected with channel liners

The following subsections discuss the various types or means of providing vegetative
stabilization.

Grass-lined Channels

General Description
Grass-lined channels, or swales, convey stormwater runoff through a stable conduit. Vegetation
lining the channel reduces the flow velocity  of concentrated runoff. Grassed channels are usually
not designed to  control peak runoff loads by themselves and are often used in combination with
other BMPs such as subsurface drains and riprap stabilization.

Applicability
Grassed channels should be used in areas where erosion-resistant conveyances are needed, such
as in areas with highly erodible soils and slopes of less than 5 percent. They should be installed
only where space is available  for a relatively large cross-section. Grassed channels have a limited
ability to control runoff from large  storms and should not be used in areas where velocity
exceeds 5 feet per second unless they are on erosion-resistant soils with dense groundcover at the
soil surface.
                                           7-9

-------
                                                              Section 7: Technology Assessment
Design and Installation Criteria
Because of their ease of construction and low cost, vegetation-lined waterways are frequently
used for diversion and collection ditches. The U.S. Department of Agriculture's (USDA's) Soil
Conservation Service's (SCS) Engineering Field Manual (1979) recommends the maximum
permissible velocities for individual site conditions shown in Table 7-3.

    Table 7-3. Maximum permissible velocities for individual site conditions for grass swales
Site location
Areas where only a sparse cover can be established or maintained
because of shale, soils, or climate
If the vegetation is to be established by seeding
Areas where a dense, vigorous sod is obtained quickly or where the
runoff can diverted out of the waterway while the vegetation is being
established
Velocity
3.00 ft/sec (0.91 m/sec)
3.00 to 4.00 ft/sec (0.91 to 1
4.00 to 5.00 ft/sec (1 .22 to 1
.22 m/sec)
.52 m/sec)
Source: USDA 1979

Grassed waterways typically begin eroding in the invert of the channel if the velocity exceeds the
sheer strength of the vegetation soil interface. Once the erosion process has started, it will
continue until an erosion-resistant layer is encountered. If erosion of a channel bottom is
occurring, rock or stone should be placed in the eroded area or the design should be changed
(UNEP 1994).

Grassed waterways on construction land must be able to carry peak runoff events from snowmelt
and rainstorms (in some areas limited to up to 1 cubic meter of water per second). The size of the
waterway depends on the size of the area to be drained. A typical grassed waterway cross-section
is parabolic with a nearly flat-bottom, a bottom width of 3 meters (m), and channel  depth of at
least 30 centimeters (cm). Side slopes usually rise about 1 m for every 10 m horizontal distance
but could be as steep as a 1 m rise for every 2 m of horizontal distance. The waterway should
follow the natural drainage path if possible (Vanderwel and Abday 1998). The design should be
site-specific and be derived using well-established procedures.

Lined channels are a means of carrying water to lower elevations along steep parts of a
waterway. Those portions of the waterway are precisely shaped and carefully lined  with heavy-
duty erosion control matting (a geotextile product).  The lining is  covered with a layer of soil and
seeded to grass. The resulting channel is highly resistant to erosion. Lined channels are
appropriate for waterways that only carry water occasionally and have slopes of up to 10 percent.
Companies that sell geotextile products provide detailed information on installation of their
products (Vanderwel and Abday 1998). The design should be site-specific and be derived using
well-established procedures. No standard procedure is available for evaluating the effectiveness
of geotextile liners for pollutant removal.

Grass-lined channels should be sited in accordance with the natural drainage system and should
not cross ridges. The channel design should not have sharp curves or significant changes in
slope. The channel should not receive direct sedimentation from disturbed areas and should be
sited only on the perimeter of a construction site to convey relatively clean stormwater runoff.
                                           7-10

-------
                                                               Section 7: Technology Assessment
They should be separated from disturbed areas by a vegetated buffer or other BMP to reduce
sediment loads.

Although exact design criteria should be based on local conditions, basic design
recommendations for grassed channels include the following:

       •  Construction and vegetation of the channel should occur before grading and paving
          activities begin.

       •  Design velocities should be less than 5 ft/sec.

       •  Geotextiles can be used to stabilize vegetation until it is fully established.

       •  Covering the bare soil with sod or geotextiles can provide reinforced stormwater
          conveyance immediately.

       •  Triangular-shaped channels should be used with low velocities and small quantities of
          runoff; parabolic grass channels are used for larger flows and where space is
          available; trapezoidal channels are used with large flows of low velocity (low
          gradient).

       •  Outlet stabilization structures might be needed if the runoff volume or velocity has
          the potential to exceed the capacity of the receiving area.

       •  Channels should be designed to convey runoff from a 10-year storm without erosion.

       •  The sides of the channel should be sloped less than 3:1, with V-shaped channels
          along roads sloped 6:1 or less for safety.

       •  All trees, bushes, stumps, and other debris should be removed during construction.

Effectiveness
Grass-lined channels can effectively transport stormwater from construction areas if they are
designed for expected flow volumes and velocities and if they do not receive sediment directly
from disturbed areas. The primary function is to carry the flow at a higher velocity without
eroding or overtopping the channel.

Limitations
Grassed channels, if improperly installed, can alter the natural flow of surface water and have
adverse effects on downstream waters. Additionally, if the  design capacity is exceeded by a large
storm event, the vegetation might not be sufficient to prevent erosion, and the channel might be
destroyed. Clogging with sediment and debris reduces the effectiveness of grass-lined channels
for stormwater conveyance.

Maintenance
Maintenance requirements for grass channels are relatively minimal. During the vegetation
establishment period, the channels should be inspected after every rainfall. Other maintenance
activities that should be carried out after vegetation is established are mowing, litter removal, and
spot vegetation replacement. The most important objective in the maintenance of grassed
channels is the maintaining of a dense and vigorous growth of turf. Periodic cleaning of
vegetation and soil buildup in curb cuts is required so that water flow into the channel is
                                           7-11

-------
                                                              Section 7: Technology Assessment
unobstructed. During the growing season, channel grass should be cut no shorter than the level of
design flow, and the cuttings should be removed promptly.

Cost
Costs of grassed channels range according to depth, with a 1.5-foot-deep, 10-foot-wide grassed
channel estimated to cost between $6,395 and $17,075 per trench, while a 3-foot-deep, 21-foot-
wide grassed channel is estimated at $12,909 to $33,404 per trench (SWRPC 1991).

As an alternative cost approximation, grassed channel construction costs can be developed using
unit cost values. Shallow trenching (1 to 4 feet deep) with a backhoe in areas not requiring
dewatering can be performed for $4 to $5 per cubic yard of removed material (R.S. Means
2000). Assuming no disposal costs (i.e., excavated material is placed on either side of the
trench), only the cost of fine grading, soil treatment, and grassing (approximately $2 per square
yard of earth surface area) should be added to the trenching cost to approximate the total
construction cost. Site-specific hydrologic analysis of the construction site is necessary to
estimate the channel conveyance requirement; however, it is not unusual to have flows on the
order of 2 to 4 cfs per acre served. For channel velocities between 1 and 3 feet per second, the
resulting range in the channel cross-section area can be as low as 0.67 square foot per acre
drained to as high as 4 square feet per acre. If the average channel flow depth is 1 foot, the low
estimate for grassed channel installation is $0.27 per square foot of channel bottom per acre
served per foot of channel  length. The high estimate is $1.63 per square foot of channel bottom
per acre served per foot of channel length.

Seeding

General Description
Permanent seeding is used to control runoff and erosion on disturbed areas by establishing
perennial vegetative cover from seed. It is used to reduce erosion, decrease sediment yields from
disturbed areas, and provide permanent stabilization. This practice is both economical and
adaptable to different site conditions, and it allows selection of the most appropriate plant
materials.  Seeding is a BMP that is particularly susceptible to local conditions such as the
climatic conditions, physical and chemical characteristics of the soil, topography, and time of
year.

Applicability
Permanent seeding is well  suited in areas where permanent, long-lived vegetative cover is the
most practical or most effective method of stabilizing the soil.  Permanent seeding can be used on
roughly graded areas that will not be regraded for at least a year. Vegetation controls erosion by
protecting bare soil surfaces from displacement by raindrop impacts and by reducing the velocity
and quantity of overland flow. The advantages of seeding over other means of establishing plants
include lower initial costs and labor inputs.

Design and Installation Criteria
Areas to be stabilized with permanent vegetation  must be seeded or planted 1 to 4 months after
the final grade is achieved unless temporary stabilization measures are in place.  Successful plant
establishment can be maximized with proper planning; considering soil characteristics; selecting
                                           7-12

-------
                                                               Section 7: Technology Assessment
plant materials that are suitable for the site; adequate seedbed preparation, liming, and
fertilization; timely planting; and regular maintenance. Climate, soils, and topography are major
factors that dictate the suitability of plants for a site. The soil on a disturbed site could require
amendments to provide sufficient nutrients for seed germination and seedling growth. The
surface soil must be loose enough for water infiltration and root penetration. Soil pH  should be
between 6.0 and 6.5 and can be increased with liming if soils are too acidic. Seeds can be
protected with mulch to retain moisture, regulate soil temperatures, and prevent erosion during
seedling establishment.

Seedbed preparation is critical in established vegetation.  Spraying seeds on a scraped slope will
generally not provide satisfactory results. Typical seedbed preparation will begin with a soil test
to determine the amount of lime or fertilizer that should be added. In addition, tillage should be
performed that will break up clods so that seed contact can be established. When the seed is
applied, it should be covered and lightly compacted. Natural or synthetic mulch is recommended
to provide surface stabilization until the vegetation is established. In addition to providing
surface stabilization, the mulch will  also retard evaporation and encourage rapid growth. A
suitable tack to hold the mulch might be necessary if the mulch is not otherwise anchored. Mulch
as an erosion control practice is covered in a subsequent  subsection.

Depending on the amount of use permanently seeded areas receive, they can be considered high- or
low-maintenance areas. High-maintenance areas are mowed frequently, limed and fertilized
regularly, and either (1) receive intense use (for example, athletic fields) or (2) require maintenance
to an aesthetic standard (for example, home lawns). Grasses used for high-maintenance areas are
long-lived perennials that form a tight sod and are fine-leaved. High-maintenance vegetative cover
is used for homes, industrial parks, schools, churches, and recreational areas.

Low-maintenance areas are mowed infrequently or not at all and do not receive lime or fertilizer
regularly. Plants must be able to persist with minimal maintenance over long periods. Grass and
legume mixtures are favored for these sites because legumes fix nitrogen from the atmosphere.
Sites suitable for low-maintenance vegetation include steep slopes, streambanks or channel
banks, some commercial  properties, and utility turf areas such as road banks.

Effectiveness
Seeding that results in a successful stand of grass has been  shown to remove between 50 and 100
percent of TSS from stormwater runoff, with an average  removal of 90 percent (USEPA 1993).

Limitations
The effectiveness of permanent seeding can be limited because of the high erosion potential
during establishment, the need to reseed areas that fail to establish, limited seeding times
depending on the season, and the need for stable soil temperature and soil moisture content
during germination and early growth. Permanent seeding does not immediately stabilize  soils—
temporary ESC measures should be  in place to prevent off-site transport of pollutants from
disturbed areas. Use of mulches or geotextiles or both could improve the likelihood of
successfully establishing vegetation.
                                           7-13

-------
                                                              Section 7: Technology Assessment
Maintenance
Grasses should emerge within 4 to 28 days and legumes within 5 to 28 days after seeding. A
successful stand should exhibit the following:

       •  Vigorous dark green or bluish green seedlings—not yellow

       •  Uniform density, with nurse plants, legumes, and grasses well intermixed

       •  Green leaves—perennials remaining throughout the summer, at least at the plant bases

Seeded areas should be inspected for failure, and necessary repairs and reseeding should be made
as soon as possible. If a stand has inadequate cover, the choice of plant materials and quantities
of lime and fertilizer should be reevaluated. Depending on the condition of the stand, areas can
be repaired by overseeding or reseeding after complete seedbed preparation. If the timing is bad,
an annual grass seed can be overseeded to temporarily thicken the stand until a suitable time for
seeding perennials. Consider seeding temporary, annual species if the season is not appropriate
for permanent seeding. If vegetation fails to grow, the soil should be tested to determine whether
low pH or nutrient imbalances are responsible. Local NRCS or county extension agents can also
be contacted for seeding and soil testing recommendations.

On a typical disturbed site, full plant establishment usually requires refertilization in the second
growing season. Soil tests should be used to determine whether more fertilizer needs to be added.
Do not fertilize cool season grasses in late May through July. Grass that looks yellow might be
nitrogen deficient. Nitrogen fertilizer should not be used if the stand contains more than 20
percent legumes.

Cost
Seeding costs range from $200 to $1,000 per acre and average $400 per acre. Maintenance costs
range from 15 to 25 percent of initial costs and average 20 percent (USEPA 1993). R.S. Means
(2000) indicates the cost of mechanical seeding to be approximately $900 per acre and
demonstrates that the coverage cost varies with the seed type, seeding approach and scale (total
acreage to be seeded). For example, hydro or water-based seeding for grass is estimated to be
$700 per acre, but seeding offield grass species is only $540 per acre (Costs include materials,
labor, and equipment,  with profit and overhead). If surface preparation is required, the
installation costs increase. R.S. Means suggests the cost of fine grading, soil treatment, and
grassing is approximately $2 per square yard.

Sodding

General Description
Sodding is a permanent erosion control practice that involves laying a continuous cover of grass
sod on exposed soils. In addition to stabilizing soils, sodding can reduce the velocity of stormwater
runoff. Sodding can provide immediate vegetative cover for critical areas and stabilize areas that
cannot be vegetated by seed. It can also stabilize channels or swales that convey concentrated
flows and reduce flow velocities. While sodding is not as dependent on local conditions as seeding
is, it does depend on soil and climatic conditions to be successful. Watering immediately after
installation and occasionally until establishment is generally beneficial.
                                           7-14

-------
                                                              Section 7: Technology Assessment
Applicability
Sodding is appropriate for any graded or cleared area that might erode, requiring immediate
vegetative cover. Locations particularly well suited to sod stabilization are the following:

       •  Waterways and channels carrying intermittent flow

       •  Areas around drop inlets that require stabilization

       •  Residential or commercial lawns and golf courses where prompt use and aesthetics
          are important

       •  Steeply sloped areas

Design and Installation Criteria
Sodding eliminates the need for seeding and mulching and produces more reliable results with
less maintenance. Sod can be laid during times of the year when seeded grasses can fail. The sod
must be watered frequently within the first few weeks of installation. Some seedbed preparation
is recommended, including smoothing to provide contact between the sod and the soil surface
and soil testing to determine liming and fertilizer application rates. Because sod provides
instantaneous cover, mulches are not typically recommended, but anchoring might be
appropriate on steep slopes.

The type of sod selected should be  composed of plants adapted to site conditions. Sod
composition should reflect environmental conditions as well as the function of the area where the
sod will be laid. The sod should be of known genetic origin and be free of noxious weeds,
diseases, and insects. The sod should be machine cut at a uniform soil thickness of 15 to 25 mm
at the time of establishment (this does not include top growth or thatch). Soil preparation and
addition of lime and fertilizer could be needed—soils should be tested to determine whether
amendments are needed. Sod should be laid in strips perpendicular to the direction of water flow
and staggered in a brick-like pattern.  The corners and middle of each strip should be stapled
firmly. Jute or plastic  netting can be pegged over the sod for further protection against washout
during establishment.

Areas to be sodded should be cleared of trash, debris, roots, branches, stones, and clods larger
than 2 inches in diameter.  Sod should be harvested, delivered, and installed within a period of 36
hours. Sod not transplanted within this period should be inspected and approved before its
installation.

Limitations
Compared to seed, sod is more expensive and more difficult to obtain, transport, and store. Care
must be taken to prepare the soil and provide adequate moisture before, during, and after installation
to ensure successful establishment. If sod is laid on poorly prepared soil or unsuitable surface, the
grass will die quickly  because it is unable to root. Sod that is not adequately irrigated after
installation can cause  root dieback because grass does not root rapidly and is subject to drying.

Effectiveness
Sod has been shown to remove between 98 and 99 percent of TSS in runoff (USEPA 1993). It is
therefore a highly effective management practice for ESC.
                                           7-15

-------
                                                               Section 7: Technology Assessment
Maintenance
Watering is very important to maintain adequate moisture in the root zone and to prevent dormancy,
especially within the first few weeks of installation, until it is fully rooted. Mowing should not result
in the removal of more than one-third of the shoot. Grass height should be maintained to be 2-3
inches long. After the first growing season, sod might require fertilization or liming.

Permanent, fine turf areas require yearly fertilization. Warm-season grass should be fertilized in
late spring to early summer, and cool-season grass in late winter and again in early fall.

Cost
Average installation costs of sod average $0.20 per square foot  and range from $0.10 to $1.10
per square foot; maintenance costs are approximately 5 percent of installation costs (USEPA
1993). R.S. Means (2000) indicates the sodding ranges between $250 and $750 per 1,000 square
feet for  1-inch deep bluegrass sod on level ground, depending on the size of the area treated (unit
costs value are for orders over 8,000 square feet and less than 1,000 square feet, respectively).
Bent grass sod values range between $350 and $500 per 1,000 square feet; again, the lower value
is more  likely for most construction  sites because it is for large  area applications. (Costs include
materials, labor, and equipment, with profit and overhead).

Mulching

General  Description
Mulching is a temporary erosion control practice  in which materials such as grass, hay, wood
chips, wood fibers, straw, or gravel are placed on exposed or recently planted soil surfaces.
Mulching is highly recommended as a stabilization method and is most effective when anchored
in place until vegetation is well established. In addition to stabilizing soils, mulching can reduce
the velocity of stormwater runoff. When used in combination with seeding or planting, mulching
can aid plant growth by holding seeds, fertilizers, and topsoil in place; by preventing birds from
eating seeds; by retaining moisture; and by insulating plant roots against extreme temperatures.

Mulch mattings are materials  such as jute or other wood fibers that are formed into sheets and
are more stable than loose mulch. They can also be easily unrolled during the installation process
and are particularly useful in steeper areas or in channels. Netting can be used to stabilize soils
while plants are growing, although netting does not retain moisture or insulate against extreme
temperatures. Mulch binders consist of asphalt or synthetic materials that are sometimes used
instead of netting to bind loose mulches, but they have been found to have limited usefulness.

Applicability
Mulching is often used in areas where temporary  seeding cannot be used because of
environmental constraints. Mulching can provide immediate, effective, and inexpensive erosion
control.  On steep slopes and critical  areas such as waterways, mulch matting is used with netting
or anchoring to hold it in place. Mulches can be used on seeded and planted areas where slopes
are steeper than 2:1 or where sensitive seedlings require insulation from extreme temperatures.
                                           7-16

-------
                                                              Section 7: Technology Assessment
Design and Installation Criteria
When possible, organic mulches should be used for erosion control and plant material
establishment. Suggested materials include loose straw, netting, wood cellulose, or agricultural
silage. All materials should be free of seed, and loose hay or straw should be anchored by
applying tackifier, stapling netting over the top, or crimping with a mulch crimping tool.
Materials that are heavy enough to  stay in place do not need anchoring (for example, gravel).
Steepness of the slope will also affect the extent of anchoring the mulch. Other examples include
hydraulic mulch products with 100 percent post-consumer paper content, yard trimming
composts, and wood mulch from recycled stumps and tree parts. Inorganic mulches such as pea
gravel or crushed granite can be used in unvegetated areas.

Mulches might require a binder, netting, or tacking. All straw and loose materials must have a
binder to hold them in place. Mulch materials that  float away during storms can clog drainage
ways and lead to flooding. The extent of binding depends on the type of mulch applied. Effective
use of netting and matting material requires firm, continuous contact between the materials and
the soil. If there is no contact, the material will not hold the soil, and erosion will occur
underneath the material. Grading is not necessary before mulching.

There must be adequate coverage, or erosion, washout, and poor plant establishment will result.
If an appropriate tacking agent is not applied or if it is applied in an insufficient amount, mulch
will not withstand wind and runoff. The channel grade and liner must be appropriate for the
amount of runoff or the channel bottom will erode. Also, hydromulch should be applied in
spring, summer, or fall to prevent deterioration of the mulch before plants can become
established.  Table 7-4 presents guidelines for installing mulches, but local conditions could
warrant additional requirements.
                  Table 7-4. Typical mulching materials and application rates
Material
Rate per acre
Requirements
Notes
Organic mulches
Straw
Wood fiber or
wood cellulose
Wood chips
Bark
1-2 tons
0.5-1 ton
5-6 tons
35yd3
Dry, unchopped, unweathered;
avoid weeds.

Air dry. Add fertilizer N, 12
Ib/ton.
Air dry, shredded or
hammermilled, or chips.
Spread by hand or machine; must be
tacked or tied down.
Use with hydroseeder; can be used to
tack straw. Do not use in hot, dry
weather.
Apply with blower, chip handler, or by
hand. Not for fine turf areas.
Apply with mulch blower, chip handler,
or by hand. Do not use asphalt tack.
Nets and mats
Jute net
Excelsior (wood
fiber) mat
Fiberglass
roving
Cover area
Cover area
0.5-1 ton
Heavy, uniform; woven of single
jute yarn. Used with organic
mulch.

Continuous fibers of drawn
glass bound together with a
nontoxic agent.
Withstands water flow.

Apply with compressed air ejector. Tack
with emulsified asphalt at a rate of 25-
35 gal/1 ,000 ft2 .
                                           7-17

-------
                                                              Section 7: Technology Assessment
Effectiveness
Mulching effectiveness varies with the type of mulch used and local conditions such as rainfall
and runoff amounts. Percent soil loss reduction for different mulches ranges from 53 to 99.8
percent, and associated water velocity reductions range from 24 to 78 percent (Harding 1990).
Table 7-5 shows soil loss and water velocity reductions for different mulch treatments.

           Table 7-5. Measured reductions in soil loss for different mulch treatments
Mulch characteristics
1 00% wheat straw/top net
1 00% wheat straw/two nets
70% wheat straw, 30% coconut fiber
70% wheat straw, 30% coconut fiber
1 00% coconut fiber
Nylon monofilament/two nets
Nylon monofilament/rigid/bonded
Vinyl monofilament/flexible/bonded
Curled wood fibers/top net
Curled wood fibers/two nets
Antiwash netting (jute)
Interwoven paper and thread
Uncrimped wheat straw-2,242 kg/ha
Uncrimped wheat straw-4,484 kg/ha
Soil loss reduction
(%)
97.5%
98.6%
98.7%
99.5%
98.4%
99.8%
53.0%
89.6%
90.4%
93.5%
91 .8%
93.0%
84.0%
89.3%
Water velocity reduction (%)
relative to bare soil
73%
56%
71%
78%
77%
74%
24%
32%
47%
59%
59%
53%
45%
59%
Source: Harding 1990, as cited in USEPA 1993

Limitations
Mulching, matting, and netting might delay seed germination because the cover changes soil
surface temperatures. The mulches themselves are subject to erosion and could be washed away
in a large storm if not sufficiently anchored with netting or tacking. Maintenance is necessary to
ensure that mulches provide effective erosion control.

Maintenance
Mulches must be anchored to resist wind displacement. Netting should be removed when
protection is no longer needed and disposed of in a landfill or composted. Mulched areas should
be inspected frequently to identify areas where mulch has loosened or been removed, especially
after rain storms. Such areas  should be reseeded (if necessary) and the mulch cover replaced
immediately. Mulch binders  should be applied at rates recommended by the manufacturer. If
washout, breakage, or erosion occurs, surfaces should be repaired, reseeded, and remulched, and
new netting should be installed. Inspections should be continued until vegetation is firmly
established.

Cost
The costs for various types of mulches vary by the type of material and also whether seeding is
incorporated. The costs of seed and mulch average $1,500 per acre and range from $800 to
                                          7-18

-------
                                                               Section 7: Technology Assessment
$3,500 per acre (USEPA 1993). Ground hydromulching applied between fiscal year 2000 and
2003 in the southwestern United States had a cost range of $1,675 to $3,000 per acre (Napper
2006). The California Stormwater Quality Association's (CSQA's) Stormwater BMP Handbook:
Construction reports the average cost of installing wood fiber mulch as $900 per acre (CSQA
2003). R.S. Means (2000) estimates the cost of power mulching to be $22.50 per 1,000 square
feet, for large volume applications.

Vegetated Buffer Strips

General Description
Vegetated buffers are areas of either natural or established vegetation that are maintained to
protect the water quality of neighboring areas. Buffer zones reduce the velocity of Stormwater
runoff, provide an area for the runoff to permeate the soil, allow ground water recharge, and act
as filters to catch sediment. The reduction in velocity also helps to prevent soil erosion.

Applicability
Vegetated buffers can be used in any area that is able to support vegetation, but they are most
effective and beneficial on floodplains, near wetlands, along streambanks, and on steep, unstable
slopes. They are also effective in separating land use areas that are not compatible and in
protecting wetlands or waterbodies by displacing activities that could be sources of nonpoint
source pollution.

Design and Installation Criteria
To establish an effective vegetative buffer, the following guidelines should be followed:

       •  Soils should not be compacted.

       •  Slopes should be less than 5 percent.

       •  Buffer widths should be determined after careful consideration of slope, vegetation,
          soils, depth to impermeable layers, runoff sediment characteristics, type and quantity
          of Stormwater pollutants, and annual rainfall.

       •  Buffer widths should increase as slope increases.

       •  Zones of vegetation (native vegetation in particular), including grasses, deciduous and
          evergreen shrubs, and understory and overstory trees, should be intermixed.

       •  In areas where flows are concentrated and velocities are high, buffer zones should be
          combined with other structural or nonstructural BMPs as a pretreatment.

Vegetated strips have been  studied extensively, with emphasis placed on their effectiveness in
removing  sediment and other pollutants. Vegetated strips are most appropriate at sites where
sediment loads are relatively low, because high sediment loads will cause large quantities of
deposition along the leading edge of the vegetation. This deposition will cause the flow to divert
around the vegetation in a concentrated flow pattern, which will cause short-circuiting and
greatly reduce removal efficiency. Variability in vegetation density and uniformity often causes
similar problems. Removal  efficiency  depends on a combination of slope, length, and width of
the filter; density of the vegetation; sediment characteristics, hydraulics of the flow; and
infiltration. The interaction of these variables is complex and prevents the process from being
                                           7-19

-------
                                                              Section 7: Technology Assessment
reduced to a simple relationship except on a local basis. For site-specific local conditions,
methods have been developed that allow trapping to be related to strip length and slope.

Effectiveness
Considerable data have been collected on the effectiveness of buffer strips for specific
conditions. Numerous factors such as infiltration rate, flow depth, slope, dimensions of the
buffer, density and type of vegetation, sediment size, and sediment density impact removal rates.
Recent studies show that even short vegetative buffers can trap high percentages of sediment and
certain chemicals. A significant concern is preventing flow from concentrating to maintain
adequate the travel time through the buffer to allow the removal of pollutants.

Several researchers have measured greater than 90 percent reductions in sediment and nitrate
concentrations; buffer/filter strips do a reasonably good job of removing phosphorus attached to
sediment,  but they are relatively ineffective in removing dissolved phosphorus (Gillman 1994).
However,  because the hydraulics of flow through buffer strips are not well defined and can vary
considerably by site conditions, it is difficult to consistently estimate the effectiveness of buffer
strips.

Limitations
Vegetated buffers require plant growth before they can be effective, and land must be available
on which to plant the vegetation. If land costs  are very high, buffer zones might not be cost-
effective.  Although vegetated buffers help to protect water quality, they usually do not
effectively mitigate concentrated stormwater flows to neighboring or downstream wetlands.

Maintenance
Keeping the vegetation in vegetated buffers healthy requires routine maintenance, which
(depending on species, soil types, and climatic conditions) can include weed and pest control,
mowing, fertilizing, liming, irrigating, and pruning. Inspection and maintenance are most
important  when buffer areas are first installed. Once established, vegetated buffers do not require
much maintenance beyond the routine procedures listed earlier and periodic inspections of the
areas, especially after any heavy rainfall and at least once  a year. Inspections should focus on
encroachment, gully erosion, density of vegetation, evidence of concentrated flows through the
areas, and any damage from foot or vehicular traffic. If there are more than 6 inches of sediment
in one place, it should be removed.

Cost
Conceptual cost estimates for grassed buffer strips can be  made on the basis of square footage
using unit cost values. R.S. Means (2000) estimates the cost of fine grading, soil treatment, and
grassing to be $2 per square yard. This cost estimate is based on applying traditional lawn seed.
The cost for field seed is lower than lawn seed, reducing the coverage price. Where gently
sloping areas need to be grassed only with acceptable species, the cost can be as low as $0.38 per
square yard.
                                           7-20

-------
                                                               Section 7: Technology Assessment
      7.2.1.3.   Non-Vegetative Stabilization

Non-vegetative practices can also be used during construction to stabilize and protect soil
exposed to the erosive forces of water, as well as post-construction to provide a filtration
mechanism for stormwater pollutants. Non-vegetative stabilization techniques operate on the
same principles as vegetative stabilization; however, these practices use a variety of synthetic on
natural materials (such as coconut fiber) to stabilize exposed soils. Non-vegetative practices are
particularly useful as temporary stabilization measures until vegetative practices have had a
chance to become established.  The following discussion refers to non-vegetative stabilization as
a construction BMP that stabilizes and protects soil from erosion. A variety of proprietary and
vendor-supplied materials are in this category, which are not discussed in detail.

Geotextiles

General Description
Geotextiles are porous fabrics also known as filter fabrics, road rugs, synthetic fabrics,
construction fabrics, or simply fabrics. Geotextiles are manufactured by weaving or bonding
fibers made from synthetic materials such as polypropylene, polyester, polyethylene, nylon,
polyvinyl chloride, glass, and various mixtures of such materials. As a synthetic construction
material, geotextiles are used for a variety of purposes such as separators, reinforcement,
filtration and drainage, and erosion control (USEPA 1992). Some geotextiles are made of
biodegradable materials such as mulch matting and netting. Mulch mattings are jute or other
wood fibers that have been formed into sheets and are more stable than normal mulch. Netting is
typically made from jute, wood fiber, plastic, paper, or cotton and  can be used to hold the
mulching and matting to the ground. Netting can also be used alone to stabilize soils while the
plants are growing; however, it does not retain moisture or temperature well.

Geotextiles can aid in plant growth by holding seeds, fertilizers, and topsoil in place. Fabrics are
relatively inexpensive for certain applications—a wide variety of geotextiles exist to match the
specific needs of the site.

Applicability
Geotextiles can be used for erosion control by using it alone. Geotextiles can be used as matting,
which is used to stabilize the flow of channels or swales or to protect seedlings on recently
planted slopes until they become established. Matting can be used  on tidal or streambanks where
moving water is likely to wash out new plantings. They can also be used to protect exposed soils
immediately and temporarily, such as when active piles of soil are left overnight.

Geotextiles are also used as separators. An example of such a use is geotextile as a separator
between riprap and soil. This sandwiching prevents the soil from being eroded from beneath the
riprap and maintaining the riprap's base.

Design and  Installation Criteria
Many types of geotextiles are available.  Therefore, the selected fabric should match its purpose.
State or local requirements, design procedures, and any other applicable requirements should be
considered. In the field, important concerns include regular inspections to determine whether
cracks, tears, or breaches are present in the fabric and to identify when repairs should be made.
                                           7-21

-------
                                                              Section 7: Technology Assessment
Effective netting and matting require firm, continuous contact between the materials and the soil.
If there is no contact, the material will not hold the soil, and erosion will occur underneath the
material.

Effectiveness
A geotextile's effectiveness depends on the strength of the fabric and proper installation. For
example, when protecting a cut slope with a geotextile, it is important to properly anchor the
fabric using appropriate length and spacing of wire  staples. This will ensure that it will not be
undermined by a storm event.

Limitations
Geotextiles (primarily synthetic types) have the potential disadvantage of being sensitive to light
and must be protected before installation. Some geotextiles might promote increased runoff and
can blow away if not firmly anchored. Depending on the type of material used, geotextiles might
need to be disposed of in a landfill, making them less desirable than vegetative stabilization. If
the fabric is not properly selected, designed, or installed, the effectiveness can be reduced
drastically.

Maintenance
Regular inspections should be made to determine whether cracks, tears, or breaches have formed
in the fabric—it should be repaired or replaced immediately. It is necessary to maintain contact
between the ground and the geotextile at all times.

Cost
Costs for geotextiles range from $0.50 to $10.00 per square yard depending on the type chosen
(SWRPC 1991).

Erosion Control Matting

General Description
Erosion control mats can be either organic or made  from a synthetic material. A wide variety of
products exist to match the specific needs of the site. Organic mats are made from such materials
as wood fiber, jute net, and coconut coir fiber. Unlike organic matter, synthetic mats are
constructed from non-biodegradable materials and remain in place for many years. These organic
mats are classified as turf reinforcement mats (TRMs) and erosion control and revegetation mats
(ECRMs) (USDOT 1995).

Erosion control matting aids in plant growth by holding seeds, fertilizers, and topsoil in place.
Matting can be used to stabilize the flow of channels or swales or to protect seedlings on recently
planted slopes until they become established. Matting can be used on tidal or streambanks where
moving water is likely to wash out new plantings. It can also be used to protect exposed soils
immediately and temporarily, such as when active piles of soil are left overnight.

Applicability
Mulch mattings, netting, and filter fabrics are particularly useful in steep areas and drainage
swales where loose seed is vulnerable to being  washed away or failing to survive dry soil (UNEP
                                           7-22

-------
                                                              Section 7: Technology Assessment
1994). Erosion control mats can also be used to separate riprap and soil. That results in a
sandwiching effect, maintaining the riprap's base and preventing the soil beneath from being
eroded.

Design and Installation Criteria
Matting is especially recommended for steep slopes and channels (UNEP 1994). Many types of
erosion control mats are available. Therefore, the selected product should match its purpose.
Effective netting and matting require firm, continuous contact between the materials and the soil.
If there is no contact, the material will not hold the soil, and erosion will occur underneath the
material.

Wood fiber or curled wood mat consists of curled wood with fibers, 80 percent of which are 150
mm or longer, with a consistent thickness and even distribution of fiber over the entire mat. The
top side of the mat is covered with a biodegradable plastic mesh. The mat should be placed in the
channel or on the slope parallel to the direction of flow and secured with staples and check slots.
It should be  applied immediately after seeding operations (USDOT 1995).

Jute net consists of jute yarn, approximately 5 mm in diameter, woven into a net with openings
that are approximately 10 by 20 mm (or 0.40 to 0.79 inches). The jute net should be loosely laid
in the channel parallel to the direction of flow. The net is secured with staples and check slots at
intervals along the channel. Placement of the jute net should be done immediately after seeding
operations (USDOT 1995).

Coconut blankets are constructed of biodegradable coconut fibers that resist decay for 5 to 10
years to provide long, temporary erosion control protection. The materials are often encased in
ultraviolet stabilized nets and sometimes have a composite, polypropylene structure to provide
permanent turf reinforcement. These materials are best used for waterway stabilization and
slopes that require longer periods to stabilize (USDOT 1995).

Within the synthetic mat category are TRMs and ECRMs. TRMs are three-dimensional polymer
nettings or monofilaments formed into a mat. They have sufficient thickness (> 13 mm or 0.5
inch) and void space (> 90 percent) to allow for soil filling and retention. The mat acts as a
traditional mat to protect the seed and increase germination. As the turf establishes, the mat
remains in place as part of the root structure. That gives the established turf a higher strength and
resistance to erosion (USDOT 1995).

ECRMs are  composed of continuous monofilaments bound by heat fusion or stitched between
nettings. They are thinner than TRMs and do not have the void space to allow for filling of soil.
They act as permanent mulch and allow vegetation to grow through the mat (USDOT 1995).

Effectiveness
The effectiveness of erosion control matting depends on the strength of the material and proper
installation.  For example, when protecting a cut slope with an erosion control mat, it is important
to anchor the mat properly. That ensures that it will not be undermined  by a storm event.

While erosion control blankets can be effective, their performance varies. Some general trends
are that organic materials tend to be the most effective (Harding 1990) and that thicker materials
                                          7-23

-------
                                                              Section 7: Technology Assessment
are typically superior (Fifield 1999), but there are exceptions to both of these trends. Information
about product testing of blankets is generally lacking. One notable exception is the Texas
Department of Transportation, which publishes the findings of its testing program in the form of
a list of acceptable and unacceptable materials for specific uses.

Limitations
Erosion control mats (primarily synthetic types) are sensitive to light and for that reason must be
protected before installation. Some erosion control mats might cause an increase in runoff or
blow away if not firmly anchored. Erosion control mats might need to be properly disposed of in
a landfill,  depending on the type of material. Effectiveness could be reduced if the matting is not
properly selected, designed, or installed.

Maintenance
Regular inspections are necessary  to determine whether cracks, tears or breaches have formed in
the matting. Contact between the ground and erosion control mat should be maintained at all
times and  trapped sediment removed after each storm event.

Cost
Costs for erosion control mats range from $0.50 to $10.00 per square yard depending on the type
chosen (SWRPC 1991). Geosynthetic TRMs are widely used for immediate erosion protection
and long-term vegetative reinforcement, usually for steeply sloped areas or areas exposed to
runoff flows.  The Erosion Control Technology  Council (ECTC—a geotextile industry support
association) estimates that TRMs cost approximately $7.00 per square yard (installed) for
channel protection (Lancaster et al. 2002). Channel protection is one of the most demanding of
installations (much more demanding than general coverage of denuded area). The ECTC
estimates the  cost to install a simple soil blanket (or rolled erosion control product), seed, and
fertilizer to be $1.00 per square  yard (Honnigford 2002).

USDA's Burned Area Emergency Response (BAER) Treatment Catalog reports that most rolled
erosion control products are priced by the square yard and sold in rolls with prices, without
installation, ranging from $0.35 to  $0.50 per square yard to more than $1 per square yard (Napper
2006). CSQA's Stormwater BMP Handbook:  Construction reports material costs of $0.50 to
$0.57 per square yard for biodegradable materials, $3.00 to $4.50 per square yard for permanent
materials,  and $0.04 to $0.05 per staple (CSQA 2003). The installation cost for jute mesh,
100 square yards per roll, 4 inches  wide, stapled is $0.49 per square yard (R.S. Means 2009).

Topsoiling

General Description
Topsoiling is the placement of a surface layer of soil enriched in organic matter over a prepared
subsoil to  provide a suitable soil medium for vegetative growth on areas with poor moisture, low
nutrient levels, undesirable pH,  or the presence of other materials that would inhibit the
establishment of vegetation. Advantages of topsoil include its high organic matter content and
friable consistency and its water-holding capacity and nutrient content. The texture and friability
of topsoil  are usually more conducive to seedling emergence and root growth. In addition to
being a better growth medium, topsoil is often less erodible than subsoils, and the coarser texture
                                           7-24

-------
                                                               Section 7: Technology Assessment
of topsoil increases infiltration capacity and reduces runoff. During construction, topsoil is often
removed from the project area and stockpiled. It is replaced on areas to be grassed or landscaped
during the final stages of the project.

Applicability
Conditions where topsoiling applies include the following:

       •  Where a sufficient supply of quality topsoil is available

       •  Where the subsoil or areas of existing surface soil present the following problems:
          o  The structure, pH, or nutrient balance of the available soil cannot be amended by
              reasonable means to provide an adequate  growth medium for the desired
              vegetation
          o  The soil is too shallow to provide adequate rooting depth or will not supply
              necessary moisture and nutrients for growth of desired vegetation
          o  The soil contains substances toxic to the desired vegetation

       •  Where high quality turf or ornamental plants  are desired
       •  Where slopes are 2:1 or flatter

Design and Installation Criteria
The topsoil should be uniformly distributed over the subsoil to a minimum compacted depth of
50 mm (2 inches) on slopes steeper than 3:1 and 100 mm (4 inches) on flatter slopes.
Thicknesses of 100 to 150 mm are preferred for vegetation establishment via seeding. The
topsoil should not be placed while in a frozen or muddy  condition or when the subsoil is
excessively wet, frozen, or in a condition that is detrimental to proper grading or seedbed
preparation. The final surface should be prepared so that any irregularities are corrected and
depressions and water pockets do not form. If the topsoil has been treated with soil sterilants, it
should not be placed until the toxic substances have  dissipated (USDOT  1995). Table 7-6
summarizes the cubic yards of topsoil required for application to various depths.

                         Table 7-6. Cubic yards of topsoil required for
                                application to various depths
Depth
(inches)
1
2
3
4
5
6
Perl, 000 sq ft
3.1
6.2
9.3
12.4
15.5
18.6
Per acre
134
268
403
536
670
804
                       Source: Smolen et al. 1988.
On slopes and areas that will not be mowed, the surface could be left rough after spreading
topsoil. A disk can be used to promote bonding at the interface between the topsoil and subsoil
(Smolen etal. 1988).
                                           7-25

-------
                                                              Section 7: Technology Assessment
Effectiveness
No information is available describing the effectiveness of applying topsoil as a BMP.

Limitations
Limitations of applying topsoil can include the following:

       •  Topsoil spread when conditions are too wet, resulting in severe compaction

       •  Topsoil mixed with too much unsuitable subsoil material, resulting in poor vegetation
          establishment

       •  Topsoil contaminated with soil sterilants or chemicals, resulting in poor or no
          vegetation establishment

       •  Topsoil not adequately incorporated or bonded with the subsoil, resulting in poor
          vegetation establishment and soil slippage on sloping areas

       •  Topsoiled areas not protected, resulting in excessive erosion

Maintenance
Newly topsoiled areas should be inspected frequently until the vegetation is established. Eroded
or damaged areas should be repaired and revegetated.

Cost
Topsoiling costs are a function of the price of topsoil, the hauling distance, and the method of
application. R.S. Means (2000) reports unit cost values of $3 and $4 per square yard, for 4 and 6
inches of topsoil cover, respectively. That price is for furnishing and placing of topsoil, and
includes materials, labor, and equipment, with profit and overhead.

   7.2.2.   WATER HANDLING PRACTICES

      7.2.2.1.   Earth Dike

General Description
An earth dike is a temporary or permanent ridge of soil designed to channel water  to a desired
location. Dikes are used to divert the flow of runoff by constructing a ridge of soil  that intercepts
and directs the runoff to the desired outlet or alternative management practice, such as a pond.
The practice serves to reduce the length of a slope for erosion control and protect downslope
areas. An earth dike can be used to prevent runoff from going over the top of a cut and eroding
the slope, directing runoff away from a construction site or building; to divert clean water from a
disturbed area; or to reduce a large drainage area into a more manageable size. Dikes should be
stabilized with vegetation after construction (NAHB, No Date).

Applicability
Earth dikes are applicable to all areas; the size of the dike is correlated to the size of the drainage
area (NAHB, No Date).
                                           7-26

-------
                                                               Section 7: Technology Assessment
Design and Installation Criteria
The location of dikes should take into consideration outlet conditions, existing land use,
topography, length of slope, soils, and development plans. The capacity of earth dikes and
diversions should be suitable for the area that is being protected, including adequate freeboard, or
extra depth that is added as a safety margin. For homes, schools, and industrial buildings, the
recommended design frequency storm is 50 years and the freeboard is 0.5 feet (NAHB, No Date).

Earth dikes can be employed as a perimeter control. For small sites, a compacted, 2-foot-tall dike
is usually suitable if hydroseeded. Larger dikes will actually divert runoff to another portion of
the site, usually to a downstream sediment trap or basin. Therefore, the designer should ensure
that they have the capacity for the 10-year storm event and that the channel created behind the
dike is properly stabilized to prevent erosion (Brown and Schueler 1997). In addition, the
downstream structure must be sized to handle the flow from the dike. Dikes should be designed
using standard hydrologic and hydraulic calculations and certified by a professional hydrologist
or engineer.  Diversion dikes should be installed  before the majority of the soil-disturbing
activity. As soon as the dike form is completed,  it should be machine compacted, fertilized,  and
either seeded and mulched or sodded. Excavated materials should be properly stockpiled for
future use or disposed of properly. Dikes should have an outlet that functions with a minimum of
erosion. Depending on site  conditions and outlet structures, the runoff directed by dikes might
need to be conveyed to a sediment-trapping device, such as a sediment basin or detention pond.
As grades increase over 4 percent, geotextile material or sod could be required to control erosion.
Slopes greater than 8 percent could require riprap. Dikes can be removed when the drainage area
and outlet are stabilized (NAHB, No Date). Dike design criteria must incorporate site-specific
conditions because dimensions depend  on expected flows, soil types, and climatic conditions. All
such inputs vary tremendously across sections of the country.

Effectiveness
No information has been found on the effectiveness of earth dikes used as BMPs, although
terraces often have sediment removal rates of up to 90 percent.

Limitations
An erosion-resistant lining in the channel might  be  needed to prevent erosion in the channel
caused by excessive grade.  In addition, the channel should be deepened and the grade realigned
if there is overtopping caused by sediment in the channel where the grade decreases or reverses.
If overtopping occurs at low points in the ridge where the diversion crosses the shallow draw, the
ridge should be reconstructed with a positive grade toward the outlet at all points. Finally, if
erosion occurs at the  outlet, an outlet stabilization structure should be installed; if sedimentation
occurs at the diversion outlet, a temporary sediment trap should be installed.

Maintenance
An earth dike should be inspected for signs of erosion after every major rain event. Any  repairs
or revegetation should be completed promptly (NAHB, No Date). The following actions can be
taken to properly maintain an earth dike:

       •  Remove debris and sediment from  the channel immediately after the storm event.

       •  Repair the dike to its original height.
                                           7-27

-------
                                                              Section 7: Technology Assessment
          Check outlets and make necessary repairs to prevent gully formation.

          Clean out sediment traps when they are 50 percent full.

          Once the work area has been stabilized, remove the diversion ridge, fill and compact
          the channel to blend with the surrounding area, and remove sediment traps, disposing
          of unstable sediment in a designated area.
Cost
The cost of an earth dike depends on the design and materials used. Small dikes can cost
approximately $2.00 per linear foot, while larger dikes can cost approximately $2.00 per cubic
yard. Earth dikes can cost approximately $4.50 per linear foot (NAHB, No Date).

An alternative means to estimate conceptual costs for earthen dikes is to use unit cost values and
a rough estimate of the quantities needed. Shallow trenching (1 to 4 feet deep) with a backhoe in
areas not requiring dewatering can be performed for $4 to $5 per cubic yard of removed material
(R.S. Means 2000). On the basis of that value, $2 per linear foot provides for 11 square feet of
flow area and $4.50 per linear foot provides for 24 square feet of flow area. That suggests that
the size of the dike is required before specifying a cost, which requires a  site-specific hydrologic
evaluation. On the basis of standards for Virginia, most small drainage areas (made up of 5 acres
or less) require 18-inch tall diversion dikes with a 4.5-foot base. Assuming the excavation
volume equals the volume of the dike, the resulting excavation volume is approximately 7 cubic
feet per linear foot, which (conservatively)  equates to $1.03 to $1.30 per  linear foot for
construction costs.

If the earthen dikes are to be permanent, additional costs are incurred to vegetate the dike. R.S.
Means (2000) estimates the cost of fine grading, soil treatment, and grassing is approximately $2
per square yard of earth surface area.  That adds approximately $6 per linear foot of dike. Where
gently  sloping areas need to be grassed only with acceptable species, the  cost can be as low as
$0.38 per square yard.

     7.2.2.2.   Temporary Swale

General Description
The term swale (grassed channel, dry swale, wet swale, biofilter) refers to a series of vegetated,
open-channel management practices designed specifically to treat and attenuate stormwater
runoff  for a specified water quality volume. As stormwater runoff flows through such channels,
it is treated by filtering through the vegetation in the channel, filtering through a subsoil matrix,
or infiltrating into the underlying soils. Variations of the grassed swale include the grassed
channel, dry swale, and wet swale. The  specific design features and methods of treatment differ
in each of these designs, but all are improvements on the traditional drainage ditch and
incorporate modified geometry and other features for use of the swale as  a treatment and
conveyance practice.

Applicability
Grassed swales can be applied in most situations with some restrictions and are very well suited
for treating highway or residential road  runoff because they are linear practices. Perimeter
                                          7-28

-------
                                                              Section 7: Technology Assessment
dikes/swales should be limited to a drainage area of no more than 1.97 acres (0.8 hectare) and
usually work best on gently sloping terrain. Perimeter dikes might not work well on moderate
slopes, and they should never be established on slopes exceeding 20 percent (UNEP  1994).

Regional Applicability. Grassed swales can be applied in most regions of the country. In arid
and semi-arid climates, however, the value of these practices needs to be weighed against the
water needed to irrigate them.

Ultra-Urban Areas. Ultra-urban areas are densely developed urban areas in which little
pervious surface exists. Grassed swales are generally not well suited to ultra-urban areas because
they require a relatively large area of pervious surface.

Stormwater Hot Spots. Stormwater hot spots are  areas where land use or activities generate
highly contaminated runoff, with concentrations of pollutants in excess of those commonly
found in Stormwater. A typical example is a gas station or convenience store. With the exception
of the dry swale design, hot spot runoff should not be directed toward grassed channels. Such
practices either infiltrate Stormwater or intersect the ground water, making use of the practices
for hot spot runoff a threat to ground water quality.

Stormwater Retrofit. A Stormwater retrofit is a Stormwater management practice (usually
structural), put into place after development has occurred, to improve water quality, protect
downstream channels, reduce flooding, or meet other specific objectives. One retrofit
opportunity using grassed swales modifies existing drainage ditches. Ditches have traditionally
been designed to convey Stormwater away from roads as quickly as possible. In some cases, it
might be possible to incorporate features to enhance pollutant removal or infiltration such as
check dams (for example, small dams along the ditch that trap sediment, slow runoff, and reduce
the longitudinal slope). Because grassed swales cannot treat a large area, using this practice to
retrofit an entire watershed would be expensive because of the number of practices needed to
manage runoff from a significant amount of the watershed's land area.

Cold Water (Trout) Streams. Grassed channels are a good treatment option in watersheds that
drain to cold water streams. Such practices do not retain water for a long period of time and often
induce infiltration. As a result, standing water will  not typically be subjected to warming by the
sun.

Design and Installation Criteria
Temporary swales should be designed using standard hydrologic and hydraulic calculations.
Designs should be certified by a professional hydrologist, engineer, or other appropriate
professional.

Perimeter dikes/swales should be established before any major soil-disturbing activity takes
place. Dikes should be compacted with construction equipment to the design height plus 10
percent to allow for settlement. If they are to remain in place for longer than 10 days, they should
be stabilized using vegetation, filter fabric, or other material. Diverted water should be directed
to a sediment trap or other sediment treatment area (UNEP 1994).
                                           7-29

-------
                                                              Section 7: Technology Assessment
In addition to the broad applicability concerns described above, designers need to consider
conditions at the site level. In addition, they need to incorporate design features to improve the
longevity and performance of the practice while minimizing the maintenance burden.

Siting Considerations
In addition to considering the restrictions and adaptations of grassed swales to different regions
and land uses, designers must ensure that this management practice is feasible at the site in
question. Depending on the design option, grassed channels can be highly restricted practices
because of site characteristics.

Drainage Area. Grassed swales generally should treat small drainage areas of less than 5 acres.
If the practices are used to treat larger areas, the flows and volumes through the swale become
too large to achieve stormwater treatment through infiltration and filtration.

Slope. Grassed swales should be used on sites with relatively flat slopes (less than 4 percent).
Runoff velocities in the channel become too high on steeper slopes. That can  cause erosion and
does not allow for infiltration or filtration in the swale.

Soils /Topography. Grassed swales can be used on most soils, with some restrictions on the
most impermeable soils. In the dry swale, a fabricated soil bed replaces on-site soils to ensure
that runoff is filtered as it travels through the soils of the swale.

Ground Water. The depth to ground water depends on the type of swale used. In the dry swale
and grassed channel options, designers should separate the bottom of the swale from the ground
water by at least 2 feet to prevent a moist swale bottom or contamination of ground water. In the
wet swale option, treatment is enhanced by a wet pool, which is maintained by intersecting the
water table.

Design Considerations
Although the grass swale has different design variations, including the grassed channel, dry
swale, and wet swale, some design considerations are common to all three.  One similarity is their
cross-sectional geometry. Swales should generally have a trapezoidal or parabolic cross-section
with relatively flat side slopes (flatter than 3:1). Designing the channel with flat side slopes
maximizes the wetted perimeter, which  is the length along the edge of the swale's cross-section
where runoff flowing through the swale is in contact with the vegetated sides and bottom of the
swale. Increasing the wetted perimeter slows runoff velocities and provides more contact with
vegetation to encourage filtering and infiltration. Another advantage to flat side slopes is that
runoff entering the grassed swale from the side receives some pretreatment along the side slope.
The flat bottom of all three should be between 2 and 8 feet wide. The minimum width ensures an
adequate filtering surface for water quality treatment, and the maximum width prevents braiding
(the formation of small channels in the swale bottom).

Another similarity among all three designs is the type of pretreatment needed. A small forebay
should be used at the inflow area of the  swale to trap incoming sediments. A pea gravel
diaphragm (a small trench filled with river run gravel) should be used to pretreat runoff entering
along the sides of the swale.
                                          7-30

-------
                                                               Section 7: Technology Assessment
Two other features designed to enhance the treatment ability of grassed swales are a flat
longitudinal slope (generally between 1 and 2 percent) and a dense vegetative cover in the
channel. The flat slope helps to reduce the velocity of flow in the channel. Dense vegetation also
helps reduce velocities, protect the channel from erosion, and act as a filter to treat stormwater
runoff. During construction, it is important to stabilize the channel before the turf has been
established, either with a temporary grass cover or with the use of natural or synthetic erosion
control products.

In addition to treating runoff for water quality, grassed swales need to convey larger storms
safely. Typical designs allow the runoff from the 2-year storm to flow through the swale without
causing erosion. Swales should also have the capacity to pass larger storms  (typically a 10-year
storm) safely.

The length of the swale necessary to infiltrate runoff can be calculated by using a mass balance
of runoff and infiltration for a triangular-shaped, cross-sectional  area.

Design Variations
The following discussion identifies three different variations of open channel practices, including
the grassed channel, the dry swale, and the wet swale.

Grassed Channel. (Discussed in more length in subsection 7.2.1.2, under Grass-lined Channels)
Of the three grassed swale designs, grassed channels are the most similar to a conventional
drainage ditch, with the major differences being flatter side slopes and longitudinal slopes and a
slower design velocity for water quality treatment of small storm events. Of all the grassed swale
options, grassed channels are the least expensive, but they also provide the least reliable pollutant
removal performance. The best application of a grassed channel  is as pretreatment to other
stormwater treatment practices.

One major difference between the grassed channel and most of the other structural practices is the
method used to size the practice. Most water quality practices for stormwater management are
sized by volume. This method sets the volume available in the practice equal to the water quality
volume, or the volume of water to be treated in the practice. The grassed channel, on the other
hand, is a flow rate-based design. On the basis of the peak flow from the water quality storm (this
varies from region to region, but a typical value is the 1-inch storm), the channel should be
designed so that runoff takes, on average, 10 minutes to flow from the top to the bottom of the
channel. A procedure for this design is in Design of Storm Water Filtering Systems (CWP 1996).

Dry Swales. Dry swales are similar in design to bioretention areas. Such practices incorporate a
fabricated soil bed into their design. The existing soil is replaced with a sand/soil mix that meets
minimum permeability requirements. An underdrain system is used under the soil bed. The
system is a gravel layer that encases a perforated pipe.  Stormwater treated in the soil bed flows
through the bottom into the underdrain, which conveys the treated stormwater to the storm drain
system. Dry swales are a relatively new design, but studies of swales with a native soil similar to
the man-made soil bed of dry swales suggest high pollutant removal rates.

Wet Swales. Wet swales intersect the ground water and behave similarly to a linear wetland cell.
Such a design variation incorporates a shallow, permanent pool and wetland vegetation to
                                           7-31

-------
                                                              Section 7: Technology Assessment
provide stormwater treatment. The design also has potentially high pollutant removal. One
disadvantage of the wet swale is that its use in residential or commercial settings is unpopular
because property owners sometimes view the shallow, standing water in the swale as a nuisance.

Regional Variations
Cold Climates. In cold or snowy climates, swales can serve a dual purpose by acting as both a
snow storage/treatment practice and a stormwater management practice. This dual purpose is
particularly relevant when swales are used to treat road runoff. If used for this purpose, swales
should incorporate salt-tolerant vegetation, such as creeping bentgrass.

Arid Climates. In arid or semi-arid climates, swales should be designed with drought-tolerant
vegetation, such as buffalo grass. As pointed out in the Applicability discussion, the value of
vegetated practices for water quality needs to be weighed against the cost of water needed to
maintain them in arid and semi-arid regions.

Effectiveness
Swales act to control peak discharges in two ways. First, the grass reduces runoff velocity,
depending on the length and slope of the swale.  Second, a portion of the stormwater runoff
volume passes through the swale and infiltrates into the soil. Table 7-7 summarizes grassed
swale pollutant removal efficiencies.
                  Table 7-7. Grassed swale pollutant-removal efficiency data
Study
Goldberg 1993
Seattle Metro and
Washington Department of
Ecology 1992
Seattle Metro and
Washington Department of
Ecology 1992
Wangetal. 1981
Dorman etal. 1989
Harper 1988
Kercher, Landon, and
Massarelli 1983
Harper 1988
Koon 1995
Yousefetal. 1985
Yousefetal. 1985
Welborn and Veenhuis 1987
Yu, Barnes, and Gerde 1993
Dorman etal. 1989
Pitt and McLean 1986
Oakland 1983
Dorman etal. 1989
Grassed swale removal efficiencies
TSS
67.8
60
83
80
98
87
99
81
67
-
-
0
68
65
0
33
-85
TP
4.5
45
29
-
18
83
99
17
39
8
-19.5
-25
60
41
-
-25
12
TN
-

-
-
-
84
99
40
-
13
8
-25
-
-
0
-
-
N03
31.4
-25
-25
-
45
80
99
52
9
11
2
-25
-
11
-
-
-100
Metals
42-62
2-16
46-73
70-80
37-81
88-90
99
37-69
-35 to 6
14-29
41-90
0
74
14-55
0
20-58
14-88
Bacteria
-100
-25
-25
-
-
-
-
-
-
-
-
-
-
-
0
0
-
Type
Grassed channel
Grassed channel
Grassed channel
Dry swale
Dry swale
Dry swale
Dry swale
Wet swale
Wet swale
Drainage channel
Drainage channel
Drainage channel
Drainage channel
Drainage channel
Drainage channel
Drainage channel
Drainage channel
                                           7-32

-------
                                                              Section 7: Technology Assessment
Limitations
Common problems associated with swales include excessive erosion along unlined channels
(usually because of excessive grade), erosion or sedimentation at the outlet point, or overtopping
of the swale at low points (UNEP 1994).

Additional limitations of the grass swale include the following:

       •  Grassed swales cannot treat a very large drainage area.

       •  Swales do not appear to be effective at reducing bacteria.

       •  Wet swales can become a nuisance because of mosquito breeding.

       •  If designed improperly (for example, improper slope), grassed channels will have
          very little pollutant removal.

       •  A thick vegetative cover is needed for the practices to function properly.

Maintenance
As with any BMP, swales must be maintained to continue to effectively remove pollutants.
Maintenance can include occasional mowing, fertilizing, and liming. In addition, any areas that
become damaged by erosion should be immediately repaired and replanted. The swales should
be protected from concentrated flows and be checked periodically for downstream obstructions.

Cost
To produce a conceptual cost approximation, grassed channel construction costs can be
developed using unit cost values.  Shallow trenching (1 to 4 feet deep) with a backhoe in areas
not requiring dewatering can be performed for $4 to $5 per cubic yard of removed material (R.S.
Means 2000). Assuming no disposal costs (i.e., excavated material is placed on either side of the
trench), only the cost of fine grading, soil treatment, and grassing (approximately $2 per square
yard) should be added to the trenching cost to approximate the total construction cost. Site-
specific hydrologic analysis of the construction  site is necessary to estimate the channel
conveyance requirement and the desired retention time in the swale. It is not unusual to have
flows on the order of 2 to 4 cfs per acre served.

For a design channel velocity of 1 foot per second, the resulting range in the channel cross-
section area can be as low as 2 but as high as 4 square feet  per acre drained. If the average
channel flow depth is 1 foot, the low estimate for grassed channel installation is $0.74 per square
foot of channel bottom per acre served per foot of channel length. The high estimate is $1.48 per
square foot of channel bottom per acre served per foot of channel length.

Table 7-8 summarizes additional costs of grass swales.
                                          7-33

-------
                                                              Section 7: Technology Assessment
         Table 7-8. Average annual operation and maintenance costs for a grass swale
Component
Mowing
General grass care
Debris/litter removal
Reseeding/
fertilization
Inspection and
general
administration
TOTAL
Estimated
unit cost ($)
0.89/1 00m2
8.8/1 00m2
0.51/m2
0.35/m2
0.74/m2

$ for swale size:
0.5m deep 0.3m
bottom width 3m
top width
145.0
162.98
93.0
5.9
231.0
638.0
$ for swale size:
1m deep 1m
bottom width 7m
top width
241.0
274.0
93.0
10.37
231.0
850.0
Comments
Mow 2-3 times per year
Grass maintenance area is
(top width + 3 m) x length

Area revegetated is 1 % of
maintenance area per year
Inspection once per year

Source: Ellis 1998.


      7.2.2.3.   Temporary Storm Drain Diversion

General Description
A temporary storm drain diversion is a pipe that reroutes an existing drainage system to
discharge flow into a sediment trap or basin. Such a practice reduces the amount of sediment-
laden runoff from construction sites that enters waterbodies without treatment. Temporary storm
drain diversions can be used when a permanent stormwater drainage system has not yet been
installed. It should be recognized that diversion channels can also be installed but are not
considered in the following discussion.

Applicability
A temporary storm drain diversion should be used to temporarily redirect discharge to a
permanent outfall and should remain in place until the area draining to the storm sewer is no
longer disturbed. Temporary storm drain diversions can also be combined with other structures
and used as a sediment-trapping device when the completion of a permanent outfall has been
delayed; alternatively, a sediment trap can be placed below a permanent outfall to remove
sediment before the final flow discharge.

Design and Installation Criteria
Because the diversion is only temporary, the layout of piping and the overall impact of the
diversion's installation on post-construction drainage patterns must be  considered. Once
construction is completed, the temporary diversion should be moved to restore the original
system. The following activities should be done at that time:

       •   The  storm drain  should be flushed before the sediment trap  is removed.

       •   The  outfall should be stabilized.

       •   Graded areas should  be restored.
                                          7-34

-------
                                                               Section 7: Technology Assessment
       •  State or local requirements should be checked for more detailed requirements and an
          appropriate professional should certify that the design meets local hydrologic and
          hydraulic requirements.

Effectiveness
If installed properly to capture the bulk of runoff from a construction site, temporary storm drain
diversions can be effective in reducing the discharge of sediment-laden, untreated water to
waterbodies. When used in combination with other ESC practices such as minimized clearing or
vegetative and chemical stabilization, the level of pollution from a construction site can be
substantially reduced or eliminated.

Limitations
Installing temporary storm drain diversion can result in the disturbance of existing storm
drainage patterns. Care must be taken to ensure that the original system is properly restored once
the temporary system is removed. The most common source of problems is excessive velocity at
the outlet. Installing an outlet stabilization structure is typically required and can be constructed
of riprap, reinforced concrete, geotextile linings, or a  combination.

Maintenance
Once installed, temporary storm drain diversions require very little maintenance. Frequent
inspection and maintenance of temporary storm drain systems, especially after large storms,
should ensure that pipe clogging does not occur and that runoff from the site is being
successfully diverted. After removing the temporary diversion, the permanent storm drain system
should be carefully inspected to ensure that drainage patterns have not been altered by the
temporary system.

Cost
Depending on the size of the construction site, a temporary storm drain diversion can be costly.
Costs include those associated with materials needed to construct the diversion and sediment trap
or basin (mainly piping, concrete, and gravel), and labor costs for installation and removal of the
system, all of which could involve excavation, re-grading, and inspections.  Because of the
variety of conditions that can affect storm drain diversion designs, typical costs per installation
are not presented here. However, site-specific cost estimates can be produced using unit cost
values along with site-specific quantity estimates. R.S. Means (2000) indicates a range of pipe
costs for surface placement, between $5.00 per linear foot for 4-inch diameter PVC piping, and
$9.20 per linear foot for 10-inch diameter PVC piping. On construction sites, temporary inlets
and outlets are usually formed by small rock-lined depressions. Assuming 4 cubic yards of
crushed rock (1.5-inch mean diameter) per opening, an inlet and outlet combine to add
approximately $200 per pipe installation, based on $25 per cubic yard of stone (R.S. Means
2000).

     7.2.2.4.   Pipe Slope Drain

General Description
Pipe slope drains are used to reduce the risk of erosion on slopes by discharging runoff to
stabilized areas. Consisting of a metal  or plastic flexible pipe if temporary, or pipes or paved
                                           7-35

-------
                                                              Section 7: Technology Assessment
chutes if permanent, the drains carry surface runoff from the top to the bottom of a slope that has
already been damaged by erosion or is at high risk for erosion. The drains are also used to drain
saturated slopes that have the potential for soil slides.

Applicability
Temporary slope drains can be used on most disturbed slopes to eliminate gully erosion
problems resulting from concentrated flows discharged at a diversion outlet.  Slope drains should
be used as a temporary measure for as long as the drainage area remains disturbed. They will
need to be moved once construction is complete and a permanent storm drainage system is
established. Appropriate restoration measures will then need to be taken, such as adjusting
grades and flushing sediment from the pipe before it is removed (UNEP 1994).

Design and Installation Criteria
Pipe slope drains can be placed directly on the ground or buried under the surface. The inlet
should be at the top of the slope and should be fitted with an apron, attached with a watertight
connection. Filter cloth should be placed under the inlet to prevent erosion. Flexible pipes, which
are positioned on top of the ground, should be securely anchored with grommets placed 10 feet
on center. The outlet at the bottom of the slope should also be stabilized with riprap. The riprap
should be placed along the bottom of a swale that leads to a sediment-trapping structure or
another stabilized structure.

Slope drain pipe sizes are based on drainage area and the size of the design storm. Pipes should
be connected to a diversion ridge at the top of the slope by covering it with compacted fill
material where it passes through the ridge. Discharge from a slope drain should be to a sediment
trap, sediment basin,  or other stabilized outlet (UNEP 1994).

Pipe slope drains should be installed perpendicular to the contour down the slope, and the design
should be able to handle the peak runoff for the 10-year storm. Recommendations of slope drain
diameter are summarized in Table 7-9 (NAFffi, No Date).

                  Table 7-9. Recommended pipe/tubing sizes for slope drains
Maximum drainage area
(acres)
0-0.5
0.5
0.75
1.0
1.5
2.5
3.5
5.0
Pipe/tubing diameter3
(inches)

12


18
21
24
30
Pipe/tubing diameter1*
(inches)

12


18

24

Pipe/tubing diameter0
(inches)

8
10
12
Individually designed



a UNEP 1994.
bUSDOT1995.
cIDNR 1992.
Recently graded slopes that do not have permanent drainage measures installed should have a
temporary slope drain and a temporary diversion installed. A temporary slope drain used in
                                           7-36

-------
                                                               Section 7: Technology Assessment
conjunction with a diversion conveys stormwater flows and reduces erosion until permanent
drainage structures are installed.

The following are design recommendations for temporary slope drains:

       •   The drain should consist of heavy-duty material manufactured for the purpose and
           have grommets for anchoring at a spacing of 10 feet or less.

       •   Minimum slope drain diameters should be observed for varying drainage areas.

       •   The entrance to the pipe should consist of a standard flare end section of corrugated
           metal. The corrugated metal pipe should have watertight joints at the ends. The rest of
           the pipe is typically corrugated plastic or flexible tubing, although for flatter, shorter
           slopes, a polyethylene-lined channel is sometimes used.

       •   The height of the diversion at the pipe should be the diameter of the pipe plus 0.5
           foot.

       •   The outlet should be placed at a reinforced or erosion-resistant location.

Temporary slope drains should be designed to adequately  convey runoff for a desired frequency
storm, typically either 2 years or 10 years depending on local regulations. Both the size and the
spacing can be determined on the basis of the contributing drainage area. Drains are spaced at
intervals corresponding to the specified drainage areas. For larger drainage areas and critical
locations, the drains should be sized on an individual basis (USDOT 1995).

Slope drains should be constructed in conjunction with diversion berms  such that the berms are
not overtopped. At the pipe inlet, the top of the berm should be a minimum of 300 mm (11.81
inches) higher than the top of the pipe. The entrance should be constructed of a standard flared
end section or a Tee section if designed properly. The entrance should be placed in a sump that is
depressed 150 mm (5.90 inches) (USDOT 1995).

The outlet of the slope drain must be protected with a riprap apron. If the slope drain is draining
a disturbed area and sufficient right-of-way is available, the drain could  empty into a sediment
trap (USDOT  1995). Table 7-10 summarizes slope drain characteristics.

                           Table 7-10. Slope drain characteristics
Capacity
Material
Inlet section
Connection to ridge at top of
slope
Outlet
2-year frequency, 24-hour-duration storm event
Strong, flexible pipe, such as heavy duty, nonperforated
, corrugated plastic
Standard Tor L flared-end section with metal toe plate
Compacted fill over pipe with minimum dimensions, 1 .5-foot depth, 4-foot top
width, and 6 inches higher than ridge
Pipe extends beyond toe of slope and discharges into a
basin unless contributing drainage area is stable
sediment trap or
Source: IDNR 1992.
Effectiveness
There is no information on the effectiveness of pipe slope drains.
                                           7-37

-------
                                                              Section 7: Technology Assessment
Limitations
The area drained by a temporary slope drain should not exceed 5 acres. Physical obstructions
substantially reduce the effectiveness of the drain. Overtopping of the inlet is a common slope
drain problem because of an undersized or blocked pipe or erosion at the outlet point from
insufficient protection (UNEP 1994). Other common failures caused by overtopping are from
inadequate pipe inlet capacity and reduced diversion channel capacity and ridge height, as well
as the following:

       •  Overtopping because the drainage area might be too large.

       •  Overtopping caused by improper grade of channel and ridge—A positive grade
          should be maintained.

       •  Overtopping caused by poor entrance conditions and trash buildup at the pipe inlet—
          Deepen and widen the channel at the pipe entrance and frequently inspect and clear
          the inlet.

       •  Washout—A washout along a pipe from seepage and piping can be caused by
          inadequate compaction, insufficient fill, or installation that might be too close to the
          edge of the slope.

       •  Erosion at outlet—The pipe should be extended to a stable grade or an outlet
          stabilization structure is needed.

       •  Displacement or separation of pipe—The pipe should be tied down and the joints
          secured.

Maintenance
Pipe slope drains must be inspected after each significant runoff event for evidence of erosion
and uncontrolled runoff. Any repairs to the drain should be made immediately. Significant
amounts of sediment trapped at the outfall should also be removed in a timely manner and
disposed of properly (NAHB, No Date).

The following actions should be taken to properly maintain a pipe slope drain (IDNR 1992):

       •  Inspect slope drains and supporting diversions once a week and after every storm
          event.

       •  Check the inlet for sediment or trash accumulation;  clear and restore to proper
          entrance condition.

       •  Check the fill over the pipe for settlement, cracking, or piping holes; repair
          immediately.

       •  Check for holes where the pipe emerges from the dike; repair immediately.

       •  Check the conduit for evidence of leaks or inadequate anchoring; repair immediately.
       •  Check the outlet for erosion or sedimentation; clean and repair, or extend if
          necessary.

       •  Once slopes have been stabilized, remove the temporary diversions and slope drains,
          and stabilize all disturbed areas.
                                          7-38

-------
                                                             Section 7: Technology Assessment
Cost
The cost of pipe slope drains and their installation varies with the design and materials used.
Site-specific cost estimates can be produced using unit cost values with site-specific quantity
estimates.

R.S. Means (2000) indicates a range of pipe costs for surface placement between $5.00 per linear
foot for 4-inch diameter PVC piping, and $9.20 per linear foot for 10-inch diameter PVC piping.
On construction sites, temporary inlets and outlets are usually formed by small, rock-lined
depressions. Assuming 4 cubic yards of crushed rock (1.5-inch mean diameter) per opening, an
inlet and outlet together add approximately $200 per pipe installation, based on $25 per cubic
yard of stone (R.S. Means 2000).

      7.2.2.5.   Check Dam

General Description
A check dam is a small, temporary barrier constructed across a drainage channel or swale to
reduce the velocity of the flow. By reducing the  flow velocity, the erosion potential is reduced,
detention times are lengthened, and more sediment is able to settle out of the water column.
Check dams can be constructed of stone, gabions, treated lumber, or logs (NAHB, No Date).
Recent work by Dr. Richard McLaughlin involves the use of fiber check dams installed at grade
with polyacrylamide (PAM) applied to the check dam (McLaughlin 2009) and PAM blocks on
the downhill side of triangular silt dikes (McLaughlin, No Date b)for passive dosing to greatly
reduce turbidity.

Check dams are inexpensive and easy to install.  They can be used permanently to settle
sediment, reduce the velocity of runoff, and provide aeration. Check dams are often used in
combination with other practices, such as sediment traps or basins.

Applicability
Check dams are commonly used (1) in channels  that are degrading but where permanent
stabilization is impractical because of their short period of usefulness and (2) in eroding channels
where construction delays or weather conditions prevent timely installation of erosion-resistant
linings (IDNR 1992).

Check dams are also useful in steeply sloped swales, in small channels, in swales where adequate
vegetative protection cannot be established, or in swales or  channels that will be used for a short
time and it is not practical to line the channel or  implement  other flow control practices (USEPA
1993). In addition, check dams are appropriate where temporary seeding has been recently
implemented but has not had time to fully develop and take root. The contributing drainage area
should range from 2 to 10 acres. Check dams should be used only in small, open channels that
will not be overtopped by flow once the dams are built. They should not be built in stream
channels, either intermittent or perennial (UNEP 1994). Check dams can be effective sediment
trapping devices when designed appropriately.
                                          7-39

-------
                                                             Section 7: Technology Assessment
Design and Installation Criteria
Check dams can be constructed from a number of different materials. Most commonly, they are
made of rock, logs, sandbags, or straw bales. Rock or stone is often preferred because of its cost-
effectiveness and longevity. Logs and straw bales will decay with time and are not recommended
because they can cause waterway blockage if they fail. When using rock or stone, the material
diameter should be 2 to 15 inches. The stones should be extended 18 inches beyond the banks,
and the side slopes should be 2:1 or flatter. Lining the upstream side of the dam with a foot of 1-
to 2-inch gravel can improve the efficiency of the dam (NAHB, No Date). Logs should have a
diameter of 6 to 8 inches. Regardless of the material used, careful construction of a check dam is
necessary to ensure its effectiveness.

The distance between rock check dams will vary depending on the slope of the ditch, with closer
spacing when the slope is steeper. The size of stone used in the check dam should also vary with
the expected design velocity and discharge. As velocity and discharge increase, the rock size
should also increase. For most rock check dams, 3 inches to 12 inches is a suitable stone size. To
improve the sediment-trapping efficiency of check dams, a filter stone can be applied to the
upstream face. A well-graded,  coarse aggregate that is less than 1 inch in size can be used as a
filter stone.

All check dams should have a  maximum height of 3 feet. The center of the dam should be at
least 6 inches lower than the edges. Such a design creates a weir effect that helps to channel
flows away from the banks and prevent further erosion. Additional stability can be achieved by
implanting the dam material approximately 6 inches into the sides and bottom of the channel
(VDCR 1995).

When installing more than one check dam in a channel, outlet stabilization measures should be
installed below the final dam in the series. Because that area is likely to be vulnerable to further
erosion, riprap or some other stabilization measure is highly recommended.

Effectiveness
Field  experience has shown that rock check dams are more effective than silt fences or straw
bales to stabilize wet-weather ditches (VDCR 1995). Straw bales have been shown to have very
low trapping efficiencies and should not be used for check dams.  For long channels, check dams
are most effective when used in a series, creating multiple barriers to sediment-laden runoff.

Dr. Richard McLaughlin reports dramatic turbidity reductions using fiber check dams with PAM,
to levels below 200 nephelometric turbidity units (NTUs) and in some cases to below 50 NTUs.
McLaughlin also reports reductions when PAM is added to conventional rock check dams
(McLaughlin, No Date a). McLaughlin (2009) states that fiber check dams are much more
effective than rock check dams, according to data presented in the study. For summaries of
studies with monitoring data, and annotated bibliographies for the journal articles and
professional conference proceedings that EPA reviewed, see DCN 43114.

Limitations
Check dams should not be used in perennial streams unless approved by an appropriate
regulatory agency (USEPA 1992; VDCR 1995). Because the primary function of check dams is
to slow runoff in a channel, they should not be used as a standalone substitute for other sediment-
                                          7-40

-------
                                                             Section 7: Technology Assessment
trapping devices. Also, leaves have been shown to be a significant problem because they clog
check dams; therefore, increased inspection and maintenance might be necessary in the fall.
Common problems with check dams include channel bypass and severe erosion when
overtopped and ineffectiveness from accumulated sediment and debris. When designing check
dams, because they reduce the capacity of a channel to transmit stormwater runoff and, thus,
need to be sized appropriately should be taken into account (UNEP 1994). The check dam could
also kill grass linings in the channel if the water level remains high after it rains or if there is
significant sedimentation. In addition, a check dam might reduce the hydraulic capacity of the
channel and create turbulence, which erodes the channel banks (NAHB, No Date).

Maintenance
Check dams should be inspected periodically to ensure that they have not been repositioned as a
result of stormwater flow. In addition, the center of a check dam should always be lower than its
edges. Additional stone might have to be added to maintain the correct height.  Sediment  should
not be allowed to accumulate to more than half the original dam height. Any required
maintenance should be performed immediately. When check dams are removed, take care to
remove all dam materials to ensure proper flow within the channel. The channel should
subsequently be seeded for stabilization (NAHB, No Date).

Cost
The cost of check dams varies according to the material used for construction and the width of
the channel to be dammed. In general, it is estimated that check dams constructed of rock cost
about $100 per dam (USEPA 1992). Brown and Schueler (1997) estimate that a rock check dam
would cost approximately $62 per installation, including the cost for filter fabric bedding. Other
materials, such as logs and sandbags, might be a less expensive alternative, but they could
require higher maintenance costs. McLaughlin estimates that fiber check dams are comparable in
cost to stone check dams, however installation costs can be much lower because fiber check
dams can be positioned and staked in place by hand (stone check dams usually require a backhoe
or other equipment to install) (McLaughlin, personal communication). Fiber check dams will
likely require periodic replacement for longer-duration projects. Costs for PAM addition to
check dams is minimal, with a new application required every few storm events. Application is
done by hand, and a predetermined quantity of dry PAM is simply applied to the surface of the
check dam. McLaughlin (2009) reports costs for installation and maintenance for standard BMPs
(stone check dams with preceding excavations) and fiber check dams with and without PAM at
two linear road projects in North Carolina. The installation cost per linear meter was $6.50  (site
1) and $5.74 (site 2) for standard BMPs, $5.59 (site 1) for fiber check dams only, and $4.33 (site
1) and $5.35 (site 2) for fiber check dams with PAM. The cost per maintenance action  was  $416
at each site for the standard BMPs and $74 to $79 for fiber check dams with PAM.

      7.2.2.6.   Lined Waterways

General Description
Lined channels convey stormwater runoff through a stable conduit. Vegetation lining the channel
reduces the flow velocity of concentrated runoff. Lined channels usually are not designed to
control peak runoff loads by themselves and are often used in combination with other BMPs
such as subsurface drains and riprap stabilization. Where moderately steep slopes require
                                          7-41

-------
                                                               Section 7: Technology Assessment
drainage, lined channels can include excavated depressions or check dams to enhance runoff
storage, decrease flow rates, and enhance pollutant removal. Peak discharges can be reduced
through temporary detention in the channel. Pollutants can be removed from stormwater by
filtration through vegetation, by deposition, or in some cases by infiltration of soluble nutrients
into the soil. The degree of pollutant removal in a channel depends on the residence time of the
water in the channel and the amount of contact with vegetation and the soil surface, but pollutant
removal is not generally the major design criterion.

Often construction increases the velocity and volume of runoff, which causes erosion in newly
constructed or existing urban runoff conveyance channels. If the runoff during or after
construction would cause erosion in a channel,  the channel should be lined or flow control
practices instituted. The first choice of lining should be grass or sod because that reduces runoff
velocity and provides water quality benefits through filtration and infiltration. If the velocity in
the channel would erode the grass or  sod, one can use riprap, concrete, or gabions (USEPA
2000). Geotextile materials can be used in conjunction with  either grass or riprap linings to
provide additional protection at the soil-lining interface.

Applicability
Lined channels typically are used in residential developments, along highway medians, or as an
alternative to curb and gutter systems. Grass-lined channels  should be used to convey runoff only
where slopes are 5 percent or less. Such channels require periodic mowing, occasional spot-
seeding, and weed control to ensure adequate grass cover (UNEP 1994).

Lined channels should be used in areas where erosion-resistant conveyances are needed, such  as
in areas with highly erodible soils and slopes of less than 5 percent.  They should be installed
only where space is available for a relatively large cross-section. Grassed channels have a limited
ability to control runoff from large storms and should be used with the recommended allowable
velocities for the specific soil types and vegetative cover.

Design and Installation Criteria
The design of a lined waterway requires proper determination of the channel dimensions. It must
ensure that (1) the velocity of the flowing water will not wash out the waterway and that (2) the
capacity of the waterway is sufficient to carry the surface flow from the watershed without
overtopping.

Vegetation-Lined Channels. Grass-lined channels have been previously discussed in detail and
are only summarized in this section. The allowable velocity  of water in the waterway depends on
the type, condition, and density of the vegetation, as well as  the erosive characteristics of the soil.
Uniformity of vegetative cover is important because the stability of the most sparsely covered
area determines the stability of the channel. Grasses are a better vegetative cover than legumes
because grasses resist water velocity more effectively.

Vegetative-lined channels can have triangular, parabolic, or  trapezoidal cross-sections. Side
slopes should not exceed 3:1 to facilitate the establishment, maintenance, and mowing of
vegetation. A dense cover of hardy, erosion-resistant grass should be established as soon as
possible following grading. This could require using straw mulch and installing protective
netting until the grass becomes established. If the intent is to create opportunities for runoff to
                                           7-42

-------
                                                              Section 7: Technology Assessment
infiltrate into the soil, the channel gradient should be kept near zero, the channel bottom must be
well above the seasonal water table, and the underlying soils should be relatively permeable
(generally, with an infiltration rate greater than 2 centimeters [0.78 inches] per hour).

Rock-Lined Channels. Riprap-lined channels can be installed on somewhat steeper slopes than
grass-lined channels. They require a foundation of filter fabric or gravel under the riprap.
Generally, side slopes should not exceed 2:1, and riprap thickness should be 1.5 times the
maximum stone diameter. Riprap should form a dense, uniform, well-graded mass (UNEP 1994).

Lined channels should be sited in accordance with the natural drainage system and should not
cross ridges. The channel design should not have sharp curves or significant changes in slope.
Channels should not receive direct sedimentation from disturbed areas and should be established
only on the perimeter of a construction  site to convey relatively clean stormwater runoff. They
should also be separated from disturbed areas by a vegetated buffer or other BMP to reduce
sediment loads.

Basic design recommendations for lined channels include the following:

       •  Construction and vegetation of the channel should occur before grading and paving
          activities begin.

       •  Design velocities should be less than 5 feet per second.

       •  Geotextiles can be used to stabilize vegetation until it is fully established.

       •  Covering the bare soil with sod or geotextiles  can provide reinforced stormwater
          conveyance immediately.

       •  Triangular-shaped channels  should be used with low velocities and small quantities of
          runoff; parabolic grass channels are used for larger flows and where space is
          available; trapezoidal channels are used with large flows of low velocity (low slope).

       •  Outlet stabilization structures might be  needed if the runoff volume or velocity has
          the potential to exceed the capacity of the receiving area.

       •  Channels should be designed to convey runoff from a 10-year storm without erosion.

       •  The sides of the  channel should be sloped less than 3:1, with V-shaped channels
          along roads sloped 6:1 or less for safety.

       •  All trees, bushes, stumps, and other debris should be removed during construction.

Effectiveness
Lined channels can effectively transport stormwater from construction areas if they are designed
for expected flow volumes and velocities and if they do not receive sediment directly from
disturbed areas.

Limitations
Lined channels, if improperly installed, can alter the natural flow of surface water and have
adverse effects on downstream waters. Additionally, if the design capacity is exceeded by a large
storm event, the vegetation  might not be sufficient to prevent erosion, and the channel might be
                                           7-43

-------
                                                              Section 7: Technology Assessment
destroyed. Clogging with sediment and debris reduces the effectiveness of grass-lined channels
for storm water conveyance.

Common problems in lined channels include channel erosion before vegetation is fully
established and gullying or head cutting in the channel if the grade is too steep. In addition, trees
and brush tend to invade lined channels, causing maintenance problems.

Riprap-lined channels can be designed to safely convey greater runoff volumes on steeper slopes.
However, they should generally be avoided on slopes exceeding 10 percent because stone
displacement, erosion of the foundation, or channel overflow and erosion resulting from a
channel that is too small can occur. Thus, channels established on slopes greater than 10 percent
will usually require protection with rock gabions, concrete, or other highly stable and protective
surfaces (UNEP  1994).

Maintenance
Maintenance requirements for lined channels are relatively minimal. During the vegetation
establishment period, the channels should be inspected after every rainfall. Other maintenance
activities that should be carried out after vegetation is established are mowing, litter removal, and
spot vegetation repair. The most important objective in the maintenance of lined channels is
maintaining a dense and vigorous growth of turf. Periodic cleaning of vegetation and soil buildup
in curb cuts is required so that water flow into the channel is unobstructed. During the growing
season,  channel grass should be cut no shorter than the level of design flow, and the cuttings
should be removed promptly.

Cost
Costs of grassed channels range according to depth, with a 1.5-foot-deep, 10-foot-wide grassed
channel estimated at $6,395 to $17,075 per trench, while a 3-foot-deep, 21-foot-wide grassed
channel is estimated at $12,909 to $33,404 per trench (SWRPC 1991).

EPA also refers readers to the discussion of costs for grass-lined channels, which contains many of
the design and cost elements required for installing lined waterways. Designers have a range of
options for lining new channels. Geosynthetic TRMs can be used for immediate erosion protection
in channels exposed to runoff flows. The Erosion  Control Technology Council (a geotextile
industry support association) suggests that TRMs  cost approximately $7.00 per square yard
(installed) for channel protection (Lancaster et al.  2002). R.S. Means indicates that machine-placed
riprap costs of approximately $40 per cubic yard.  The riprap maximum size is typically between 6
and 12 inches, depending  on the channel design velocity.  A cubic yard of riprap will cover
between 36 and 18 square feet of channel bed for those riprap sizes (assuming depth of riprap is
1.5 times the maximum size). Such estimates suggest that riprap lining will be between $10 and
$20 per  square foot of channel (costs include  materials, labor, and equipment, with overhead and
profit).

   7.2.3.   SEDIMENT-TRAPPING DEVICES

The devices listed under this group of BMPs trap sediment primarily through impounding water
and allowing for settling to occur (Haan et al.  1994). Silt fence, super silt fence,  straw bale dikes,
sediment traps, and sediment basins all  control flow through a porous flow control system such as
                                          7-44

-------
                                                              Section 7: Technology Assessment
filter fabric or straw bales, or they use a dam to impound water with a pipe, open channel, or rock
fill outlet. The filtering capacity of silt fence (filter fabric) contributes only a small amount of
trapping, but it serves to make the fence less porous and hence increases ponding. For steady-state
flows, the trapping that occurs behind the flow-control device can be shown to be directly
proportional to the surface area and indirectly proportional to flow through the system (Haan et
al. 1994). The ratio of the surface area to flow is known as the overflow rate, and trapping in
such systems is predicted by the ratio of overflow rate to particle settling velocity. Although flows
in nature are inherently non-steady-state and more complex than steady-state systems, studies have
shown that the best predictor of trapping in such systems is still the ratio of settling velocity to
overflow rate (Hayes et al.  1984). In the case of non-steady-state, the overflow rate is best defined
by the ratio of peak discharge to surface area (Hayes et al.  1984; McBurnie et al. 1990).

The amount of trapping in these structures depends on the size of the structure, flow rates into
the system, hydraulics of the flow control system, the size distribution of the sediment flowing
into the structure, and the chemistry of the sediment-water system (Haan et al. 1994). Trapping
can be enhanced by chemical treatment of flows into the structure, but the effects have not been
widely defined for varying mineralogy and chemistry of the sediment-water system (Haan et al.
1994; Tapp and Barfield 1986). Recent studies have been conducted on applying PAM to
disturbed areas for enhancing settling (Benik et al. 1998; Masters et al. 2000; Roa-Espinosa et al.
2000), but results have not been definitive.  No known studies have evaluated the effects of PAM
application to disturbed areas on settling in sediment trapping devices.

Sediment flowing into sediment trapping devices is composed of primary particles and
aggregated particles. Aggregates are formed when clays, silts, and sands are cemented together
to form  larger particles that have settling velocities far greater than those of any individual
particles alone, although the  degree of aggregation depends on the amount of cementing material
present  (typically clays and organic matter). Because the aggregates have higher settling
velocities than primary particles, the degree of aggregation that is present has a large effect on
the trapping that occurs. Procedures are available to measure the combined size distribution of
aggregate and primary particle size distribution (Barfield  et al. 1979; Haan et al. 1994).
Procedures are also available to predict particle size distributions of aggregates and primary
particles (Foster et al. 1985).

In the absence of chemical treatment, the sediment that can be captured in sediment trapping
devices is typically the larger settleable solids. In many cases, to trap the smaller-sized clay
particles, structures with surface areas larger than the construction site itself would have to be
built (Barfield 2000). Chemical treatment can be used to reduce the size captured, but it has not
been widely adopted because of the cost and complexity of the operation (Tapp et al. 1981).

Sediment-trapping devices also provide some stormwater detention by virtue of detaining flows
long enough to allow sediment to settle out and be deposited. However, to operate as a
storm water-detention structure, the design should include adequate volume for detention.

Virtually all the available information on sediment-trapping structures, both theoretical and
experimental, is on effects on receiving waters and not downstream effects. In a very limited
analysis, Barfield (2000) combines the SEDIMOT II computer model with the FLUVIAL model
                                           7-45

-------
                                                             Section 7: Technology Assessment
to theoretically evaluate the effect of sediment trapping structures on downstream
geomorphology in a Puerto Rican watershed.

      7.2.3.1.   Silt Fence and Compost Filter Berms/Socks

General Description
Silt fences are used as temporary sediment barriers consisting of filter fabric anchored across and
supported by posts. Their purpose is to retain sediment from small, disturbed areas by reducing
the velocity of sediment-laden runoff and promoting sediment deposition (Smolen et al. 1988).
Silt fences capture sediment by ponding water and allowing for deposition, not by filtration. Silt
fence fabric first screens silt and sand from runoff, resulting in clogging of the lower part of the
fence. The pooling water allows sediments to settle out of the runoff. Silt fences work best in
conjunction with temporary basins, traps, or diversions. Compost filter berms and socks can also
be used in lieu of silt fences. A compost filter berm is a dike of compost or a compost product
that is placed perpendicular to sheet flow runoff to control erosion in disturbed areas and retain
sediment. A compost filter sock is a mesh or geotextile tube filled with composted material.
Compost filter berms are commonly used as perimeter controls, and are sometimes used in
combination with silt fence to provide redundant control of perimeter ditch.

Applicability
Silt fences are generally placed at the toe of fills, along the edge of waterways, and along the site
perimeter. The fences should not be used in drainage areas with concentrated and high flows, in
large drainage areas, or in ditches and swales where concentrated flow is present.

The drainage area for the fence should be selected on the basis of design storms and local
hydrologic conditions so that the silt fence is not expected to overtop. A typical design calls for
no greater than one-quarter acre of drainage area per 100 feet offence, but that is highly variable,
depending on climate. The fence should be stable enough to withstand runoff from a 10-year
peak storm. Table 7-11  lists the maximum slope length specified by the U.S. Department of
Transportation (USDOT). The slope lengths should be based on sediment load and flow rates.
That would mean that the values given below should be adjusted for climatic conditions instead
of one size fits all to ensure maximum effectiveness.

                      Table 7-11. Maximum slope lengths for silt fences
Slope
(%)
<2
5
10
20
25
30
35
40
45
50
18-inch (460 mm) fence
250 ft (75 m)
100 ft (30m)
50 ft (15m)
25 ft (8 m)
20 m (6 ft)
15 ft (5m)
15 ft (5m)
15 ft (5m)
10 ft (3m)
10 ft (3m)
30-inch (760 mm) fence
500 ft (150m)
250 ft (75 m)
150 ft (45m)
70 ft (21 m)
55 ft (17m)
45 ft (14m)
40 ft (12m)
35 ft (10m)
30 ft (9 m)
25 ft (8 m)
               Source: USDOT 1995.
                                          7-46

-------
                                                               Section 7: Technology Assessment
Typical standards and specifications call for the silt fence to be on fairly level ground and follow
the land contour, although it is recognized that a slight slope can occur along the fence in spite of
the best installation practices. Runoff can move down the contour until a weak spot occurs in the
buried toe and undercuts the fence. Alternatively, flow could move to a low spot where it
accumulates and causes an overtopping. In either case, trapping by the silt fence is essentially
zero, and flows will then have been concentrated, causing downslope erosion.

Design and Installation Criteria
Design criteria are of two types:

       •  Hydrologic design for a required trapping of sediment and flow rate to pass the design
          storm

       •  Selecting appropriate installation criteria such that the silt fence performs as designed

Hydrologic Design
The fence should be designed to pass the design storm without causing damage, while trapping
the required amount of sediment. It is necessary to use either a database or some type of model to
develop the appropriate hydrologic design. Efforts to model the sediment trapping that occurs
through with a silt fence have resulted in models that predict the settling in the ponded area
upstream from the fence (Barfield et al. 1996; Lindley et al. 1998). The results from model
simulations show that trapping depends primarily on the surface area of the impounded water
and the flow rate through the filter. The models use a clear water flow rate, typically  specified by
the manufacturer, to predict discharge. However, numerous studies have shown that sediment-
laden flows cause clogging of the geotextiles used to construct the fence, depending on the
opening size and size of the sediment (Britton et al. 2001; Wyant 1980; Barrett et al.  1995;
Fisher and Jarrett 1984). Thus, results from model studies to date are suspect and need to be
modified to account for the effects of clogging on flow rate. Barfield et al. (2001) developed a
model of flow rate using conditional probability concepts, but the results have not been
experimentally verified.

Design aids have been developed for silt fence, using simulations from the SEDEVIOT III model
(Hayes and Barfield 1995). In the model, predictions  are made about trapping efficiency using the
ratio of settling velocity for the dis of the eroded sediment, divided by the ratio of discharge to
ponded surface area.3 The design aids yield conservative estimates as  compared to the  SEDEVIOT
III model, but the database used for generating the design aid is based on the assumption that
clogging does not affect flow rates. The discussion above shows that assumption to be  erroneous.

SEDCAD takes the approach of using a slurry flow rate, not a clean water flow rate, when it
simulates fence effectiveness, reporting slurry rates ranging between 0.1 and 15 gallons per
minute (gpm) per sqare foot. On the basis of this discussion, one can conclude that it is difficult
to predict with accuracy the trapping efficiency of silt fence under a given set of conditions. In
addition, the quality of installation and maintenance are important to the long-term performance
3 d15:15 percent by weight of suspended solids are smaller than those that are trapped by this device; similarly dso
indicates that 50 percent by weight of suspended solids are smaller than those trapped.
                                           7-47

-------
                                                               Section 7: Technology Assessment
of the fence. The best available estimate of sediment trapping obtained from modeling of
hydrologic events should be applied with care in any site design problem.

Installation Criteria
General installation criteria for the silt fence should incorporate the following factors:

       •   The fabric must have sufficient strength to counter forces created by contained water
           and sediment (Sprague 1999).

       •   The posts must have sufficient strength to counter the forces transferred to them by
           the fabric (Sprague 1999).

       •   The fabric must be installed to ensure that the loads are all adequately transferred
           through the fabric to the posts or the ground without overstressing (Sprague 1999).

       •   The fence must be designed according  to site-specific hydrologic and soil conditions
           such that it will not overtop during design events.

       •   The fence must be installed (anchored) with a buried toe of sufficient depth so that it
           does not become detached from the soil surface.

       •   In general, the fence requires a metal wire backing to provide sufficient strength to
           prevent failure from the weight of trapped sediment and to prevent the toe of the
           fabric from being removed from the ground.

       •   Maximum drainage area behind the fence should be determined on the basis of the
           local rainfall and the infiltration characteristics of the soil and cover.

Silt fence material is typically synthetic filter fabric or a pervious sheet of polypropylene, nylon,
polyester, or polyethylene yarn. The fabric should have ultraviolet ray inhibitors and stabilizers
to provide for a minimum useful construction life  of 6  months or the duration of construction,
whichever is greater. The height of the fence fabric should not exceed 3 feet. If standard strength
filter fabric is used, it should be reinforced with a  wire fence, extending down into the trench  that
buries the toe. The wire should be of sufficient strength to support the weight of the deposited
sediment and water. In general, a minimum 14 gauge and a maximum mesh spacing of 6 inches
is called for (Smolen et al. 1988). Typical requirements for the  silt fence physical properties, as
specified in selected local BMP standards and specifications, are presented in Table 7-12.

                      Table 7-12. Typical requirements for silt fence fabric
Physical property
Filtering Efficiency
Tensile Strength
at 20% (maximum)
Elongation
Slurry Flow Rate
Water Flow Rate
UV Resistance
Requirements
Woven fabric
85%
Standard Strength — 30 pound/linear inch
Extra Strength — 50 pound/linear inch
0.3 gallon/square feet/minute
15 gallons/square feet/minute
70%
Non-woven fabric
85%
Standard Strength — 50 pound/linear inch
Extra Strength — 70 pound/linear inch
4.5 gallons/square feet/minute
220 gallon/square feet/minute
85%
Source: NCDNR 1988; IDNR 1992.
                                           7-48

-------
                                                              Section 7: Technology Assessment
Note that those numbers, particularly the flow rates, can vary widely depending on the local soil
condition because of possible clogging of the filter material.

Material for the posts used to anchor the filter fabric can be constructed of either wood or steel.
Wooden stakes should be buried at a depth sufficient to keep the fence, when loaded with
sediment and water, from falling over. The depth of burial should depend on post diameter and
soil strength characteristics when saturated. Many standards and specifications set a minimum
post length of 5 feet with 4-inch diameter for posts composed of softwood (e.g., pine) and 2-inch
diameter for posts composed of hardwood (e.g., oak) (Smolen et al.  1988). Steel posts should
also be designed according to local wet soil strength characteristics.  Some standards and
specifications for the posts set a minimum weight of 1.33 pounds per linear feet with a minimum
length of 4 feet. Steel posts should also have projections to adhere filter fabric to the post
(Smolen etal. 1988).

A silt fence should be erected continuously from a single roll of fabric to eliminate unwanted
gaps in the fence. If a continuous roll of fabric is not available, the fabric should overlap from
both directions only at posts with a minimum overlap of 6 inches and be rolled together with a
special flexible rod to keep the ends from separating. Fence posts should be spaced at a distance
on the basis of wet soil strength characteristics and post size and strength; generally, the posts are
spaced approximately 4 to 6 feet apart. If standard strength fabric is  used in combination with
wire mesh, the spacing can be larger. Typically, standards and specifications call for the posts to
be no more than  10 feet apart. If extra-strength fabric is used without wire mesh reinforcement,
some standards call for the support posts to be spaced no more than  6 feet apart (VDCR 1995).
Again, the spacing depends on wet soil strength characteristics and post size.

A silt fence must provide sufficient storage capacity or be stabilized over flow outlets such that
the storage volume of water will not overtop the fence. The return period event (size of the
rainfall event managed) used for design is typically a prerogative of the regulatory agency. For
temporary fences, a 2-year storm event is typically used as a design  standard. Fences that will be
in place for 6 months or longer are commonly designed for a 10-year storm event (Sprague
1999). The space behind the fence used for impoundment volume must be sufficient to
adequately contain the sediment that will be deposited. Each storm will deposit sediment behind
the fence, and after a time, the amount of sediment accumulated will render the fence useless.
Frequency offence management is a function of its sizing (i.e.,  whether the fence was installed
for a 2-year or a  10-year storm event) (Sprague 1999) and the amount of erosion that occurs in
the area draining to the fence.

Effectiveness
The performance of silt fences has not been well defined. Laboratory studies using carefully
controlled conditions have shown trapping efficiencies in the range of 40 to 100 percent,
depending on the type of fabric, overflow rate, and detention time (Barrett et al. 1995; Wyant
1980; Wishowski et al. 1998). Field studies have been limited and quite inadequate; however, the
results show that field-trapping efficiencies are very low. In fact, Barrett et al. (1995) obtained a
value of zero percent trapping averaged over several samples with a standard error of 26 percent.
Barrett et al. (1995) cite the following reasons for the field tests not  showing the expected
results:
                                           7-49

-------
                                                              Section 7: Technology Assessment
       •  Inadequate fabric splices

       •  Sustained failure to correct fence damage resulting from overtopping

       •  Large holes in the fabric

       •  Under-runs because of inadequate toe-ins

       •  Silt fence damaged and partially covered by the temporary placement of stockpiles of
          materials

Silt fences are effective at removing large particle sediment, primarily aggregates, sands, and
larger silts. Sediment is removed through impounding of water to slow velocity. It is argued that
the silt fence will not contribute to a reduction in small particle sediment and is not effective
against other pollutants (WYDEQ 1999). EPA (1993) reports the following effectiveness ranges
for silt fences constructed of filter fabric: average TSS removal of 70 percent, sand removal of
80 to 90 percent, silt-loam removal of 50 to 80 percent, and silt-clay-loam removal of 0 to 20
percent. However, EPA numbers from the Nationwide Urban Runoff Program should not be
considered to apply to every location. The actual trapping will vary widely for a given design
because of differences in hydrologic regimes and soil types.

The advantages of using silt fences include minimal labor requirement for installation, low cost,
high efficiency in removing sediment, durability, and sometimes reuse (Sprague 1999). Silt
fences are the most readily available and cost-effective control options where options such as
diversion are not possible. Silt fences are also a popular choice because contractors have used
them extensively and their familiarity makes silt fence use more likely for future construction
activities. The visibility of a silt fence is also an advantage (i.e., the fence is advertising the use
of ESC practices). In addition, the silt fence visibility makes site inspection easier for contractors
and government inspectors (CWP 1996).

EPA's National Menu of Best Management Practices for Stormwater Phase II reports that
compost filter socks and berms are at least as effective as other traditional ESC BMPs in
controlling sediment; however, the results of the studies vary depending on the site conditions
(USEPA 2008).

Limitations
Silt fences should not be installed along areas where rocks or other hard surfaces prevent
uniform anchoring offence posts and entrenching of the filter fabric. An insufficient anchor
greatly reduces their effectiveness and might create runoff channels. In addition, open areas
where wind velocity is high could present a maintenance challenge because high winds can
accelerate deterioration of the filter fabric (Smolen et al. 1988). When the pores of the silt fence
fabric become clogged with sediment, pools of water are likely to form uphill of the fence. Siting
and design of the silt fence should account for this problem, and care should be taken to avoid
unnecessary diversion of Stormwater from the pools that might cause further  erosion damage. Silt
fences can act as a diversion if placed slightly off-contour and can control shallow, uniform
flows from small, disturbed areas and deliver sediment-laden water to deposition areas.

Silt fences will sag or collapse if a site is too large, if too much sediment accumulates, if the
approach slope is too steep, or if the fence was not adequately supported. If the fence bottom is
                                           7-50

-------
                                                             Section 7: Technology Assessment
not properly installed or the flow velocity is too fast, fence undercuts or blowouts can occur
because of excess runoff. Erosion around the end of the fence can occur if the fence ends do not
extend upslope to prevent flow around the fence (IDNR 1992).

Maintenance
Site operators should inspect silt fences after each rainfall event to ensure that they are intact and
that there are no gaps at the fence-ground interface or tears along the length of the fence. If gaps
or tears are found, they should be repaired or the fabric should be replaced immediately.
Accumulated sediments should be removed from the fence base when the sediment reaches one-
third to halfway up the height of the fence. Sediment removal should occur more frequently if
accumulated sediment is creating a noticeable strain on the fabric, and there is the possibility that
the fence could fail from a sudden storm event.

Cost
There is a wide range of data on installation costs for silt fences. EPA estimates the costs at
approximately $6.00 per linear foot (USEPA 1992) while Southeastern  Wisconsin Regional
Planning Commission (SWRPC) estimates unit costs between $2.30 and $4.50 per linear foot
(SWRPC 1991). Silt fences have an annual maintenance cost that is 100 percent of installation
cost (Brown and Schueler 1997). Those values are significantly greater than that reported by R.S.
Means (2000), which indicates a 3-foot-tall silt fence installation costs between $0.68 and $0.92
per linear foot (for favorable and challenging installations). Note that the R.S. Means value
covers only installation, without the expected costs of maintenance (e.g., removal of collected
sediment). In addition, the type of silt fence fabric employed also affects the total installation
costs.

The Texas Commission on Environmental Quality reports that the cost  of a  12-inch diameter
compost filter sock ranges from $1.40 to $1.75 per linear foot when used as a perimeter control
(McCoy 2005). The costs for an 18-inch diameter sock used as a check  dam range from $2.75 to
$4.75 per linear foot (McCoy 2005).  Those costs do not include the cost of removing the
compost filter sock and disposing of the mesh after construction ends; however, filter socks are
often left on-site to provide slope stability  and post-construction stormwater control. The cost to
install a compost filter sock varies, depending on the availability of the  required quality and
quantity of compost and the availability of an experienced installer (USEPA 2008).

The Texas Commission on Environmental Quality reports that compost filter berms cost $1.90 to
$3.00 per linear foot when used as a perimeter control  and $3 to $6 per  linear foot when used  as
a check dam (McCoy 2005). The Oregon Department of Environmental Quality reports that
compost filter berms cost approximately 30 percent less to install than silt fences (ODEQ 2004).
Those costs do not include the cost of removal  and disposal of the silt fence or the cost of
dispersing the compost berm after construction ends. The cost to install a compost filter berm
varies, depending on the availability of the required quality of compost  in an area (USEPA
2008).
                                          7-51

-------
                                                               Section 7: Technology Assessment
      7.2.3.2.   Super Silt Fence

General Description
Super silt fence is a modification of a standard silt fence. The two main differences between the
standard silt fence and the super silt fence is that the super silt fence has a toe that is buried more
deeply, and the backing material is chain link fence held in place by steel posts—a concept that
originated in Maryland. The Maryland super silt fence requires a Geotextile Class F fabric over a
chain link fence to intercept sediment-laden runoff from small drainage areas. The super silt
fence provides a barrier that can collect and hold debris and soil more effectively than a standard
silt fence, preventing material from entering critical areas. It is best used where installing a dike
would  destroy sensitive areas, woods, and wetlands.

Applicability
Super silt fences can be used in the same conditions as a silt fence. Fences should follow the
contour of the land. Table 7-13  lists the distance a super silt fence should be from a slope to
ensure maximum effectiveness  (MDE 1994).
                        Table 7-13. Slope lengths for super silt fences
Slope
(%)
0-10
10-20
20-33
33-50
50+
Slope length
Minimum
Unlimited
200 feet
1 00 feet
1 00 feet
50 feet
Maximum
Unlimited
1 ,500 feet
1 ,000 feet
500 feet
250 feet
Design and Installation Criteria
As with the standard silt fence, design criteria are of two types:

       •  Hydrologic design for a required trapping of sediment and flow rate to pass the design
          storm

       •  Selecting appropriate installation criteria such that the silt fence performs as designed.

Hydrologic Design
Hydrologic design criteria are the same as those for the standard silt fence.

Installation Criteria
The criteria used for the Maryland super silt fence indicate the following, although they have not
been tested with field data:

       •  The fence should be placed as close to the contour as possible, with no section of the
          silt fence exceeding a grade of 5 percent for a distance of more than 50 feet.

       •  Fabric should be no more than 42  inches in height and should be held in place with a
          6-foot chain link fence.
                                           7-52

-------
                                                               Section 7: Technology Assessment
       •  Fabric should be attached to the steel pole using wire ties or staples. Fabric should be
          securely fastened to the chain link fence with ties spaced every 24 inches at the top
          and midsection.

       •  Fabric should be embedded into the ground at a minimum of 8 inches.

       •  Edges of fabric should overlap by 6 inches.

Table 7-14 describes the physical properties of Geotextile Class F fabric (MDE 1994).

                           Table 7-14. Minimum requirements for
                          super silt fence Geotextile Class F fabric
Physical properties
Tension strength
Tensile modulus
Flow rate
Filtering efficiency
Requirements
50 pounds/inch
20 pounds/inch
0.3 gallon/ft2/minute
75%
Effectiveness
EPA did not identify any performance data for super silt fences.

Limitations
Super silt  fences are not as likely to fail structurally as are standard silt fences, but they are more
expensive than standard silt fences.

Maintenance
Maintenance requirements for super silt fences are generally the same as for standard silt fences.

Cost
The cost of the super silt fence is more than the standard silt fence because of deeper burial at the
toe and the cost of chain linked fencing. R.S. Means (2000) indicates a rental price of $10 to $11
per linear  foot of chain linked fence for periods up to  1 year. Overall, rental is expected for most
construction site installation because rental rates are approximately half the price of permanent
chain link fencing.

      7.2.3.3.   Straw Bale Dike

General Description
The straw bale dike is a temporary measure used to trap sediment from small, sloping disturbed
areas. It is constructed of straw bales (not hay bales) wedged tightly together and placed along
the contour downslope of disturbed areas. The bales are placed in a shallow excavation, and the
upslope side is sealed with soil. Stakes are driven through the bales into the soil to help hold the
bales in place. The dike works by impounding water, which allows sediment to settle out in the
upslope area (Haan et al. 1994). Straw bale dikes are recommended for short duration application
and are usually effective for less than 3 months because of rapid decomposition (USDOT 1995).
                                           7-53

-------
                                                              Section 7: Technology Assessment
Applicability
Straw bale dikes are generally placed at the toe of fills to provide for a broad shallow sediment
pool. The dikes should not be used in drainage areas with concentrated and high flows, in large
drainage areas, or in ditches and swales. The location of the straw bale dike should be fairly
level, at least 10 feet from the toe, and should follow the land contour. Table 7-15 lists the
distance a straw bale dike should be placed from a slope to ensure maximum effectiveness.

             Table 7-15. Maximum land  slope and distances above a straw bale dike
Land slope
(%)
Less than 2%
2%-5%
5%-10%
10%-20%
More than 20%
Maximum distance above dam
(ft)
100
75
50
25
15
                        Source: USDOT1995.

Design and Implementation Criteria
Hydrologic Design
Hydrologic design dictates the structure necessary to withstand a storm without causing damage
while trapping the required amount of sediment. Either a database or some type of model is
needed to find the appropriate design. Efforts to model the sediment trapping that occurs in straw
bale dikes have resulted in models that predict the settling in the ponded area upstream from the
dike (Barfield et al. 1996; Lindley et al. 1998). The results from model simulations show that
trapping depends primarily on the surface area of the impounded water and flow rate through the
filter.  The models use a clear water slurry flow rate to  predict discharge. It is anticipated, on the
basis of visual observations, that sediment clogs the straw bale barrier, reducing the slurry flow
rate. Thus, results from model studies to date are suspect and need to be modified to account for
the effect of clogging on flow rate.

Installation Criteria
The USDOT's BMP  manual and the Indiana BMP manual call for bales to be

       •  Anchored by driving two 36-inch long (minimum) steel rebars or 2 x 2-inch
          hardwood stakes through each bale

       •  Sized according to the standard bale size of 14 inches x 18 inches x 35 inches

       •  Placed in  an excavated trench at least 4 inches deep, a bale's width, and long enough
          that the end  bales are somewhat upslope  of the sediment pool

       •  Abutted tightly against each other

       •  Sized so that impounded water depth does not exceed 1.5 feet

The USDOT BMP manual does not require that straw  bale dikes be designed; however, the
Indiana manual limits the drainage area to one-quarter acre per 100  feet of dam and the total
drainage area draining to a straw bale dike to 2 acres.
                                          7-54

-------
                                                              Section 7: Technology Assessment
Effectiveness
The information on straw bale dikes performance is very limited. In laboratory studies of bales at
varying orientations, Kouwen (1990) found that trapping efficiencies range from 60 to 100
percent. While field data on trapping have not been collected, bales deteriorate rapidly and need
to be replaced frequently. Because of such problems, using straw bale dikes as a perimeter
control is not recommended, except in special circumstances. Only 27 percent of ESC experts
rate the straw bale dike as an effective ESC practice, although their use is still allowed in half of
the communities surveyed (Brown and Caraco 1997).

Limitations
Straw bale dikes should not be used as a diversion, in streams, in channels, or in areas with
concentrated flow. The bales are not recommended for paved areas because of the inability to
anchor the bales (IDNR 1992).

Care must be taken to ensure that the bales are not installed in an area where there is a
concentrated flow of runoff, in a drainage area that is too large, or on an excessive slope (IDNR
1992). Under such conditions, erosion around the end of the bales, overtopping and undercutting
of the bales, and bale collapsing and dislodging are likely to occur. Overtopping also occurs if
the storage capacity is underestimated  and where provisions are not made for safe bypass of
storm flow (IDNR 1992). Undercutting occurs if the bales are not entrenched at least 4 inches
and backfilled with compacted soil or are not abutted or chinked properly. Straw bale dikes are
likely to collapse or dislodge if the bales are not adequately staked or if too much sediment is
allowed to accumulate before cleanout (IDNR 1992).

Maintenance
For the straw bale dike to be most effective, it is important to replace deteriorated bales when
appropriate.

Cost
The cost of straw bale dikes is relatively low, making their use attractive. R.S. Means (2000)
indicates a staked straw bale unit cost of $2.61 per linear foot (including materials, labor, and
equipment, with profit and overhead).

      7.2.3.4.   Sediment Trap

General Description
A sediment trap is a temporary control device used to intercept sediment-laden runoff and to trap
sediment to prevent or reduce off-site sedimentation. It is normally a more temporary type of
structure than a sediment pond and is constructed to control sediment on the construction area
during a selected phase of the construction operation. A sediment trap can be formed by
excavation or embankments or both constructed at designated locations accessible for cleanout.
The outlet for a sediment trap is typically a porous rock fill structure that detains the flow, but a
pipe structure can also be used. A temporary sediment trap can placed be in a drainageway,  at a
storm drain inlet, or at other points  of discharge from a disturbed area.  They can be constructed
independently or in conjunction with diversions and can be used in most drainage situations to
prevent excessive siltation of pipe structures (USEPA 1992).
                                           7-55

-------
                                                              Section 7: Technology Assessment
Applicability
Sediment traps can simplify the stormwater control plan design process by trapping sediment at
specific spots at a construction site (USEPA 1992). They should be installed as early in the
construction process as possible and are primarily effective as a short-term solution to trapping
sediment from construction sites (WYDEQ 1999). Natural drainage patterns should be noted,
and sites where runoff from potential erosion can be directed into the traps should be selected.
Traps are most effective when capturing runoff from areas where 2 to 5 acres drain to one
location. Sediment traps should not be in areas where their failure resulting from excess runoff
can lead to further erosive damage of the landscape. Alternative diversion pathways should be
designed to accommodate potential overflows. Traps should be accessible for clean-out and
placed so that they do not interfere with construction activity. In addition, the traps are easily
adaptable to most conditions.

Design and  Implementation Criteria
Hydrologic Design
A sediment trap should be designed to maximize surface area and sediment settling. That will
increase the effectiveness of the trap and decrease the likeliness of backup during and after
periods of high runoff intensity. The design of a trap includes determining the storage volume,
surface area, dimensions of spillway or outlet, and elevations of embankment (USDOT 1995).
Sediment traps should be designed to meet a 2-year, 24-hour storm event, but selecting a return
period varies among regulatory agencies (IDNR 1992).

Storage volume is created by a combination of excavation of land and construction of an
embankment to detain runoff (USDOT 1995). Trap storage volume and length of spillway are
determined as a function of the runoff volume and rate for the design storm. Such parameters
will vary depending on return period rainfall and watershed hydrologic characteristics. Some
standards specify a storage volume per acre disturbed. For example, Smolen et al. (1988) specify
that approximate storage capacity  of each trap should be at least 67 cubic yards per acre
disturbed draining into the trap, but more recent guidelines suggest 134 cubic yards per acre  of
drainage area (VDCR 2001). Any  national standard, however, should be based on runoff volume
and peak discharge to be generally applicable. Local regulations can translate that into applicable
volume and area standards.

A more important criterion than storage volume relates to sediment trapping. If a trapping
efficiency is specified, as in the case of South Carolina (SCDHEC 1995), it is necessary to
design for trapping efficiency. If a TSS or settleable solids effluent criterion is adopted
(SCDHEC 1995), settleable solids must be estimated. In both cases, a national standard should
address how to estimate trapping efficiency or settleable solids. Efforts to model the sediment
trapping that occurs in sediment traps have resulted in models that predict the settling in the
ponded area (Barfield et al.  1996;  Lindley et al. 1998). The results from model simulations show
that trapping depends  primarily on surface area of the impounded water and flow rate through
the rock fill outlet. In fact, the ratio of peak outflow rate to surface area is the best  simple
predictor of trapping. The models  use a modification of the Herrera and Felton (1991)
relationship developed by Haan et al. (1994) to predict discharge rates. The predicted flow rates
do not take into account clogging that can occur in rock fill. No models or procedures are
available to estimate this clogging or its effect on flow criteria.
                                           7-56

-------
                                                              Section 7: Technology Assessment
Design aids have also been developed for sediment traps, using simulations from the SEDEVICT
III (Barfield et al. 2001; Hayes et al. 2001). In the model, predictions are made of trapping
efficiency using the ratio of settling velocity for the dis of the eroded sediment, divided by the
ratio of discharge to ponded surface area. The design aid yields conservative estimates, but the
database used for generating the design aid is based on the assumption that flow rates are not
affected by clogging. That latter assumption is not likely to be a critical issue but should be
addressed in future research.

Installation Specifications
USDOT standards call for the embankment to be constructed of compacted earth, at a maximum
height of 5 feet (1.5 meters), a width of 4 to 5 feet (1.2 meters), and side slopes of 2:lor flatter.
Those values might change as a result of local criteria and with changing soil characteristics.
Temporary vegetation should be applied to the embankment.

Two types of outlet structures are typically used for sediment traps, a rock outlet and a pipe
outlet. Spillways of large stones or aggregate are the most common type of outlet designed for
sediment traps.  The crest of the spillway should be constructed 1 foot below the top of the
embankment and the spillway depth 1.5 feet below the top of the embankment. Weir length of
the spillway is determined on the basis of the contributing drainage area (Table 7-17) (USDOT
1995). The outlet apron should be  a minimum of 5 feet long, and placed on level ground with a
filter fabric foundation to ensure exit velocity of drainage to receiving stream is nonerosive
(IDNR 1992).

The length of the rock outlet should be determined on the basis of peak discharge required and
rock characteristics, typically rock diameter. Flow rate calculations can be made with the
relationship of Herrera and Felton (1991) as modified by Haan et al. (1994).  Alternatively,
USDOT has specified the weir length for a given drainage area as shown in Table 7-16.
However, the values  should be adjusted for each climatologic area to account for local
hydrologic and  return period rainfall.

                         Table 7-16. Weir length for sediment traps
Contributing drainage area
1
2
3
4
5
Weir length
(ft)
4
5
6
10
12
                      Source: USDOT 1995.

The pipe outlet, constructed of corrugated metal or PVC pipe riser, is an alternative to the rock
outlet. Pipe diameter is based on the peak discharge rate required. To obtain appropriate
freeboard, the top of pipe should be placed 1.5 feet below embankment elevation. Perforated
pipe is sometimes used. USDOT suggests perforations of 1-inch (25 mm) diameter holes or 0.5 x
6 inch (13 x 15 mm) slits in the upper two-thirds of the pipe; however, the discharge should be
calculated for this pipe specification to ensure that it matches the required peak discharge.
                                          7-57

-------
                                                              Section 7: Technology Assessment
The pipe should be placed vertically and horizontally above wet storage elevation (USDOT
1995). Riprap should be used as an outlet protection and placed at the outlet of the barrel to
prevent scour from occurring (USDOT 1995). A stable channel should be provided to convey
discharge to the receiving channel (USDOT 1995).

Effectiveness
If it is assumed that the flow can be accurately controlled by the rock fill outlet, sediment traps
should operate as effectively as sediment basins, with trapping efficiencies reduced as a result of
smaller surface areas. The NURP study (USEPA 1983),  Stahre and Urbonas (1990), and Haan, et
al. (1994), reports that sediment basins effectively trap sediment and chemicals as shown in
Table 7-17.

         Table 7-17. Range of measured pollutant removal for sediment detention basins
Item
Total suspended solids (TSS)
Total phosphorus (TP)
Nitrogen
Organic matter
Lead
Zinc
Hydrocarbons
Bacteria
Removable percentage
50%-70%
10%-20%
10%-20%
20%-40%
75%-90%
30%-60%
50%-70%
50%-90%
               Source: Stahre and Urbonas 1990.

Information on the actual effectiveness of sediment traps is limited. The discussion should start
first with the flow hydraulics of the rock fill outlet typically employed as a principal spillway for
sediment traps. Procedures for estimating flow through rock fill have been developed by Herra
and Felton (1991) to estimate flow as a function of average rock diameter, standard deviation of
rock size, and flow length. If those parameters could be controlled in an actual situation, the flow
could be accurately predicted. However, given that standard construction practices consist of
end-dumping the rock fill in place, one would expect little correlation between design and
construction, and the actual discharge and trapping efficiency would be expected to be
dramatically different from the design. This analysis does not mean that sediment traps are
ineffective but that a given design could not be guaranteed to meet the effluent criteria, even
though the predictions indicate compliance. Sediment trapping efficiency is a function of surface
area and inflow rate (Smolen et al. 1988). Those traps that provide pools with large length-to-
width ratios have a greater chance of success.

Sediment traps remove larger-sized sediment, primarily sized from silt to sands, by slowing
water velocity  and allowing for sediment settling in ponded water (Haan et al. 1994). Although
sediment traps allow for settling of eroded soils, because of their short detention periods for
stormwater they typically do not remove fine particles such as silts and clays without chemical
treatment. Sediment settling ability is related to the square of the particle size; halving particle
sizes quadruples the time needed to achieve settlement (WYDEQ  1999). To increase overall
effectiveness, traps should be constructed in smaller areas with low slopes. Sediment traps are
                                          7-58

-------
                                                             Section 7: Technology Assessment
typically designed to remove only sediment from surface water, but some non-sediment
pollutants are trapped as well (Haan et al. 1994).

Limitations
Common concerns associated with sediment traps are included in Table 7-18.

                Table 7-18. Common concerns associated with sediment traps
Common concern
Inadequate spillway size
Omitted or improperly installed geotextile fabric
Low point in embankment caused by inadequate
compaction and settling
Stone outlet apron does not extend to stable grade
Stone size too small or backslope too steep
Inadequate vegetative protection
Inadequate storage capacity
Contact slope between stone spillway and earth
embankment too steep
Outlet pipe installed in the vertical side of the trench
Corrugated tubing used as an outlet pipe
Result
Results in overtopping of the dam and possible failure
of the structure
Results in piping under the sides or bottom of the
stone and outlet section
Results in overtopping and possible failure
Results in erosion below the dam
Results in stone displacement
Results in erosion of embankment
Results in a less than adequate settling time (can also
be caused by an insufficient amount of sediment being
removed from the basin)
Results in piping failure
Results in piping failure of embankment
Results in crushed pipe and inadequate outlet capacity
Source: IDNR 1992.

Maintenance
The primary maintenance consideration for temporary sediment traps is removing accumulated
sediment from the basin, which must be done periodically to ensure the continued effectiveness
of the sediment trap. Sediments should be removed when the basin reaches approximately 50
percent sediment capacity.

A sediment trap should be inspected after each rainfall event to ensure that the trap is draining
properly. Inspectors should also check the structure for damage from erosion or piping. The
depth of the spillway should be checked and maintained at a minimum of 1.5 feet below the low
point of the trap embankment.

Cost
The cost of installing temporary sediment traps ranges from $0.20 to $2.00 per cubic foot of
storage (about $1,100 per acre of drainage). EPA estimates the following costs for sediment
traps, which vary as a function of the volume of storage: $513 for 1,800 cubic yards, $1,670 for
3,600 cubic yards, and $2,660 for 5,400 cubic yards (USEPA 1993). Evaluation of a series of
more recent data sources (USEPA 2003) indicates that sediment traps have an average cost of
$0.30 per cubic foot of storage. In addition, it has been reported that a sediment trap has an
annual maintenance cost of 20 percent of installation cost (Brown and Schueler 1997).
                                          7-59

-------
                                                              Section 7: Technology Assessment
      7.2.3.5.   Sediment Basin

General Description
A sediment basin is a stormwater detention structure formed by constructing a dam across a
drainageway or excavating a storage volume at other suitable locations and using it to intercept
sediment-laden runoff. Sediment basins are generally larger and more effective in retaining
sediment than temporary sediment traps and typically remain active throughout the construction
period. Jurisdictions that require post-development flow to be less than or equal to
predevelopment flow during construction could employ the designed detention facilities as a
temporary sediment basin during construction.

When sediment basins are designed properly, they can control sediment pollution through the
following functions (Faircloth 1999):

       •  Sediment-laden runoff is caught to form an impoundment of water and create
          conditions where sediment will settle to the bottom of the basin.

       •  Treated runoff is released with less sediment concentration than when it entered the
          basin.

       •  Storage is provided for accumulated sediment, and resuspension by subsequent
          storms is limited.

Applicability
Sediment basins should be located at a convenient concentration point for sediment-laden flows
(NCDNR 1988). Ideal sites are areas where natural topography allows a pond to be formed by
constructing a dam across a natural swale; such sites are preferred to those that require
excavation (Smolen et al. 1988).

Sediment basins are also applicable in drainage areas where it is anticipated that other erosion
controls, such as sediment traps, will not be sufficient to prevent off-site transport of sediment.
Choosing to construct a sediment basin with either an earthen embankment or a stone/rock dam
will depend on the materials available, location of the basin, and desired capacity for stormwater
runoff and settling of sediments.

Rock dams are suitable where earthen embankments would be difficult to construct or where
riprap is readily available. Rock structures are also desirable where the top of the dam structure
is to be used as an emergency overflow outlet. Such riprap dams are best  for drainage areas of
less than 50 acres. Earthen damming structures are appropriate where failure of the dam will not
result in substantial damage or loss of property or life. If properly constructed, sediment basins
with earthen dams can handle stormwater runoff from drainage basins as  large as 100 acres.

Design and Implementation Criteria
Hydrologic Design
A sediment basin can be constructed by excavation or by erecting an earthen embankment across
a low area or drainage swale. Sediment basins can be designed to drain completely during dry
periods, or they can be constructed so that a shallow, permanent pool of water remains between
                                          7-60

-------
                                                             Section 7: Technology Assessment
storm events. Depending on the size of the basin constructed, the basin might be subject to
additional regulation, particularly state and federal regulations related to dam safety.

Sediment basins can be used for any size watershed, but USDOT recommends a drainage area
range of 5 to 100 acres (USDOT 1995). Components of a sediment basin that must be considered
in the hydrologic design include the following (Haan et al. 1994):

       •  A sediment storage volume sized to contain the sediment trapped during the life of
          the structure or between cleanouts

       •  A permanent pool volume (if included) above the sediment storage to protect trapped
          sediment and prevent resuspension as well as providing a first flush of discharge that
          has been subjected to an extended detention period

       •  A detention volume that contains storm runoff for a period sufficient to trap the
          necessary quantity of suspended solids

       •  A principal spillway that can be a drop-inlet pipe and barrel, a trickle tube, or other
          type of controlled release structure

       •  An emergency spillway that is designed to handle excessive runoff from the rarer
          events and prevent overtopping

The following recommended procedures for conducting the hydrologic design are summarized
from Haan et al. (1994).

Sediment Storage Volume. This volume should be sufficient to store the sediment trapped
during the life of the structure or between cleanouts. Sediment storage volume can be calculated
on the basis of sediment yield using relationships such as the RUSLE with an appropriate
delivery ratio (Renard et al.  1994) or a computer model such as SEDIMOT III (Barfield et al.
1996) or SEDCAD (Warner et al. 1999). Many design specifications, however, base the
sediment storage volume on a volume per acre disturbed. For example, Pennsylvania specifies a
sediment storage volume of 1,000 cubic feet per acre drained (see DCN 43050, Pennsylvania
Erosion and Sediment Pollution Control Program Manual). This volume is highly site-specific,
depending on rainfall distributions, soil types, and construction techniques.

Permanent Pool Volume. Providing a first flush of discharge that has been subjected to an
extended detention period can help to minimize degradation of water quality and justify some
permanent pool. The recommended capacity of the permanent pool varies with the regulatory
agency. USDOT, for example, recommends 67 cubic yards per acre (126 m3/ha) (USDOT 1995).
That standard has been adopted by many states as well. If an effluent criterion such as allowable
peak TSS or peak settleable solids is used, the final design of both permanent pool and detention
volume should be selected only after using a computer model to predict the expected peak
effluent concentrations.

Detention Volume. Storm runoff must be contained for a period of time sufficient to trap the
necessary quantity of suspended solids. Because inflow is occurring simultaneously with outflow,
the detention time for each plug of flow is different and should be considered individually. The
size of the detention volume, as stated above, should also be developed in concert with determining
the size of the permanent pool volume and the size of the principal spillway. When effluent TSS
                                          7-61

-------
                                                              Section 7: Technology Assessment
and settleable solids criteria are used, the size of the detention volume and permanent pool volume
should be determined through a computer model calculation of expected effluent concentrations
for a given design. The return period used to size the detention volume depends on the regulatory
agency, but a return period of 10 years is typical for sediment basins that eventually become
stormwater detention ponds (i.e., are used to limit future flooding due to stormwater). EPA's
review of state construction site regulations found that the majority of states specify detention
volume in terms of cubic feet per acre that drains to the sediment basin. State design values range
between 1,800 and 5,400 cubic feet per acre, with 3,600 cubic feet per acre or expected runoff
from the local 2-year, 24-hour storm event as the typical value.

Principal Spillway. The principal spillway is a hydraulic outlet structure sized to provide the
appropriate outflow rate to meet the effluent or trapping efficiency criteria. The principal
spillway should have a dewatering device that slowly releases water contained in the detention
storage over an extended period and at a rate determined to trap the required amount of sediment
or provide for the appropriate effluent concentration in the design storm. The more common
outlet structures are the drop-inlet structure and the trickle tube. Sizing of the principal spillway
should follow standard design procedures with respect to hydrology and sediment considerations,
but sizing the structure to simply pass the design storm is inappropriate and will not result in
meeting an effluent or trapping efficiency standard. The size to be used in a given structure
should be determined on the  basis of the effluent or trapping efficiency standard being targeted
and site-specific hydrologic and soil conditions. Appropriate design will require the use of a
computer model such as SEDEVIOT III (Barfield et al. 1996) or design aids such as those
developed for South Carolina (Hayes and Barfield 1995). In general, the design is developed to
maximize surface area, which will minimize peak discharge. Because failure of the dam could
result in downstream damage, the design should be done and certified by a licensed engineer
with expertise in hydrologic computation.

For discussion of skimmers in lieu of rock and perforated outlets in sediment traps and basins,
see Section 7.2.3.6.

Emergency Spillway. Because overtopping of the dam can cause failure and downstream damage,
an emergency spillway is necessary to handle excessive runoff from the larger, less frequent events
and prevent overtopping. The design storm for the emergency spillway will depend on the hazard
classification of the sediment basin. Typical return periods vary between 25 and 100 years, with 25
years recommended by USDOT. Sizing of the emergency spillway is typically accomplished to
simply transmit the rare event without eroding the base of the spillway. Procedures for making the
hydrologic and hydraulic computations are summarized in Haan et al. (1994). Again, because
failure of the dam could result in downstream damage, the design should be done and certified by a
licensed engineer with expertise in hydrologic computation.

Installation Criteria
The embankment for permanent sediment basins should be designed using standard geotechnical
construction techniques. The fill is typically constructed of earthen fill material placed and
compacted in continuous layers over the entire length of the fill. USDOT recommends 6- to 8-
inch layers (USDOT  1995). The embankment should be stabilized with vegetation after
construction of the basin. A cutoff trench should be excavated along the centerline of the dam to
prevent excessive seepage beneath the  dam and be sized using standard geotechnical
                                          7-62

-------
                                                               Section 7: Technology Assessment
computations. USDOT recommends that a minimum depth of the cutoff trench be approximately
2 feet (600 mm), the height should be to the riser crest elevation, the minimum bottom width
should be 4 feet (1.2 m) or wide enough for compaction equipment, and slopes should be no
steeper than 1:1.

Sediment basins can also be constructed with rock dams in a design that is similar to a sediment
basin with an earthen embankment. It is important to remember that rock fill is highly
heterogeneous and that flow rates calculated with any available procedure are not likely to match
those that will actually occur. Because sediment trapping is inversely proportional to flow rate,
the trapping efficiency will be affected  significantly. No data are available to determine the
variability of rock fill in actual installations so that confidence intervals can be placed on
predicted flow rates. Such data should be collected and the confidence intervals calculated before
recommending  the use of rock dams as  outlets on any structures other than sediment traps.

Effectiveness
The effectiveness of a sediment basin depends primarily on the sediment particle size and the
ratio of basin surface area to inflow rate (Smolen et al. 1988; Haan et al. 1994). Basins with a
large surface area-to-volume ratio will be most effective. Studies by Barfield and Clar (1985)
show that a surface area-to-peak discharge ratio of 0.01 acre per cubic foot would trap more than
75 percent of the sediment coming from the Coastal Plain and Piedmont regions in Maryland.
That efficiency might vary for other regions of the country and should not be used as a national
standard. Studies by Hayes et al. (1984) and Stevens et al. (2001), however, show that similar
relationships can be developed for other locations.

Laboratory data collected on pilot-scale facilities are available on the trapping efficiency of
sediment basins, effluent concentrations, dead storage and flow patterns, and the effects of
chemical flocculants on sediment trapping  (Tapp et al. 1981; Wilson and Barfield 1984; Griffin
et al. 1985; Jarrett 1999; Ward et al.  1977,  1979). In general, the laboratory studies show that
pilot-scale ponds can be expected to trap 70 to 90 percent of sediment, depending on the
sediment characteristics, pond volume,  and flow rate.  The trapping efficiency and effluent
concentration are, in general, related to  the overflow rate and can be reasonably well predicted
using a plug flow model (Ward et al. 1977, 1979) and a Continuously Stirred Tank Reactor
(CSTR) model  (Wilson et al. 1982; Wilson et al. 1984). Extensive field-scale data are available
on long-term trapping efficiency in stormwater detention basins in which the annual trapping
efficiency is related to the annual capacity  inflow ratio of the basin. These structures are not
representative of those used for sediment ponds but would be representative of those used for
regional detention. A more limited database is available on single storm sediment trapping in the
larger structures (Ward, et al.  1979) and on a field laboratory structure at Pennsylvania State
University (Jarrett et al. 1999).

For maximum trap efficiency, Smolen et al. (1988) recommend the following:

       •  Allow the largest surface area possible, maximize the length-to-width ratio of the
          basin to prevent short circuiting, and ensure use of the entire design settling area.

       •  Locate inlets for the basin at the maximum distance from the principal spillway
          outlet.
                                           7-63

-------
                                                              Section 7: Technology Assessment
       •  Allow the maximum reasonable time to detain water before dewatering the basin.

       •  Reduce the inflow rate into the basin and divert all sediment-free runoff.

Jarett (1999) has shown that the smaller the depth of the basin, the more sediment is discharged.
A 0.15-meter-deep (0.49-foot-deep) basin lost twice as much sediment as a 0.46-meter-deep
(1.5-foot-deep) basin. Jarrett also found that the performance of a sediment basin will increase
with the use of a skimmer in the principal spillway. The sediment discharged was 1.8 times
greater with only a perforated riser than with a skimmer in the principal spillway. In addition,
increasing the dewatering time, which allows for more sediment deposition, decreases the
sediment loss from the basin (Jarrett 1999).

Table 7-19 presents a summary of sediment basin monitoring or modeling data. Table 5-1 shows
corresponding influent TSS data when available. For summaries of studies with monitoring or
modeling data, and annotated bibliographies for the journal articles and professional conference
proceedings that EPA reviewed, see DCN 44321.

  Table 7-19. Studies of TSS in sediment basin effectiveness and effluent from construction sites
Site
Seattle, Washington
SR204
Mercer Island
SB1
SB2
SB4
Pennsylvania Test
Basin
Georgia Model
Maryland Model
Hamilton County, Ohio
Johnston County, North
Carolina SkB1
Mean TSS (mg/L)
Mean effluent TSS
concentration
(mg/L)
154
626
63
322
91
875
800
600
700
3,507
1,042
798
Mean TSS
reduction
(percent)
98.6%
86.7%
75.1%
54.7%
80.3%
66.8%
94.2%
65%
84%
35%
87%
75%
Source
Horner et al. 1990
Horneretal. 1990
Horner etal. 1990
Schueler and Lugbill 1990
Schueler and Lugbill 1990
Schueler and Lugbill 1990
Jarrett 1 996
Sturm and Kirby 1991
Barfield and Clar 1985
Islam etal. 1998
Markusic and McLaughlin 2008;
McLaughlin and Markusic 2007.
N/A
 N/A - Not Applicable

Limitations
Neither a sediment basin with an earthen embankment nor a rock dam should be used in areas of
continuously running water (live streams). Using sediment basins is not intended for areas where
failure of the earthen or rock dam will result in loss of life, damage to homes or other buildings,
or interference with the use of public roads or utilities.

Because sediment basins are usually temporary structures, they are often designed poorly and
rarely receive adequate attention and maintenance. As a result, such basins will not achieve the
function for which they were designed, especially when conventional outlets cannot properly
meter outflow to create an impoundment, thus allowing rapid release of sediment-laden water
from the bottom of the basin to escape (Faircloth 1999).
                                          7-64

-------
                                                             Section 7: Technology Assessment
Common concerns associated with sediment basins are included in Table 7-20.
               Table 7-20. Common concerns associated with sediment basins
Common concern
Improper compaction, omission of anti-seep collar,
leaking pipe joints, or use of unsuitable soil
Inadequate vegetation or improper grading and
sloping
Inadequate compaction or use of unsuitable soil
Steep side slopes
Inadequate outlet protection
Basin not located properly for access
Sediment not properly removed
Lack of anti-flotation
Principal and emergency spillway on design plans
Gravel clogging the dewatering system
Principal spillway too small
Result
Results in piping failure along conduit
Results in erosion of spillway or embankment slopes
Results in slumping or settling of embankment
Results in bank failure due to slumping
Results in erosion and caving below principal spillway
Results in difficult, ineffective, and costly maintenance
Results in inadequate storage capacity and potential
resuspension
Results in the riser and barrel being blocked with debris
Results in improper disposal of accumulated sediment
Results in safety or health hazard from pond water
Results in frequent operation of emergency spillway
and increased erosion potential
Source: IDNR 1992.

Maintenance
Routine inspection and maintenance of sediment basins is essential to their continued
effectiveness. Basins should be inspected after each storm event to ensure proper drainage from
the collection pool and determine the need for structural repairs. Erosion from the earthen
embankment or stones moved from rock dams should be repaired or replaced immediately.

Sediment basins must be in an area that is easily accessible to maintenance crews for removal of
accumulated sediment. Sediment should be removed from the basin when its storage capacity has
reached approximately 50 percent. Trash and debris from around dewatering devices should be
removed promptly after rainfall events.

Cost
If constructing a sediment basin with less than 50,000 cubic feet of storage space, the cost of
installing the basin ranges from $0.20 to $1.30 per cubic foot of storage (approximately $1,100
per acre of drainage) with an average cost of approximately $0.60 per cubic foot of storage
(USEPA 1993). If constructing a sediment basin with more than 50,000 cubic feet of storage
space, the  cost of installing the basin ranges from $0.10 to $0.40 per cubic foot of storage
(approximately $550 per acre of drainage) with an average cost of approximately $0.30 per cubic
foot of storage (USEPA 1993).  A review of state highway project bids and county bonding
estimates conducted in 2003 confirms the value of $0.30 per cubic foot (USEPA 2003). Annual
maintenance costs are  25 percent of installation costs (Brown and Schueler 1997).

R.S. Means (2000) suggests the cost to remove the eroded sediment collected in a small basin
during construction is  approximately $4 per cubic yard (that value includes a 100 percent
surcharge  for wet  excavation). Disposal of material on-site will result in an additional cost that
can be computed only  from site-specific conditions. The cheapest management of dredged
                                          7-65

-------
                                                              Section 7: Technology Assessment
material is application to land areas adjacent to the basin followed with application of a
vegetative cover.

      7.2.3.6.   Faircloth Skimmer

General Description
 A Faircloth Skimmer® is a surface drain that floats on top of the water in a sediment basin. The
skimmer inlet controls the rate of outflow and rises and falls as the basin fills and drains. It
releases the cleanest water in the basin from near the surface. Although the Faircloth Skimmer is
a proprietary device, the same concept applies to any device that withdraws water from the
surface of the basin as opposed to dewatering through a perforated riser or stone outlet structure.

Applicability
A Faircloth Skimmer is used instead of the rock  and perforated riser outlets in sediment traps and
basins.

Design and Implementation Criteria
The Faircloth Skimmer can be attached directly to an outlet pipe that drains through the dam or
attached to an outlet pipe through a riser. The key design parameters in sizing a Faircloth
Skimmer is volume to drain and the length of time for the basin to drain. As the size of the
skimmer increases, the basin drainage time decreases. Faircloth recommends 3 days in the
absence of state specifications. North Carolina specifies 1 to 3 days and Pennsylvania 4 to 7
days.

Effectiveness
The skimmer allows water to be released from the top of the basin, which is the cleanest water
(Faircloth 1999). EPA summarizes skimmer basin performance data from an active construction
site in Johnston County, North Carolina (see DCN 44321).

Limitations
There are many factors in addition to a surface drain for a basin to be efficient. For limitations of
sediment basins, see Section 7.2.3.5.

Maintenance
Routine inspection and maintenance of sediment basins with or without skimmers is essential to
their continued effectiveness. For maintenance of sediment basins, see Section 7.2.3.5.

Cost
EPA obtained Faircloth  Skimmer equipment and shipping costs and added costs for additional
required ancillary equipment (e.g., PVC pipe, glue), as well as labor for installation. Assuming a
3-day drainage time, EPA developed the following cost equation:

Total  skimmer cost (2009 dollars) = 0.0138 x (basin volume in cubic feet) +  1,049.

DCN 43113 documents the development of that  cost equation.
                                           7-66

-------
                                                             Section 7: Technology Assessment
      7.2.3.7.   Enhanced Sediment Trapping

General Description
Work in recent years has focused on a number of passive, PAM-based systems to enhance
pollutant removal in sediment basins. Other chemicals used in such passive systems include
chitosan acetate, chitosan lactate, gypsum, and alum. PAM, available mfloc logs, has also seen
increased placement in conveyance channels. Chitosan lactate gel socks have also been used in
that application. As water flows through the channel, the chemical dissolves, and the turbulence
in the channel aids in the flocculation process. Floes can then settle out in sediment control
devices, such as check dams, sediment traps or basins. PAM (and other flocculants) can also be
added in  liquid form to stormwater and is commonly used to dose sediment basins to help
remove sediment. For discussion of fiber check dams installed at grade with PAM applied to the
check dam for passive dosing, see Section 7.2.2.5. At least one vendor is also using a tube settler
coupled with polymer addition before  filtration to help remove sediment.

Applicability
Treatment chemicals can enhance sediment removal when traditional BMPs are not capable of
meeting numeric standards (e.g., because of fine-grained, suspended sediment or colloidal
particles, or a retention device design is not optimal because of site limitations). Auckland
Regional Council (2004) notes that passive treatment using flock blocks requires no power and is
less expensive and less complex than active systems.

Design and Implementation Criteria
For information on commonly available coagulant/flocculants and toxicity information, see
Section 7.2.5.

Effectiveness
McLaughlin demonstrated the ability to meet a 50-NTU limit at  a research site in North Carolina
by adding PAM to a basin equipped with baffles and a surface skimmer (see DCN 43082, The
Potential for Substantial Improvements in Sediment and Turbidity Control). North Carolina now
requires skimmers on all sediment basins, and the North Carolina Department of Transportation
has developed draft standards for the use of porous baffles in sediment basins (see DCNs 43083
and 43045, NCDOT draft baffles standards and North Carolina Erosion and Sediment Control
Planning and Design Manual with requirements for skimmers). Bhardwaj and McLaughlin
(2008a) found that both active and passive-dosed PAM systems  significantly reduced turbidity,
with the active dosing being slightly more effective. Bhardwaj and McLaughlin (2008a, 2008b)
and McLaughlin (2006) reports that basin modifications (e.g., baffles, outlet type)  have minor
effects on turbidity in comparison to PAM addition. For summaries of studies with monitoring
data and annotated bibliographies for the journal articles and professional conference
proceedings that EPA reviewed, see DCN 43114.

Limitations
McLaughlin (2006) reports that blocks that were allowed to dry  were much less effective.
McLaughlin (No Date b) notes that blocks in ditches without slope tend to become buried more
easily than blocks placed in stepper ditches. Auckland Regional  Council (2004) notes that the
                                          7-67

-------
                                                               Section 7: Technology Assessment
primary disadvantage of the passive blocks is that the exact dosage is unknown, dependent on
flow and condition of the block.

Maintenance
Routine inspection and maintenance of BMPs with or without passive chemical treatment is
essential to their continued effectiveness.

Cost
PAM can be used in a centralized treatment system (e.g., at a sedimentation basin) to treat larger
areas, or dispersed in granular or liquid form. In Tobiason et al. (2000), the startup costs for the
batch treatment system at a large airport construction project amounted to $90,000, although the
author notes that costs for some of the initial piping might have been unwarranted. Monthly
expenses average $18,000 for operations and maintenance and $13,000 for materials and
equipment, but the author notes that high monthly costs were driven by record rainfall  and
extremely wet weather experienced. The author states that "passive dosing systems being tested
as a complementary BMP present considerable cost savings and may provide similar
effectiveness." The total costs for this phase totaled about $245,000, less than 1 percent of total
construction costs. Auckland Regional Council (2004) reports a total cost of approximately
$2,400 per installation for a rainfall-driven, liquid dosing system that does not require flow
runoff measurement or a dosing pump.

   7.2.4.   OTHER CONTROL PRACTICES

      7.2.4.1.   Stone Outlet Structure

Description
A stone outlet structure is a temporary stone  dike installed in conjunction with and as a part of an
earth dike. The purpose of the stone outlet structure is to impound sediment-laden runoff,
provide a protected outlet for an earth dike, provide for diffusion of concentrated flow, and allow
the area behind the dike to dewater slowly. The stone outlet structure can extend across the end
of the channel behind the dike or be placed in the dike itself. In some cases, more than one stone
outlet structure can be placed in a dike.

Applicability
Stone outlet structures apply to any point of discharge where there is a need to discharge runoff
at a protected outlet or to diffuse concentrated flow for the duration of construction. The drainage
area to this practice is typically limited to one-half acre or less to prevent excessive flow rates.
The stone outlet structure should be located so as to discharge onto an already stabilized area or
into a stable watercourse. Stabilization should consist of complete vegetative cover and paving
that are sufficiently  established to be erosion resistant.

Design and Installation Criteria
Design criteria are of two types: hydrologic design for a required trapping of sediment or flow
rate to pass the design storm; and selecting appropriate installation criteria such that the stone
outlet performs as designed.
                                           7-68

-------
                                                               Section 7: Technology Assessment
Hydrologic Design
The hydrologic design should be based on the design storm and standard hydraulic calculations.
It should include the following considerations:

       •  Design rainfall and design storm. The design storm should be specified by the
          regulatory authority. Typically a return period of 2 to 5 years is used. Runoff rates
          should be calculated with standard hydrologic procedures as allowed by the
          regulatory authority.

       •  Drainage area. The drainage area to this structure is  typically limited to less than half
          an acre to ensure that the flow rates are not excessive.

       •  Length of crest and height of stone fill. The crest length and height of stone fill
          should be of sufficient size to transmit the design storm without overtopping. The
          volume of water stored behind the dike can be estimated but would require routing
          the storm flow in the design storm. Flow through the stone outlet can be calculated
          using the relationships of Herrera and Felton (1991) as modified by Haan et al.
          (1994). The height of the fill should be small enough to prevent excessive flow
          velocities through the stone fill and prevent undercutting.

       •  Outlet stabilization. The discharge from the stone outlet should be stabilized with
          vegetated waterways or riprap until the flow reaches a stable channel. Design of the
          stabilized outlet should follow procedures presented earlier.

Installation Criteria Specifications
A stone outlet structure should conform to the following specifications:

       •  The outlet should be composed of 2- to 3-inch stone or recycled concrete, but clean
          gravel can be used if stone is not available.

       •  The crest of the stone dike should be at least 6 inches lower than the lowest elevation
          of the top of the earth dike and should be level.

       •  The stone outlet structure should be embedded into the soil a minimum of 4 inches.

       •  The minimum length of the crest  of the stone outlet  structure should be 6 feet.

       •  The baffle board should extend 1  foot into the dike and 4 inches into the ground and
          be staked in place.

       •  The drainage area to this structure should be less than one-half acre.

      7.2.4.2.   Rock Outlet Protection

Description
Rock outlet structures are rocks that are placed at the outfall of channels or culverts to reduce the
velocity of flow in the receiving channel to nonerosive rates.
                                           7-69

-------
                                                               Section 7: Technology Assessment
Applicability
This practice applies where discharge velocities and energies at the outlets of culverts are
sufficient to erode the next downstream reach and is applicable to outlets of all types such as
sediment basins, stormwater management ponds, and road culverts.

Design and Installation Criteria
Hydrologic Design
Hydrologic design consists primarily of selecting the design runoff rate and sizing outlet
protection.  Standard hydrologic calculations should be used with an appropriate return period
storm for the outlet being protected (typical return periods range from 2 to 10 years).

The process for sizing outlet protection involves selecting the type and geometry of the outlet
protection and the size of the rock lining. The outlet protection could consist of a plunge pool
(scour hole), an apron-type arrangement, or an energy dissipation basin (Haan et al. 1994). The
design of each differs. Plunge pools are typically used for outlet pipes that are elevated above the
water surface. Aprons are used for other types of outlets. Plunge pool geometry is based on the
flow rate, pipe size and slope, tailwater depth, and size of the riprap lining (Haan et al. 1994).
Apron dimensions are determined by the ratio of the tailwater depth to pipe diameter (Haan et al.
1994). Energy dissipation basins are used as an alternative to the plunge pool. Dimensions are a
function of the brink depth in the pipe at the design flow, pipe diameter, and size of riprap (Haan
et al. 1994). The size of the rock lining is a function of the discharge, pipe size, tailwater depth,
and geometry selected. Details on sizing the rock are given in Haan et al. (1994).

The design method presented here applies to the sizing of rock riprap and gabions to protect a
downstream area. It does not apply to rock lining of channels or streams.  The design of rock
outlet protection depends entirely on the location. Pipe outlets at the top of cuts or on slopes
steeper than 10 percent cannot be protected by rock aprons or riprap sections because of
reconcentration of flows and high velocities encountered after the flow leaves the apron.

Installation Criteria
The following criteria should be considered:

       •  Bottom grade: The outlet protection apron should be  constructed with zero slope
          along its length. There should be no obstruction at the end of the apron.  The elevation
          of the downstream end of the apron should be equal to the elevation of the receiving
          channel or adjacent ground.

       •  Alignment: The outer protection apron should be located so that there are no beds in
          the horizontal alignment.

       •  Materials:  The  outlet protection can be accomplished using rock riprap or gabions.
          Riprap should be composed of a well-graded mixture of stone sized so that 50 percent
          of the pieces, by weight, should be larger than the size determined using charts. The
          minimum dso size to be used should be 9 inches. A well-graded mixture is defined as
          a mixture composed primarily of larger stone sizes but with a  sufficient mixture of
          other sizes to fill the smaller voids between the stones. The diameter of the largest
          stone in such a mixture should be two times the size selected in Table 7-21 (MDE
           1994).
                                           7-70

-------
                                                        Section 7: Technology Assessment
   Thickness: Riprap specification values are summarized in Table 7-21.
                   Table 7-21. Riprap sizes and thicknesses

Class I
Class II
Class III
D50
(inches)
9.5
16
23
Duo
(inches)
15
24
34
Thickness
(inches)
19
32
46
 Source: USDOT1995

•   Stone Quality: Stone for riprap should consist of field stone or rough-hewn quarry
    stone. The stone should be hard and angular and of a quality that will not disintegrate
    on exposure to water or weathering. The specific gravity of the individual stones
    should be at least 2.5. Recycled concrete equivalent can be used, provided it has a
    density of at least 150 pounds per cubic foot and does not have any exposed steel or
    reinforcing bars.

•   Filters:  A layer of material placed between the riprap and the underlying soil surface
    can prevent soil movement into and through the riprap to prevent piping, reduce uplift
    pressure, and collect water. Riprap should have a filter placed under it in all cases. A
    filter can be of two general forms: a gravel layer or a geotextile.

•   Gabions: Gabion baskets can be used as rock outlet protection,  provided they are
    made of hexagonal, triple-twist mesh with heavily galvanized steel wire. The
    maximum lined dimension of the mesh opening should not exceed 4.5 inches. The
    area of the mesh opening should not exceed 10 square inches. Gabions should be
    fabricated in such a manner that the sides, ends, and lid can be assembled at the
    construction site into a rectangular basket of the specified sizes.
    Gabions should be of a single-unit construction and installed according to the
    manufacturer's specifications. Foundation conditions should be the same as for
    placing rock riprap. Geotextiles should be placed under all gabions, and gabions must
    be keyed in to prevent undermining of the main gabion structure.

•   The subgrade for the filter, riprap, or gabion should be prepared to the required lines
    and grades. Any fill required in the subgrade should be compacted to a density of
    approximately that of the surrounding undisturbed material.

•   The rock or gravel should conform to the specified grading limits when installed in
    the riprap or filter, respectively.
•   Geotextiles should be protected from punching, cutting, or tearing. Any damage other
    than occasional small holes should be repaired by placing another piece of geotextile
    fabric over the damaged part or by completely replacing  the geotextile fabric. All
    overlaps, whether for repairs or for joining two pieces of geotextile fabric, should be a
    minimum of 1 foot in length.

•   Stone for the riprap or gabion outlets can be placed by equipment. They should be
    constructed to the full course thickness in one operation and in  such a manner as to
    avoid displacement of underlying materials. Care should be taken to ensure that the
                                    7-71

-------
                                                               Section 7: Technology Assessment
          stone is not placed so that rolling would cause segregation of stone by size, i.e., the
          stone for riprap or gabion outlets should be delivered and placed in a manner that will
          ensure that it is reasonably homogeneous, with smaller stones filling the voids
          between larger stones. Riprap must be placed so as to prevent damage to the filter
          blanket or geotextile fabric. Hand placement will be required to the extent necessary
          to prevent damage to the permanent works.

       •  Stone should be placed so that it blends in with the existing ground and the depth to
          the stone surface is sufficient to transmit the flow without spilling over onto the
          unprotected surface.

Effectiveness
No information is available on the effectiveness of rock outlet structures.

Limitations
Common problems with rock outlet structures include the following:

       •  If the foundation is not excavated deeply or wide enough, the flow cross-section
          could be restricted,  resulting in erosion around the apron and scour holes at the outlet.
          Also, the riprap apron should be placed on a suitable foundation to prevent
          downstream erosion.

       •  If the riprap that is installed is  smaller than specified, rock displacement might result;
          selectively grouting over the rock materials could stabilize the installation.

       •  If the riprap is not extended enough to reach a stable section of the channel,
          downstream erosion could result.

       •  If a filter is not installed under the riprap, stone displacement and erosion of the
          foundation might result.

Maintenance
Once a riprap outlet has been installed, the maintenance needs are very low. It should be
inspected after high flows to see if scour has occurred beneath the riprap, if flows have occurred
outside the boundaries of the riprap and caused scour, or if any  stones have been dislodged.
Repairs should be made immediately.

Cost
R.S. Means  (2000) indicates machine-placed riprap costs of approximately $40 per cubic  yard.
For a riprap  maximum size between 15 and 24 inches, a cubic yard of riprap will cover between
13.5 and 17  square feet at channel bed (assuming depth of riprap as given in Table 5-22).  This
suggests that riprap lining will be between $21 and $27 per square foot of outlet (which includes
materials,  labor, and equipment, with overhead and profit). R.S. Means (2000) provides a cost
range for gabions ($2.80 to $9  per square  foot of coverage) for stone fill depths of 6 to 36 inches,
respectively. Those costs include all costs of materials, labor, and installation.
                                           7-72

-------
                                                              Section 7: Technology Assessment
      7.2.4.3.   Sump Pit

Description
A sump pit is a temporary pit from which pumping is conducted to remove excess water while
minimizing sedimentation. The purpose of the sump pit is to filter water being pumped to reduce
sedimentation to receiving streams.

Applicability
Sump pits are constructed when water collects and must be pumped away during excavating,
cofferdam dewatering, maintenance or removal of sediment traps and basins, or other uses as
applicable, such as for concrete wash out.

Design and Installation Criteria
Hydrologic Design
The only hydrologic calculation is determining the expected flow rate and volume to be handled.
That should follow standard hydrologic computational procedures based  on design rainfall,
surface and soil conditions, and the size of the pump.

Installation Criteria and Specifications
The number of sump pits and their locations should be determined by the designer and included
on the plans. Contractors can relocate sump pits to optimize use, but discharge location changes
should be coordinated with inspectors.

A perforated, vertical standpipe  should be wrapped with 1/2-inch hardware cloth and geotextiles
and then placed in the center of an excavated pit, which is then backfilled with filter material
ranging from clean gravel to stone. Water is then pumped from the center of the standpipe to a
suitable discharge area such as into a sediment trap, sediment basin, or stabilized area.

A sump pit should conform to the following specifications:

       •  Pit dimensions are variable, with the minimum  diameter being twice the diameter of
          the standpipe.

       •  The standpipe should be constructed by perforating a 12- to 36-inch diameter pipe,
          then wrapping it with 1/2-inch hardware cloth and geotextiles. The perforations
          should be 1/2-inch slits or 1-inch diameter holes placed 6 inches on center.

       •  The standpipe should extend 12 to 18 inches above the lip of the pit or riser crest
          elevation (basin dewatering), and filter material should extend 3 inches minimum
          above the anticipated standing water level.

Effectiveness
No information is available on the effectiveness of the sump pit.

Limitations
The sump pit must be properly maintained and pumped regularly to avoid clogging.
                                          7-73

-------
                                                             Section 7: Technology Assessment
Maintenance
To maintain performance, sump pits must be removed and reconstructed when water can no
longer be pumped out of the standpipe.

Cost
R.S. Means (2000) provides information appropriate for assessing a wide range of dewatering
scenarios (i.e., different sump sizes, dewatering durations, and discharge conditions). In general,
installing earthen sump pits are listed as costing approximately $1.50 per cubic foot of sump
volume. Piping to and away from the sump ranges from $30 to $60 per linear foot. Pump rentals
and operation range between $150 and $500 per day of pumping, depending on the rate of
dewatering. All costs include materials, labor, and equipment,  with overhead and profit.

      7.2.4.4.   Sediment Tank

Description
A sediment tank is a compartmented container through which sediment-laden water is pumped to
trap and retain sediment before pumping the water to drainageways, adjoining properties, and
rights-of-way below the sediment tank site.

Applicability
A sediment tank should be used on sites where excavations are deep and space is limited, such as
urban construction, where direct discharge of sediment-laden water to streams and storm
drainage systems should be avoided.

Design and Installation Criteria
The location of sediment tanks  should facilitate easy cleanout and disposal of the trapped
sediment to minimize interference with construction activities  and pedestrian traffic. The tank
size should be determined according to the storage volume of the sediment tank, with 1 cubic
foot of storage for each gallon per minute of pump discharge capacity.

Effectiveness
No information is available on the effectiveness of sediment tanks.

Limitations
The sediment tank does not provide any natural infiltration; thus, the trapped sediment and
stormwater must be disposed of properly.

Maintenance
To facilitate maintenance of sediment tanks, they need to be located with easy access for regular
pump out. The rate at which a tank is pumped depends on site-specific considerations such as
rainfall and sediment loads to the system. Regular inspections will help to determine pump out
frequency and prevent overloading and failure of the system.

Cost
No information is available on the cost of sediment tanks.
                                          7-74

-------
                                                               Section 7: Technology Assessment
      7.2.4.5.   Stabilized Construction Entrance

Description
The purpose of stabilizing entrances to a construction site is to minimize the amount of sediment
leaving the area as mud attached to tires. Installing a pad of gravel over filter cloth where
construction traffic leaves a site can help stabilize a construction entrance. As a vehicle drives
over the gravel pad, mud and other sediments are removed from the vehicle's wheels (sometimes
by washing) and off-site transport of sediment is reduced. The gravel pad also reduces erosion
and rutting on the soil beneath the  stabilization structure. The fabric reduces the amount of
rutting caused by vehicle tires by spreading the vehicle's weight over a larger soil area than just
the tire width. The filter fabric also separates the gravel from the soil below, preventing the
gravel from being ground into the soil.

Applicability
Stabilized construction entrances typically are installed at locations where construction traffic
leaves or enters an existing paved road. However, the applicability of site entrance stabilization
should be extended to any roadway or entrance where vehicles will access or leave the site.

From a public relations point of view, stabilizing construction site entrances can be a worthwhile
exercise. If the site entrance is the most publicly noticeable part of a construction site, stabilized
entrances can improve the appearance to passersby and improve public perception of the
construction project by reducing the amount of mud tracked onto adjacent streets.

Design and Installation Considerations
Hydrologic Design
Not applicable.

Installation Criteria and Specifications
All entrances to a site should be stabilized before construction begins and further disturbance of
the site area occurs. The stabilized site entrances should be long enough and wide enough so that
the largest construction vehicle that will enter the site will fit in the entrance with room to spare.
If many vehicles are expected to use an entrance in a day, the site entrance should be wide
enough for the passage of two vehicles at the same time with room on either side of each vehicle.
For optimum effectiveness, a rock construction entrance should be at least 50 feet long and at
least 10 to 12 feet wide (USEPA 1992). If a  site entrance leads to a paved road, the  end of
entrance should be flared (made wider as in  the shape of a funnel) so that long vehicles do not go
off the stabilized area when turning onto or off of the paved roadway.

If a construction site entrance crosses a stream, swale, roadside channel, or other depression, a
bridge or culvert should be provided to prevent erosion from unprotected banks.

Stone and gravel used to stabilize the construction site entrance should be large enough so that
nothing is carried off-site with vehicle traffic. In addition, sharp-edged stone should be avoided
to reduce the possibility of puncturing vehicle tires. Stone or gravel should be installed at a depth
of at least 6 inches for the entire length and width of the stabilized construction entrance.
                                           7-75

-------
                                                               Section 7: Technology Assessment
Effectiveness
Stabilizing construction entrances to prevent sediment transport off-site is effective only if all
entrances to the site are stabilized and maintained. Also, stabilizing construction site entrances
might not be very effective unless a wash rack is installed and routinely used (Corish 1995),
although that can be problematic for sites with multiple entrances that have high vehicle traffic.

Limitations
Although stabilizing a construction entrance is a good way to help reduce the amount of
sediment leaving a site, some sediment can still be deposited from vehicle tires onto paved
surfaces. To further reduce the chance that sediments will pollute stormwater runoff, sweeping of
the paved area adjacent to the stabilized entrance is recommended.

For sites using wash stations, a reliable water source to wash vehicles before leaving the site
might not be initially available. In such a case, water might have to be trucked to the site at an
additional cost. Discharge from the wash station should be directed to an appropriate sediment
control structure.

Maintenance
Stabilization of site entrances should be maintained until the remainder of the construction site
has been fully stabilized. Stone and gravel might need to be periodically added to each stabilized
construction site entrance to maintain its effectiveness. Soil that is tracked off-site should be
swept up immediately and disposed of properly.

For sites with wash racks at each site entrance, sediment traps will have to be constructed and
maintained for the life of the project. Maintenance will entail the periodic removal of sediment
from the traps to ensure their continued effectiveness.

Cost
Without a wash rack,  construction site entrance stabilization costs range from $1,000 to $4,000.
On average, the initial construction cost is approximately $2,000 per entrance. When
maintenance costs are included, the average total annual cost for a 2-year period is
approximately $1,500. If a wash rack is included in the construction site entrance stabilization,
the initial construction costs range from $1,000 to $5,000, with an average initial cost of $3,000
per entrance. Total annual cost, including maintenance for an estimated 2-year life span, is
approximately $2,200 per year (USEPA 1993).

      7.2.4.6.   Land Grading

Description
Land grading involves reshaping the ground surface to planned grades as determined by an
engineering survey, evaluation, and layout. Land grading provides more suitable topography  for
buildings, facilities, and other land uses  and helps to control  surface runoff, soil erosion, and
sedimentation both during and after construction.
                                           7-76

-------
                                                               Section 7: Technology Assessment
Applicability
Land grading is applicable to sites with steep topography or easily erodible soils because it
stabilizes slopes and decreases runoff velocity. Grading activities should maintain existing
drainage patterns as much as possible.

Design and Installation Criteria
Before grading activities begin, decisions should be made regarding the steepness of cut-and-fill
slopes and how the slopes will be protected from runoff, stabilized, and maintained. A grading
plan should be prepared that establishes which areas of the site will be graded, how drainage
patterns will be directed, and how runoff velocities will affect receiving waters. The grading plan
also includes information regarding when earthwork will start and stop, establishes the degree
and length of finished slopes, and dictates where and how excess material will be disposed of (or
where borrow materials will be obtained if needed). Berms, diversions, and other stormwater
practices that require excavation and filling should also be incorporated into the grading plan.

One low-impact development technique that can be incorporated into a grading plan is site
fingerprinting. This involves clearing and grading only those areas necessary for building
activities and equipment traffic. Adhering to strict limits of clearing and grading helps to
maintain undisturbed temporary or permanent buffer zones in the grading operation and provides
a low-cost sediment control measure that will help reduce runoff and off-site sedimentation. The
lowest elevation of the site should remain undisturbed to provide a protected stormwater outlet
before storm drains or other construction outlets are installed.

Effectiveness
Land grading is an effective means of reducing steep slopes and stabilizing highly erodible soils
when implemented with stormwater management and ESC practices in mind. Land grading is not
effective when drainage patterns are altered or when vegetated areas on the perimeter  of the site
are destroyed.

Limitations
Construction sites are routinely graded to prepare a site for buildings and other structures.
Improper grading practices that disrupt natural stormwater patterns can lead to poor drainage,
high runoff velocities, and increased peak flows during storm events.  Clearing and grading of the
entire site without vegetated buffers promotes off-site transport of sediments and other
pollutants. Grading plans should be designed with ESC and stormwater management goals in
mind; grading crews should be carefully supervised to ensure that  the plan is implemented as
intended.

Maintenance
All graded areas and supporting ESC practices should be periodically checked, especially after
heavy rainfalls. All sediment should be promptly removed from diversions or other stormwater
conveyances. If washouts or breaks occur, they should be repaired immediately.  Prompt
maintenance of small-scale, eroded areas is essential to prevent them from becoming significant
gullies.
                                           7-77

-------
                                                               Section 7: Technology Assessment
Cost
Land grading is practiced at virtually all construction sites—additional site planning to
incorporate stormwater and ESCs in grading plans can require several hours of planning by a
certified engineer or landscape architect. Extra time might be required to excavate diversions and
construct berms, and fill materials might be needed to build up low-lying areas or fill
depressions.

Where grading is performed to manage on-site stormwater, R.S. Means (2000) suggests the cost
of fine grading, soil treatment, and grassing to be approximately $2 per square yard of earth
surface area. Shallow excavation/trenching (1  to 4 feet deep) with a backhoe in areas not
requiring dewatering can be performed for $4  to $5 per cubic yard of removed material. Larger
scale grading requires a site-specific assessment of an alternative grading apparatus and a
detailed fill/excavation material balance to retain as much soil on site as possible.

      7.2.4.7.   Temporary Access Waterway  Crossing

Description
A temporary stream crossing is a structure erected to provide a safe and  stable way for
construction vehicle traffic to cross a running watercourse. The primary  purpose  of such a
structure is to provide streambank stabilization, to reduce the risk of damaging the streambed or
channel, and to reduce the risk of sediment loading from construction traffic. A temporary  stream
crossing could be a bridge, culvert, or ford.

Applicability
Temporary stream crossings are applicable wherever heavy construction equipment must be
moved from one side of a stream channel to the other or where lighter construction vehicles will
cross the stream  a number of times during the  construction period. In either case, an  appropriate
method for ensuring the stability of the streambanks and preventing large-scale erosion is
necessary.

A bridge or culvert is the best choice for most temporary stream crossings. If properly designed,
each can support heavy loads, and materials used to construct most bridges and culverts can be
salvaged after they are removed. Fords are appropriate in steep areas subject to flash flooding,
where normal flow is shallow or intermittent across a wide channel. Fords should be used only
where stream crossings are expected to be infrequent.

Design and Installation Criteria
Because of the potential for stream degradation, flooding, and safety hazards,  stream crossings
should be  avoided on a construction site whenever possible.  Consideration should be given to
alternative site access routes before arrangements are made to erect a temporary stream crossing.
If it is determined that a stream crossing is necessary, an area where the  potential for erosion is
low should be selected. The stream crossing structure should be installed during a dry period if
possible to reduce sediment transport into the stream.

If needed, over-stream bridges are generally the preferred temporary stream crossing structure.
The expected load and frequency of the stream crossing, however, will govern the selection of a
                                           7-78

-------
                                                              Section 7: Technology Assessment
bridge as the correct choice for a temporary stream crossing. Temporary bridges usually cause
minimal disturbance to a stream's banks and cause the least obstruction to stream flow and fish
migration. They should be constructed only under the supervision and approval of a qualified
engineer.

As general guidelines for constructing temporary bridges, clearing and excavation of the stream
shores and bed should be kept to a minimum. Sufficient clearance should be provided  for
floating objects to pass under the bridge. Abutments should be parallel to the stream and be
placed on stable banks. If the stream is less than 8 feet wide at the point where a crossing is
needed, no additional in-stream supports should be used. If the crossing is to extend across a
channel wider than 8 feet (as measured from the top of one bank to the other), the bridge should
be designed with one in-water support for each 8 feet of stream width.

A temporary bridge should be anchored by steel cable or chain on one side only to a stable
structure on shore. Examples of anchoring structures include trees with a large diameter, large
boulders, and steel anchors. By anchoring the bridge on one side only, there is a decreased risk of
causing a downstream blockage or flow diversion if a bridge is washed out.

When constructing a culvert, filter cloth should be used to  cover the streambed and streambanks
to reduce settlement and improve the stability of the culvert structure. The filter cloth should
extend a minimum of 6 inches and a maximum of 1 foot beyond the end of the culvert and
bedding material. The culvert piping should not exceed 40 feet in length and should be of
sufficient diameter to allow for complete passage of flow during peak flow periods. The culvert
pipes should be covered with a minimum of 1 foot of aggregate. If multiple culverts are used, at
least 1 foot of aggregate should separate the pipes.

Fords should be constructed of stabilizing material such as large rocks.

Effectiveness
Both temporary bridges and culverts provide an adequate path for construction traffic crossing a
stream or watercourse.

Limitations
Bridges can be considered the greatest safety hazard of all  temporary stream crossing structures
if not properly designed and constructed. Bridges can also  prove to be more costly in terms of
repair costs and lost construction time if they wash out or collapse (Smolen et al. 1988).

The construction and removal of culverts are usually very disturbing to the surrounding area, and
erosion and downstream movement of sediments are often great. Culverts can also create
obstructions to flow in a stream and inhibit fish migration.  Depending on their size, culverts can
be blocked by large debris and are therefore vulnerable to frequent blockage and washout.

If given a choice between building a bridge or a culvert as  a temporary stream  crossing, a bridge
is preferred because of the relative minimal disturbance to  streambanks and the opportunity for
unimpeded flow through the channel. The approaches to fords often have high erosion potential.
In addition, excavating the streambed and approach to lay riprap or other stabilization  material
                                          7-79

-------
                                                               Section 7: Technology Assessment
causes major stream disturbance. Mud and other debris are transported directly into the stream
unless the crossing is used only during periods of low flow.

Maintenance
Temporary stream crossings should be inspected at least once a week and after all significant
rainfall events. If any structural damage is reported to a bridge or culvert, construction traffic
should be excluded  until appropriate repairs are made. Streambank erosion should be repaired
immediately.

Fords should be inspected closely after major storm events to ensure that stabilization materials
remain in place. If the material has moved downstream during periods of peak flow, the lost
material should be replaced immediately.

Cost
In general, temporary bridges are more expensive to design and construct than culverts. Bridges
are also associated with higher maintenance and repair costs should they fail. Temporary
bridging costs vary  as a function of the width of the bridge span and the amount of time the
bridge is installed. If the bridging is permanent, a mean cost of $50 per square foot for an 8-foot
wide steel arch bridge (no foundation costs included) can be used for conceptual cost estimation
(R.S. Means 2000).  If rental bridging is employed, rates are probably on the order of 20 to 50
percent of the bridge (permanent)  cost but will vary according to the rental duration and
mobilization distance.

      7.2.4.8.    Dust Control

General Description
Dust control measures are practices that help reduce ground  surface and air movement of dust
from disturbed soil  surfaces. Construction sites are good candidates for dust control measures
because land disturbance from clearing and excavation generates a large amount of soil
disturbance and open space for wind to pick up dust particles. To illustrate this point, research at
construction sites has established an average dust emission rate of 1.2 tons/acre/month for active
construction (WDEC 1992). These airborne particles pose a  dual threat to the environment and
human health. First, dust can be carried off-site, thereby increasing soil loss from the
construction area and increasing the likelihood of sedimentation and water pollution.  Second,
blowing dust particles can contribute to respiratory health problems and create an inhospitable
work environment.

Applicability
Dust control measures are applicable to any construction site where dust is created and there is
the potential for air  and water pollution from dust traveling across the landscape or through the
air. Dust control  measures are particularly important in arid or semiarid regions where soil can
become extremely dry and vulnerable to transport by high winds.

Also,  dust control measures should be implemented on all construction sites where there will be
major soil disturbances or heavy construction activity, such as clearing, excavation, demolition,
or excessive vehicle traffic. Earthmoving activities are the major source of dust from
                                           7-80

-------
                                                               Section 7: Technology Assessment
construction sites, but traffic and general disturbances can also be major contributors (WDEC
1992).

The specific dust control measures implemented at a site will depend on the topography, land
cover, soil characteristics, and amount of rainfall at the site.

Design and Installation Criteria
When designing a dust control plan for a site, the amount of soil exposed will dictate the quantity
of dust generation and transport. Therefore, construction sequencing and disturbing small areas
at one time can greatly reduce problematic dust from a site. If land must be disturbed, additional
temporary stabilization measures should be considered before disturbance.

A number of methods can be used to control dust from a site. The following is a brief list of
control measures and their design criteria. Not all control measures will be applicable to a site.
The owner, operator, and contractors responsible for dust control should determine which
practices accommodate their needs on the basis of specific site and weather conditions.

Sprinkling/Irrigation: Sprinkling the ground surface with water until it is moist is an effective
dust control method for haul roads and other traffic routes (Smolen et al. 1988). This practice can
be applied to almost any site.

Vegetative Cover: In areas not expected to handle vehicle traffic, vegetative stabilization of
disturbed soil is often desirable. Vegetative cover provides protection to surface soils and slows
wind velocity at the ground surface, thus reducing the potential for dust to become airborne.

Mulch: Mulching can be a quick and effective means of dust control for a recently disturbed
area (Smolen et al. 1988).

Wind Breaks:  Wind breaks are barriers (either natural or constructed) that reduce wind velocity
and therefore reduce the possibility of carrying suspended particles. Wind breaks can be trees or
shrubs left in place during site clearing or constructed barriers such as a wind fence,  snow fence,
tarp curtain, hay bale, crate wall, or sediment wall (USEPA 1992).

Tillage: Deep tillage in large open areas brings soil clods to the surface where they rest on top of
dust, preventing it from becoming airborne.

Stone: Stone can be an effective dust deterrent for construction roads and entrances.

Spray-on Chemical Soil Treatments (palliatives): Examples of chemical adhesives include
anionic asphalt emulsion, latex emulsion, resin-water emulsions, and calcium chloride. Chemical
palliatives should be used only on mineral soils.  When considering chemical application to
suppress dust, consideration should be taken as to whether the chemical is biodegradable or
water-soluble and what effect its application could have on the surrounding environment,
including waterbodies and wildlife.
                                           7-81

-------
                                                               Section 7: Technology Assessment
Table 7-22 shows application rates for some common spray-on adhesives as recommended by
Smolenetal. (1988).

                     Table 7-22. Application rates for spray-on adhesives
Spray on adhesive
Anionic Asphalt Emulsion
Latex Emulsion
Resin in Water
Water dilution
7:1
12.5:1
4:1
Type of nozzle
Coarse spray
Fine spray
Fine spray
Application
(gal/acre)
1,200
235
300
    Source: Smolen et al. 1988.

Effectiveness
Sprinkling/Irrigation: Not available.

Vegetative Cover: Not available.

Mulch: Mulch can reduce wind erosion by up to 80 percent.

Wind Breaks/Barriers: For each foot of vertical height, an 8- to 10-foot deposition zone develops
on the leeward side of the barrier. The barrier density and spacing will change its effectiveness at
capturing windborne sediment.

Tillage: Roughening the soil can reduce soil losses by approximately 80 percent.

Stone: The sizes of the stone can affect the amount of erosion that will take place. In areas of
high wind, small stones are not as effective as 20-cm stones.

Spray-on Chemical Soil Treatments (palliatives): Effectiveness of polymer stabilization methods
ranges from 70 to 90 percent.

Limitations
In areas where evaporation rates are high, water application to  exposed soils could require near
constant attention. If water is applied in excess, runoff can result from the site and possibly
create conditions where vehicles can track mud onto public roads.

Chemical applications should be used sparingly and only on mineral soils (not high organic
content soils) because their misuse can create additional surface water pollution from runoff or
can contaminate ground water if infiltrated. Chemical applications can also present a health risk
if excessive amounts are used.

Maintenance
Because dust controls are dependent on specific site conditions including the weather, inspection
and maintenance are unique for each site. Generally, however,  dust control measures involving
application of either water or chemicals require more monitoring than structural or vegetative
controls to remain effective. If structural controls are used, they should be inspected for
deterioration regularly to ensure that they are still achieving their intended purpose.
                                           7-82

-------
                                                               Section 7: Technology Assessment
Cost
Chemical dust control measures can vary widely in cost depending on specific needs of the site
and level of dust control desired. One manufacturer of a chloride product estimates a cost of
$1,089 per acre for application to road surfaces but cautioned that cost estimates without a
specific site evaluation can be inaccurate.

      7.2.4.9.   Storm Drain Inlet Protection

Description
Storm drain inlet protection measures are controls that help prevent soil and debris from on-site
erosion from  entering storm drain inlets. Typically, such measures are temporary controls that
are implemented before large-scale disturbance of the surrounding site. The controls are
advantageous because their implementation allows storm drains to be used during even the early
stages of construction activities. The early use of storm drains during project development
significantly reduces the occurrence of future erosion problems (Smolen et al.  1988).

Three temporary control measures to protect storm drain drop inlets are as follows:

       •  Excavation around the perimeter of the drop inlet

       •  Fabric barriers around inlet entrances
       •  Block and gravel protection

Excavation around a storm drain inlet creates a settling pool to remove sediments. Weep holes
protected by gravel are used to drain the shallow pool of water that accumulates around the inlet.
A filter fabric barrier erected around an inlet can create an effective shield to sediment while
allowing water to flow into the storm drain. This type of barrier can slow runoff velocity while
catching soil  and other debris at the  drain inlet. Block and gravel inlet protection uses standard
concrete blocks and gravel to form a barrier to sediments while permitting water runoff through
select blocks  that are laid sideways.  In addition to these materials, limited temporary stormwater
drop inlet protection can also  be achieved using straw bales or sandbags to create barriers to
sediment.

For permanent storm drain drop inlet protection after the surrounding area has been stabilized,
sod can be installed as a barrier to slow stormwater entry to the inlets and capture sediments
from erosion. This final inlet protection measure can be used as an aesthetically pleasing way to
slow stormwater velocity near drop inlet entrances and remove sediments and other pollutants
from runoff.

A new technology that uses an insert trap into the  inlet itself has been developed (Adams et al.
2000). The technique shows good results on initial tests, trapping more than 50 percent of the
incoming sediment in flows typical of those into urban storm drains. The technique is being
further developed with a pending patent application.

Applicability
All temporary controls should have a drainage area no greater than 1 acre of drainage area per
inlet.  It is also important for temporary controls to be constructed before disturbing the
                                           7-83

-------
                                                               Section 7: Technology Assessment
surrounding landscape. Excavated drop inlet protection and block and gravel inlet protection are
applicable to areas of high flow where overflow is anticipated into the storm drain. Fabric
barriers are recommended for smaller, relatively flat drainage areas (slopes less than 5 percent
leading to the storm drain).

Temporary drop inlet control measures are often used in combination with each other and with
other stormwater control techniques.

Design and Installation Considerations
Hydrologic Design
Hydrologic computations are not necessary with present technologies. A specified limitation of
one drainage acre per inlet limits flow rates, depending on local rainfall and runoff
considerations.

Installation Criteria and Specifications
The following criteria should be followed until future research establishes better techniques:

       •  With the exception of sod drop inlet protection, these controls should be installed
          before any soil disturbance in the drainage area.

       •  Excavation around drop inlets should be dug a minimum of 1 foot deep (2 feet
          maximum) with a minimum excavated volume of 35 cubic yards per acre disturbed.
          Side slopes leading to the  inlet should be no steeper than 2:1. The shape of the
          excavated area should be designed such that the dimensions fit the area from which
          stormwater is anticipated to drain. For example, the longest side of an excavated area
          should be along the side of the inlet expected to drain the largest area.

       •  Fabric inlet protection is essentially a filter fence placed around the inlet. The fabric
          should not be used as a standalone sediment control measures. To increase inlet
          protection effectiveness, these practices should be used in combination with other
          measures, such as small impoundments or sediment traps (USEPA 1992).  Temporary
          storm drain inlet protection is not intended for use in drainage areas larger than 1
          acre.  Generally, stormwater inlet protection measures are practical for relatively low
          sediment and low volume flows.

       •  Frequent maintenance of storm drain controls is necessary to prevent clogging. If
          sediment and other debris clog the water intake, drop intake control measures can
          actually cause erosion in unprotected areas.

Maintenance
All temporary control measures must be checked after each storm event. To maintain the
sediment capacity of the shallow settling pools created from these techniques, accumulated
sediment should be removed from the area around the drop inlet (i.e., from the excavated area,
around the fabric barrier, or around the block structure) when the sediment storage is reduced by
approximately 50 percent. Additional debris should be removed from the shallow pools
periodically.
                                           7-84

-------
                                                              Section 7: Technology Assessment
Weep holes in excavated areas around inlets can become clogged and prevent water from
draining from the shallow pools that form. Should this happen, unclogging the water intake can
be difficult and costly.

Cost
The cost of implementing storm drain drop inlet protection measures will vary depending on the
control measure chosen. Generally, initial installation costs range from $50 to $150 per inlet,
with an average cost of $100 (USEPA 1993). Maintenance costs can be high (annually, up to 100
percent of the initial construction cost) because of frequent inspection and repair needs. The
SWRPC has estimated that the cost of installing inlet protection devices ranges from $106 to
$154 per inlet (SWRPC 1991).

      7.2.4.10. Polyacrylamide (RAM)

General Description
The term polyacrylamide (PAM) is a generic term that refers to a broad class of compounds.
There are hundreds of specific PAM formulations, and all have unique properties that depend on
polymer chain length and number and kinds of functional group substitutions along the chain.
PAMs are classified according to their molecular weight and ionic charge and are available in
solid, granular, liquid, or emulsion forms.

The effectiveness of PAMs to prevent or reduce erosion is due to its affinity for soil particles,
largely via coulombic and Van der Waals attraction. Such surface attractions enhance particle
cohesion, stabilizing soil structure against shear-induced detachment and transport in runoff. In a
soil application, PAM aggregates soil particles, increasing pore space and infiltration capacity
and resulting in reduced runoff. The larger particle aggregates are less susceptible to raindrop
and scour erosion, thus reducing the potential to mobilize sediments.

Applicability
Because of ease in application, PAM is well suited as a short-term erosion prevention BMP,
especially for areas with limited access or steep slopes that hinder personnel from applying other
cover materials. PAM can be used to  augment other cover practice BMPs, though it can be
effective when applied alone.  Thus, the ease of application, low maintenance, and relatively low
cost associated with PAM make it a practical solution to soil stabilization during construction.

Application Criteria
PAM can be applied to soil through either a dry granular powder or a liquid spray form. Optimal
application rates to prevent erosion on construction sites are generally less than 1 kg/ha
(approximately 1 Ib/ac) (Tobiason et al. 2000). However, the concentration required can vary for
specific soil properties and construction phases. WDOT (2002) suggests a dosage of 60 mg/L for
roadway ESC. This is higher than the rate recommended by the University  of Nebraska for an
agricultural application (10 parts per million). To put this into context, one-half pound of PAM
in 1,000 gallons of water results in a PAM concentration of 60 mg/L, which treats  1 acre of
exposed soil, according to WDOT recommendations.
                                           7-85

-------
                                                              Section 7: Technology Assessment
Effectiveness
A study performed in Dane County, Wisconsin, analyzed 15-meter-square plots for runoff and
sediment yield on a construction site. The study concludes that when a solution of PAM-mix
with mulch/seeding is applied to dry soil and compared with the control (no PAM-mix
application to dry soil), the PAM-mix has an average reduction of 93 percent in sediment yield.
The lowest performance (average reduction in sediment yield of 77 percent) occurred when
PAM-mix in solution was applied to moist soil. The application of dry PAM-mix to dry soil
reduced sediment by 83 percent and decreased runoff by 16 percent when compared to the
control. The results show that regardless of the application method, PAM-mix is effective in
reducing sediment yield in the test plots (Roa-Espinosa et al. 2000).

A second study performed in Washington analyzed the runoff from three different construction
sites: an erosion control test facility, a highway construction site, and an airport runway. Table
7-23 summarizes the 225 samples analyzed by Tobiason et al. (2000).

                       Table 7-23. Turbidity reduction values from PAM

Maximum
Median
Minimum
Volume
(m3)
350
285
133
Turbidity reduction
(%)
99.97%
97.6%
46%
Limitations
PAMs are most commonly produced as dry granules. They completely dissolve and remain
dissolved if mixed properly. If added too quickly or if not stirred vigorously, the granules rapidly
form nondissolvable gels on contact with water or collect in low turbulence areas as syrupy
concentrations that dissolve slowly in an uncontrolled pattern over a period of hours or days
(Sojka and Lentz 1994). In addition, when spilled on hard surfaces, PAM solutions are extremely
slippery and hazardous to foot and vehicle traffic. PAM dust is highly hygroscopic and, if
inhaled, could impair breathing. Certain neutral and cationic PAMs at very high exposure levels
produce irritation in humans and are somewhat toxic to certain aquatic organisms; therefore,
PAM should be used in strict compliance with state and federal label requirements.

Cost
The cost of PAM ranges from $1.25 per pound to $5.00 per pound (Entry and Sojka 1999). The
cost of PAM application depends on the system employed. If dispersed through irrigation
systems (for agriculture), the seasonal cost of PAM treatment is $9 to $15 per acre
(Kay-Shoemake et al. 2000), where a season probably requires between 5 and 10 applications.
For construction sites, it is more likely that PAM would be applied as an additive to the
hydroseed mix and applied when final grade is established and cover vegetation is installed.
Numerous suppliers provide PAM as a low-cost additive for hydroseeding, suggesting PAM
application costs can be incorporated into that of hydroseeding ($540 to $700 per acre depending
on which seed is applied). An additional cost would be incurred to sample site soils to customize
the dosage and delivery mechanisms for individual sites. In addition, reapplication of PAM in
granular or liquid form to areas with rill development (poor vegetation cover) would require
                                          7-86

-------
                                                             Section 7: Technology Assessment
additional funds. Where reapplication of granular PAM is used, R.S. Means (2000) suggests a
cost of approximately $5 per 1,000 square feet for spreading soil admixtures by hand.

    7.2.5.   ADVANCED TREATMENT AND CONTROL TECHNOLOGIES

      7.2.5.1.   Active Treatment Systems (ATS) Technologies

EPA researched technologies available for treating construction stormwater runoff, with the
specific goal of identifying technologies that could reliably meet a low-effluent turbidity limit.
EPA primarily identified active treatment systems (ATS) that use coagulation/flocculation and
filtration for treating stormwater runoff from active construction sites as the most reliable
technology for meeting a low (i.e., less than 10 NTUs) effluent limit. Technologies used at
construction sites to control suspended sediment and turbidity in stormwater runoff from
discharging typically include erosion control, storage/containment, gravitational settling,
chemical treatment (i.e., coagulation/flocculation), and filter media. For an ATS to be effective,
many (if not all)  of the abovementioned treatment technologies need to be incorporated into an
ATS before treated effluent discharge. This section provides an ATS process description and
costs and discusses applicability, demonstration status, and limitations. Treatment chemical
addition and filtration are separately discussed in detail in subsequent sections.

ATS Process Description and Costs

EPA assumed that the key components of an ATS  would include the following:

       •  On-site storage by using a combination of sediment basins, tanks or other
          impoundments
       •  Chemical addition (see Section 7.2.5.2 )

       •  Mix tank/clarification tank and/or basin

       •  Media filtration (see Section 7.2.5.3)

       •  Instrumentation (e.g., monitoring of influent and effluent)

The ATS capital  costs include purchased (or leased) equipment cost, including ancillary
equipment (e.g.,  piping, valves, and controllers), delivery cost, and installation/construction cost
(including labor and site work). The ATS annual (operation and maintenance) costs include
treatment chemicals, operating labor  and material, maintenance labor and material, energy, waste
disposal, monitoring, and rented equipment.

The ATS are typically equipped with automated instrumentation to monitor water quality, flow
rate, and dosage  control for both influent and effluent flows. Following the
coagulation/flocculation process, the densified floe is settled out via gravitational settling,
skimming, or media filtration (e.g., sand, gravel, bag filters).

EPA determined that some vendors offer gel socks containing treatment chemical, often for
pretreatment of stormwater runoff. For example, the StormKlear Gel-Floe™ is a fabric sock
containing a flake form of chitosan that slowly dissolves as the influent stormwater flows over it.
The gel sock is typically anchored within the influent pipe of the ATS.
                                          7-87

-------
                                                                 Section 7: Technology Assessment
An ATS can be in either a batch or flow-through design, as described in Table 7-24. The ATS
design depends on factors including existing structures (e.g., detention basins, storm sewer
systems, sump areas), influent turbidity, flow rate, and space limitations. Clear Creek Systems,
Inc., in a comment letter to the California State Water Resources Control Board (Gannon 2007)
regarding the draft construction general permit, stated that "Batch treatment is a relatively
outdated and inefficient method of operations."

                               Table 7-24. ATS operating modes
Operation mode
Batch (Pump-Treat-Hold-Test
Release)
Flow-through or continuous
treatment
Description
Stormwater runoff is collected, stored or contained in a basin or tank until
treatment is complete before discharging.
Involves pumping stormwater runoff from a collection, storage or
containment basin, treating the water, and directly discharging.
Source: ATS Industry Task Force 2007.


Figure 7-1 presents a general ATS batch operating mode process diagram. The batch treatment
process incorporates a period for treatment in a settling, mixing, or holding tank(s) before
discharge. This is different from the continuous flow or flow-through treatment process in which
treatment and discharge occurs continuously. Figure 7-2 shows an ATS using continuous mode.
                                                        Shut-off valve
  I  STORM WATER RUNOFF
    COLLECTION/SETTLING
      [Sediment Basin (if
     available), Sump Area,
        Piping, etc.]
Instrumentation (e.g.,
\ pH, turbidity, flow
\ meter, etc.)
V
Pnmn I f*) 1 ' •""*

1
/ Chemical
/ coagularit/flocculant

Settling, Mixing and
Holding Tank (multiple
tanks in series or parallel if
applicable)
Rhn
1
-^
                                                                              RECEIVING WATERS
                             metering pump
Shown without  Instrumentation (e.g., /
optional media   pn, turbidity, etc.)
  filtration
Figure 7-1. General ATS batch-operating mode
                                             7-88

-------
                                                                Section 7: Technology Assessment
 SITE STORM WATER RUNOFF
                                    Backwash Return
STORM WATER RUNOFF
DETENTION STRUCTURE
                          nstrumentation (e.g
                           pH, turbidity, flow
                             meter, etc.)
       STORM WATER
    DETENTION STRUCTURE  Pump-4-4
                                         Effluent Diversion
                                                               Shut-off valve
                                              Media Filtration
                                                (e.g., bag,
                                              cartridge, sand
                                                 filters)
                                                                      Effluent Line f
                                                                      RECEIVING WATERS
                                                      Instrumentation (e.g.,
                                                        pH, turbiditv, flow
                                                         meter, etc.)
                               Chemical
                            coagulant/floccularrt
                             metering pump
Figure 7-2. Flow-through ATS operating mode
ATS Applicability

The ATS is well suited as a method of runoff control when traditional BMPs are not capable of
meeting numeric standards. ATS provides quick and efficient removal of fine-grained,
suspended sediment or colloidal particles and can be custom tailored for site-specific
requirements. Gravitational settling of fine or colloidal soil particles can have limited
effectiveness and might not be completed in a timely manner. Therefore, ATS could be
necessary to enhance  small particulate solids removal and minimize project timelines and costs.
The ATS can minimize potential adverse environmental effects on receiving water through
automated water quality measurements. ATS generally produces very low turbidity values (often
< 10 NTUs) in the effluent discharge.

The vendors contacted by EPA stated that ATS were typically meeting discharge standards of
10 NTU  or less. Therefore, ATS would work well for a low-NTU standard. Vendors reported
little cost savings in designing an ATS for higher, less stringent, NTU limits (e.g., 50, 100,  150,
200, 250 NTU).

Demonstration Status

EPA determined from information obtained from vendor calls that ATS using chemical treatment
with polymer coagulation/flocculation is prevalent in the industry. The majority of the vendors
contacted are using the polymer chitosan in conjunction with gravitational settling and filtration
for treating stormwater runoff. EPA did not obtain information on how many of the systems are
                                            7-89

-------
                                                                Section 7: Technology Assessment
batch or flow-through treatment systems. The polymer Diallydimethyl-ammonium chloride
(DADMAC) is also used frequently by vendors. Following the chemical treatment, media
filtration is commonly used. Sand filters in combination with small-micron (e.g., 0.5) particulate
filters appeared to be the media of choice by many of the vendors for removing the floe material
and polishing. However, bag and cartridge filters are also being used either as a standalone
treatment or in combination with the sand filters for treatment purposes. Many of the treatment
technologies used are very site-specific according to the water quality (i.e., turbidity, chemical
composition) and footprint available.

The vendors  contacted have implemented ATS primarily in the west (California, Oregon, and
Washington). Washington State's Department of Ecology (WDEC) has a new technology
evaluation program in which vendors complete a Chemical Technology Assessment Protocol—
Ecology (CTAPE) for new and emerging technologies. Following a performance evaluation,
vendors may receive a conditional use designation (CUD) or a general use level designation
(GULD) for a particular chemical treatment technology. For construction sites, WDEC has
approved conditional use or general use designations for Chitosan-enhanced sand filtration using
StormKlear™ Liquifloc™, FlocClear™, and Chitovan™ chemical treatments. Table 7-25 shows
known or draft state ATS requirements or recommendations at the time of this writing.
                     Table 7-25. ATS state requirements/recommendations
 State
ATS requirements and/or recommendations
 California
The 2009-0009-DWQ Construction General Permit, effective July 1, 2010, provides specific
requirements for dischargers who choose to use an ATS, including numeric effluent limits
(NELs) and design to capture and treat a volume equivalent to the runoff from a 10-year,
24-hour storm event in a 72-hour period with a runoff coefficient of 1.0.
The NELs for discharges from an ATS include the following:
o   Turbidity less than 10 NTU (for daily flow-weighted average) and 20 NTU (for any single
    sample)
o   Residual chemical must be less than 10% of the  maximum allowable threshold
    concentration for the most sensitive species of the chemical used
Complete ATS requirements are presented in Attachment F to the permit (see DCN 43115).
 Oregon
Mr. Dennis Jurries with the Oregon Department of Environmental Quality (DEQ) stated that
ATS is not required; however, for sites with difficulty reaching water quality standards, it is
recommended that a chitosan-enhanced sand filtration treatment be implemented.
Water Quality Requirements
If discharging to a 303(d) listed waterbody or a waterbody with a TMDL for sediment and
turbidity, sampling for turbidity is required to meet a 160-NTU benchmark. If unable to meet
benchmark, an Action  Plan using a BMP such as water treatment using electro-coagulation,
chemical flocculation or filtration must be implemented.
                                            7-90

-------
                                                                Section 7: Technology Assessment
 State
ATS requirements and/or recommendations
 Washington
BMP C250 Construction Storm Water Chemical Treatment states that formal, written
approval from the Department of Ecology is required for using chemical treatment regardless
of site size. Through the use of the Washington CTAPE, new technology evaluation program,
the following have been accepted with use designations:
o   Construction Site Treatment Technologies
o   Chitosan-Enhanced Sand Filtration Using StormKlear™ LiquiFloc™ (GULD)
o   Chitosan-Enhanced Sand Filtration Using FlocClear™(GULD and CUD)
o   Chitosan Enhanced Sand Filtration Using ChitoVan™ (CULD and GULD)
o   Water Tectonics Electrocoagulation Subtractive Technology (CUD)
o   GULD—General Use Level Designation
o   CUD—Conditional Use Designation

Water Quality Requirements
Turbidity shall be no more than 5 NTU over background (if background  is 50 NTU or less), or
no more than 10% over background (if background is 50 NTU or greater). Sites that disturb
more than 1 acre are required to sample for turbidity. Turbidity exceeding the benchmark of
25 NTU but less than 250 NTU requires BMP and SWPPP review, and  additional treatment if
three consecutive days exceed the benchmark. Turbidity exceeding 250 NTU requires
notifying the Department of Ecology and additional treatment.
ATS Limitations

Treatment chemicals must have the proper dose and contact time to avoid potential toxicity in
effluent discharges. Many of the polymers used in ATS precipitate only in a designated pH range
(e.g., 6.5 to 8.5).

ATS Costs

EPA obtained ATS costs (i.e., chitosan-enhanced sand filtration) from three vendors. The ATS
costs associated with treating active construction site stormwater runoff, as provided by vendors,
is included in Table 7-26. For estimating compliance costs for the regulatory options that
incorporate ATS, EPA determined system flowrate for the various model projects and estimated
equipment rental costs, operating costs and other ancillary costs using the data supplied by Rain
for Rent. EPA also added costs for providing storage for impounding runoff from the 2-year/24-
hour storm event as an approximation of the additional costs for storage that might be required
on-site. Additional details of this analysis are in the ATS Cost Spreadsheet Model (DCN 43119).
The  costs associated with ATS vary by site-specific factors such as turbidity, available footprint
area for basins or equipment, effluent discharge requirements, and the like. In addition to the data
provided by vendors, EPA obtained cost data on specific projects incorporating ATS from
several reports. Table 7-27 summarizes ATS costs for a number  of case studies reflecting
varying site sizes and different system configurations.

For additional details on ATS and on ATS costing, see DCNs 43000 through 43011, DCNs
41130 and 41131,  and Section 9.
                                            7-91

-------
                                                              Section 7: Technology Assessment
                           Table 7-26. Summary of vendor costs
Cost type
Total Cost
Total Cost
Labor
Chemical
Equipment Rental
(18 month rental)
Media Filter
System
2 Pumps
2 Tanks
4 Hoses
2 Elbows
Generator
Sand and Gravel
Mobilization/Demobilization
System Calibration
Pipes, valves, and electrical
Install pipes, valves, and
electrical
Misc. lab equipment and supplies
Fuel
Large Site
Small Site
Large/Small
Large Site
Small Site
All Sites
1 0-acre Site
50-acre Site
1 00-acre Site
500-gpm system.
1 6 acres @ 24
inches of rainfall
during 6-month
project period
(10,248,192
gallons).
Cost
(2008 dollars)
$0.005/gal
$0.01/gal
$0.01-0.03/gal
$1 ,250/Mgal
$5,000/Mgal
$1 ,000-$8,000/Mgal
$130,000
$250,000
$500,000
$4,000/month
$4,000/month
$3,300/month
$2,520/month
$392/month
$72/month
$1 ,050/month
$1 ,222
$2,970
$1,300
$6,000
$7,500
$4,250
10 gallons/hour
Source
StormKlear
Clear Water
Clear Creek
Rain for Rent
                         Table 7-27. Summary of ATS case studies
Project
City of Redmond
City of Redmond
(2 sites)
Confidential
Builder Project #1
Confidential
Builder Project #2
Type of ATS*
CESF
Electrocoagulation
CESF
DADMAC
Approximate
treated
volume
1 million
gallons
6.2 million
gallons
~ 100 million
gallons
~ 15 million
gallons
Project
duration
(months)
2.5
Unknown
Unknown

Project size
(acres)
32
8 and 23
acres
>500
~ 300 acres
Cost per gallon
$10.22 per thousand
gallons
$5.83 and $8.00 per
thousand gallons
$16. 00 per thousand
gallons
$36.00 per thousand
gallons
*CESF = Chitosan-Enhanced Sand Filtration, DADMAC = Diallyldimethyl Ammonium Chloride

McLaughlin (2008) noted that less complex systems (e.g., introducing a polymer such as PAM
into a pump intake followed by a sediment basin) can reduce chemical cost and might not require
media filtration (compared to active chitosan systems). McLaughlin (2008) also noted that small
quantities of water could be treated with polymer introduced into the pump intake and pumped
through geotextile sediment bags.
                                          7-92

-------
                                                                   Section 7: Technology Assessment
      7.2.5.2.   ATS Coagulation/Flocculation

The effective design of an ATS relies heavily on an analysis of site conditions (e.g., land use,
soils, toxins, water chemistry, flowrate, receiving water chemistry). Coagulants and flocculants
function as the primary treatment process used in ATS. Treatment chemical addition to influent
stormwater runoff is to destabilize the suspended particles by various mechanisms, aggregating
into larger particles that are easier to remove through settling or filtering. Coagulation is the
reduction of the net electrical repulsive forces at particle surfaces by adding coagulating
chemicals, whereas flocculation is the agglomeration of the destabilized particles by chemical
joining and bridging.

The coagulants/flocculants are typically added to the influent via an injection pump in a metered
dose just upstream of the clarifier tank or basin. The treatment chemicals are allowed to mix to
maximize the formation of a dense floe. Proper dosing of the treatment chemicals is critical to
minimize toxicity, maximize system efficiency, and  ensure proper effluent water quality. The
optimum dose is very site-specific (e.g., varying with changing types of soils, flow rate) and
should be based on a series of jar tests.

Water treatment chemicals are predominately water  soluble and classified as cationic (positively
charged), anionic (negatively charged), nonionic (neutral), or amphoteric (changeable depending
on the pH of water). Table 7-28 lists common coagulants, regulatory status,  and available
residual tests. Several of these  common coagulants and toxicity information are discussed below.

                  Table 7-28. Examples of some commonly available coagulants
Coagulant
Description
Regulatory status
Approved dosage
(or dosage where
no toxic effects
are observed)
Residual test
available3
Method detection
limit of residual
Chitosan
Chitosan
acetate
based
cationic
biopolymer
Approved in
Washington
N/A
Presence/
absence
0.1 mg/L
presence/
absence
PAC
Poly-
aluminum
chloride
N/A
N/A
DADMAC
Diallyl-
di methyl-
ammonium
chloride
N/A
N/A
PAM
Poly-
acrylamide
Approved in
Florida, New
Hampshire
Florida has no limit
New Hampshire
has limit of one-
half of NOECb or
IC25C
PASS
Poly-
aluminum
chloride
Silica/sulfate
modified
N/A
N/A
Alum
Aluminum
sulfate
Approved
in Florida
No limit
Presence/absence and quantitative
< 0.5 mg/L presence/absence
0.5 mg/L quantitative
Source: ATS Industry Task Force 2007.
a. Residual tests can be presence/absence tests or quantitative tests. A presence/absence test verifies that a chemical is or is not
present at or above a method detection limit; it does not quantify (with a numerical value) how much is present above the method
detection limit. A quantitative test yields the concentration of the chemical at or above the method detection limit; it t typically yields a
concentration in mg/L.
b. NOEC: No Observed Effect Concentration. Highest concentration of effluent where the effect (e.g., reproduction) is not
significantly different from the control.
c. IC25: 25 Percent Inhibition Concentration. Concentration causing a 25 percent reduction in the effect.
N/A - Not Available
                                              7-93

-------
                                                               Section 7: Technology Assessment
Chitosan acetate
This polymer is widely used at active construction sites in ATS for stormwater runoff.
Specifically, Washington, Oregon, and California have had numerous projects using this
polymer. It is an approved, general-use-level designated polymer for treating construction site
runoff in Washington State. Chitosan is derived from chitin, the major component of crustacean
shells and is a cationic polyelecrolyte. It is a very plentiful  natural polymer with supply
stemming from shellfish wastes. Chitosan is able to coagulate/flocculate non-polar hydrocarbons
(e.g., oil), suspended sediment,  and to chelate heavy metals (Nichols 1997).

Table 7-29 presents information from  several  studies regarding toxicity  of chitosan acetate to
aquatic organisms, chemical hazard information, and filter pass-through results.
                          Table 7-29. Chitosan acetate study results
Vendor/source
MacPherson 2006a
(references Nautilus
Environmental, Redmond,
Washington 2004)
Bullock etal. 2000
ProTech GCS 2004
MacPherson 2006b
Blandford 2006
Ray 2001
Results
Toxicitv. Chitosan acetate (1 % solution) was reported to have an LC50 for
Daphnia pulex of 1 ,370 mg/L, 642 mg/L for fathead minnow, and 168 mg/L
for rainbow trout in clean water and 452 mg/L in 500-NTU water.
Toxicitv. The toxicity of chitosan has generally been considered to be
nontoxic; however, Chitosan when dissolved in acetic acid and added to a
culture system at 1 .0 part per million (ppm) to remove organic solids was
found to have acute toxicity to rainbow trout, related to gill lesions, and the
severity was dose dependent.
Toxicitv. ProTech GCS in conjunction with GE Betz conducted research on
the polymer chitosan (1 % solution). The test was conducted using > 1 ,000
NTU water from a Sacramento, CA, project site. Survival rates for daphnia
magna, rainbow trout, and fathead minnow were 100% at a dose of 1 ,100
ppm and 2,200 ppm.
Toxicitv/Filter pass-through. Chitosan is trapped in the sand filter and not
released into the receiving water.
Filter pass-through. A study evaluating the retention of chitosan acetate in a
mixed media filter (anthracite, sand, and garnet) was conducted by GE Betz.
The results upon a side-by-side comparison for Klaraid™ PC 1 192
(DADMAC) with chitosan acetate demonstrated that both products pass
through a standard mixed media filter without any retention in the layers of
the filter.
Hazard. This polymer is listed as Resource Conservation and Recovery Act
(RCRA) hazardous because of the acidity (at a pH of about 4).
Hazard. Chitosan acetate (1 % solution) has a pH of 3.9 to 4.0 and is
reported to be mildly irritating to the eyes.
                                           7-94

-------
                                                             Section 7: Technology Assessment
Diallydimethyl-ammonium chloride (DADMAC)
This polymer is also used in ATS for treating construction site stormwater runoff. DADMAC is
considered to be water soluble over a wide pH range. It has a high affinity for suspended
sediment but can have the ability to pass through treatment media to the receiving water
(MacPherson 2006a).

Table 7-30 presents information from several studies regarding toxicity of DADMAC to aquatic
organics, and filter pass-through results.

                         Table 7-30. DADMAC acetate study results
Vendor/source
Macpherson 2006a
Macpherson 2006a
ProTech GCS, Inc., in
conjunction with GE Betz
Results
Toxicity. Tramfloc, Inc., reports daphnia magna, 48-hour LC50 of 0.23 mg/L
for the Tramfloc Polydadmac 552, 553 and 557 products; however, aquatic
toxicity is reduced by factors of 1 0 to 1 00 times in the presence of 5 to 1 0
mg/L organics found in most surface waters.
Filter pass-throuqh. Hiqh affinity for suspended sediment but miqht have the
ability to pass through treatment media to the receiving water
Toxicitv. Demonstrated that, on a dose/response basis, DADMAC reduced >
1 ,000 NTU-water to 2 NTU at a dose of 25 ppm. In addition, aquatic toxicity
testing revealed a 95%, 1 00%, and 1 00% survival rate for daphnia magna,
rainbow trout, and fathead minnow, respectively. The ProTech and GE study
reports that the polymer DADMAC was the most economical for its removal
of suspended sediment and disposal costs.
Polyacrylamide (PAM)
PAM are a broad class of compounds that include cationic (positively charged) and anionic
(negatively charged) polyacylamides. PAMs are water soluble over a wide pH range and exhibit
a high affinity for suspended sediment.

Table 7-31 presents information from several studies regarding toxicity of PAMs to aquatic
organics and hazard information.

                               Table 7-31. PAM study results
Vendor/source
(see Section 7.2.4.10)
MacPherson 2006a
Results
Toxicitv. At very high doses, irritation in humans and toxicity to certain
aquatic organisms can be observed; however, in general PAMs are
considered to be nontoxicto aquatic organisms.
Hazard. PAM in the solid state has hiqhly hyqroscopic dust and, if inhaled,
could impair breathing.
Toxicitv. Anionic PAMs are not expected to be toxic to aquatic life at normal
dose rates (LC50 for most aquatic species is greater than 1 00 mg/L).
PAMs have been approved for use in Florida and New Hampshire (ATS Industry Task Force
2007). In California, Washington, Michigan, and Oregon, cationic PAM cannot be used for
construction site soil stabilization practices (MacPherson 2006a). McLaughlin has conducted
                                          7-95

-------
                                                             Section 7: Technology Assessment
extensive research on the use of PAM on construction sites in North Carolina, as well as PAM
toxicity.

Aluminum-Based Coagulants
The aluminum-based coagulants do not appear to be as widely used in ATS at construction sites
for stormwater runoff. Table 7-32 presents aluminum toxicity on aquatic organisms. Note that
aluminum floe will be in various aluminum complexes, which can become aqueous aluminum
depending on time and site-specific physical and environmental conditions (e.g., pH,
temperature, hardness and alkalinity, release of trapped sediment).

                     Table 7-32. Aluminum-based coagulant study results
Vendor/source
MacPherson 2006a
Sutherland 1999 (references
Oughton 1992)
ProTech GCS, Inc., in
conjunction with GE Betz
MacPherson 2006a
Results
Toxicity. Specifically, studies with juvenile striped bass indicate that this
species is extremely sensitive to several forms of aqueous aluminum
(referenced Driscoll et al. 1980; Palawski et al. 1985; Skogheim and
Rosseland 1986; Rosselan etal. 1992).
Toxicitv. An in situ study (Hall et al. 1985) with larval striped bass found 90%
to 99% mortality in river water with 0.48 to 4.1 mg/L aluminum and pH levels
between 6.0 and 6.8.
Toxicitv. Klauda et al. (1 989) support the theory that monomeric aluminum
(mAI), the inorganic fraction, is potentially the most toxic to early life stages of
migratory fish.
Toxicitv. Polymers created from aluminum and water collect on gills and limit
respiration.
Toxicitv. ProTech GCS and GE Betz conducted a study using the
coagulant/flocculant Aluminum Chlorhydroxide and found that at optimum
dose (75 ppm) survival rates for daphnia magna, rainbow trout, and fathead
minnow were 95%, 100%, and 95%, respectively. At two times, the optimum
dose (150 ppm) results were similar, showing no increased toxicity.
Hazard. These aluminum-based water treatment aqents also pose a risk to
human eyes and skin if not properly handled.
      7.2.5.3.   ATS Filtration

Filtration is a final treatment step in ATS designed to remove residual, low concentrations of
target pollutants before discharge. Multimedia filtration (mixed-media filtration) is one of the
oldest and most widely applied types of filtration for removing suspended solids from aqueous
liquid streams. This form of filtration uses a bed of granular particles as the filter medium.
Granular media filters are used to remove suspended solids from construction stormwater after
chemical addition creates a floe to filter. The bed can consist of one type of medium (e.g., sand)
of the same particle size, or multiple particle sizes. Different types of media (e.g., sand and
gravel, sand and anthracite) with differing densities and different particle sizes compose the bed
of a multimedia filter. Multimedia filters  can be more efficient but more expensive and complex
than single-media filters. For that reason, sand filters are most commonly used in construction
ATS. The  filter bed is inside a basin or tank and is supported by an underdrain system, which
allows the filtered liquid to be drawn off while retaining the filter medium in place.  As
suspended particle-laden water passes through the bed of the filter medium, particles are trapped
                                          7-96

-------
                                                            Section 7: Technology Assessment
on top of and within the bed. Once the pressure drop across the filter is large enough to impede
flow, the filter is backwashed, and the backwash water is typically recirculated to the influent
flow.

Vendors are also marketing bag and cartridge filters that can be used as a final filtration step.
Bag filters are available in a range of pore sizes, and cartridge filters are available with various
media types. The filters can be used as a final polishing step before discharge.

     7.2.5.4.  Other Emerging Treatment Technologies

While EPA's analysis was based primarily on chitosan-enhanced filtration, several other
advanced technologies are available to treat construction site stormwater runoff.
Electrocoagulation has been successfully used on a number of construction sites to meet turbidity
limits. At least one vendor is using a tube settler coupled with polymer addition before filtration
to help remove sediment. In addition, several commenters provided information to EPA about
other advanced turbidity control technologies that are in use or in development (see DCNs 43122
and 43123, and Docket Numbers EPA-HQ-OW-2008-0465.0525 and EPA-HQ-OW-2008-
0465.0527/0527.1).

 7.3.     REFERENCES

Adams, J, J. Janatzen, and D. Loudenslager.  2000. A Device to Alleviate Pollution from Urban
   Stormwater. Design Report. Oklahoma State University, Biosystems and Agricultural
   Engineering, Stillwater, OK.

Auckland Regional Council. 2004. The  Use of Flocculants and Coagulants to Aid the Settlement
   of Suspended Sediment in Earthworks Runoff: Trials, Methodology and Design [draft].
   Technical Publication 227. Auckland Regional Council, Auckland, New Zealand. (DCN
   41112.)

ATS Industry Task Force. 2007. Guidelines for the Design, Construction,  Operation and
   Maintenance of Active Treatment Systems for Stormwater. August 27,  2007. See Clear Creek
    Systems DCN 43003 in Section 4 of the Public Record.

Barfield, B.J. 2000. Presentation on Nonpoint Source Sediment and Stormwater. University of
   Puerto Rico, San Juan, PR.

Barfield, B.J., and M. Clar. 1985. Development of New Design Criteria for Sediment  Traps and
   Basins. Prepared for the Maryland Resource Administration, Annapolis, MD.

Barfield, B.J., C.T. Haan, E. Stevens, J.C. Hayes., and K.F. Holbrook. 2001. Engineering design
   aids for sediment control practices. In Proceedings of the International Symposium on Soil
   Erosion Research for the 21st Century. Honolulu, Hawaii, January 3-5, 2001.

Barfield, B.J., ID. Moore, and R.G. Williams. 1979. Sediment yield in surface mined
   watersheds. In Proceedings Symposium on Surface Mine Hydrology, Sedimentology, and
   Reclamation. University of Kentucky, College of Engineering, Lexington, KY.
                                         7-97

-------
                                                            Section 7: Technology Assessment
Barfield, B.J., J.C. Hayes, A.W. Fogle, and K.A. Kranzler. 1996. The SEDIMOT III - Model of
   Watershed Hydrology and Sedimentology. In Proceedings of'Fifth Federal Interagency
   Sedimentation Conference, Las Vegas, NV,  1996.

Barrett, M.E., J.E Kearney., T.G. McCoy, J.F. Malina, RJ. Charbeneau, and G.H. Ward. 1995.
   An Evaluation of the Use and Effectiveness of Temporary Sediment Controls. Technical
   Report CRWR 261. The University of Texas at Austin, Center for Research in Water
   Resources, Austin, TX.

Benik, S.R., B.N. Wilson, D.D. Biesboer, B. Hanse, and D. Stenlund. 1998. The Efficacy of
   Erosion Control Products at a MN/DOT Construction Site. Paper No. 982156. American
   Society of Agricultural Engineers, St. Joseph, MI.

Bhardwaj, A.K., and R.A. McLaughlin. 2008a. Simple Polyacrylamide Dosing Systems for
   Turbidity Reduction in Stilling Basins. Transactions of the American Society of Agricultural
   Engineers 51 (5): 1653-1662. Docket Number EPA 2008-0465-0984.4.

Bhardwaj, A.K., and R.A. McLaughlin. 2008b. Energy Dissipation and Chemical Treatment to
   Improve Stilling Basin Performance.  Transactions of the American Society of Agricultural
   Engineers 51 (5): 1645-1652. Docket Number EPA 2008-0465-0984.5.

Blandford, N. 2006. Mixed Media Filtration Study ofKlaraidPC 1192 and Chitosan. GE
   Infrastructure Water & Process Technologies. RTS #4454100. February 9, 2006. (DCN
   43006.)

Blench,  T. 1970. Regime Theory Design  of Canals with Sand Beds. Journal of the Irrigation and
   Drainage Division 96(2):205-213.

Britton,  S.L., K.M. Robinson, and BJ. Barfield. 2001. Modeling the effectiveness of silt fence.
   In Proceedings 7th Federal Interagency Sedimentation Conference, Reno, NV,  March 25-
   29, 2001. Volume 2, pp. 75-82.

Brown, W., and D. Caraco. 1997. Muddy water in - muddy water out? Watershed Protection
   Techniques 2(3): 1997.

Brown, W., and T. Schueler. 1997. The Economics ofStormwaterBMPs in the Mid-Atlantic
   Region. Prepared for the Chesapeake  Research Consortium, Edgewater, MD, by the Center
   for Watershed Protection, Ellicott City, MD.

Bullock, G., V. Blazer, S. Tsukuda, and S. Summerfelt. 2000. Toxicity of acidified  chitosan for
   cultured rainbow trout (Oncorhynchus mykiss). Aquaculture 185(2000):273-280. See
   ProTech GCS, Inc. DCN 43006 in Section 4 of the Public Record.

Chang, H.H. 1988. Fluvial Processes in River Engineering. Wiley Interscience, New York.

Claytor, R.  1997. Practical tips for construction  site phasing. Watershed Protection Techniques
   2(3): 1997.
                                         7-98

-------
                                                             Section 7: Technology Assessment
Corish, K. 1995. Clearing and Grading Strategies for Urban Watersheds. Metropolitan
   Washington Council of Governments, Washington, DC.

CSQA (California Stormwater Quality Association). 2003. Stormwater BMP Handbook:
   Construction. California Stormwater Quality Association, Menlo Park, CA.
   . Accessed June 12, 2008.

CWP (Center for Watershed Protection). 1996. Design of Stormwater Filtering Systems.
   Prepared for the Chesapeake Research Consortium, Solomons, MD, and U.S. Environmental
   Protection Agency Region 5, Chicago, IL, by Center for Watershed Protection, Ellicott City,
   MD.

Dorman, M.E., J. Hartigan, R.F. Steg, and T. Quasebarth. 1989. Retention, Detention and
   Overland Flow for Pollutant Removal from Highway Stormwater Runoff. Vol. 1. Research
   Report. Federal FHWA/RD 89/202. Highway Administration, Washington, DC. Also in
   Performance of Grassed Swales along East Coast Highways. Article No. 114 in The Practice
   of Watershed Protection. Center for Watershed Protection, Ellicott City,  MD. 2000.

Ellis, B. 1998. A further guide to swales: costs of operation and maintenance. PERMEATE.
   Scottish Environmental Protection Agency, Stirling, Scotland.
   .

Entry, J.A., and R.E. Sojka.  1999. Polyacrylamide Application to Soil Reduces the Movement of
   Microorganisms in Water. In Proceedings of the International Irrigation Show. Irrigation
   Associations, Orlando, FL, November, 1999.

Ettinger, C.E., and I.E. Lichty.  1979.  Evaluation of Performance Capability of Surface Mine
   Sediment Ponds. Final Report on  Contract Number 68-03-2677. U.S. Environmental
   Protection Agency, Industrial Environmental Research Laboratory.

Faircloth, W. 1999. Searching for a practical, efficient, economical sediment basin. In
   Proceedings of Conference  30. International Erosion Control Association. Nashville, TN,
   February 22-26, 1999, pp. 272-282.

Fifield, J. 1999. Designing Effective Sediment and Erosion Control for Construction Sites.
   Forester Press, Santa Barbara, CA.

Fisher, L.S., and A.R. Jarrett. 1984. Sediment retention efficiency of synthetic filter fabric.
   Transactions of the American Society of Agricultural Engineers 30(l):82-89.

Foster, G.R., R.A. Young, and W.H. Neibling. 1985. Sediment composition  for nonpoint source
   pollution analyses. Transactions of the American Society of Agricultural Engineers
   28(1):133-146.

Gannon, J. 2007. Comment Letter—Draft Construction General Permit. Clear Creek  Systems,
   Inc. April 2007. See Clear Creek  Systems DCN 43003 in Section 4 of the Public Record.
                                          7-99

-------
                                                             Section 7: Technology Assessment
Gillman, J.W. 1994. Riparian wetlands and water quality. Journal of Environmental Quality.
   23:896-900.

Goldberg, J. 1993. Dayton Avenue Swale Biofiltration Study. Seattle Engineering Department,
   Seattle, WA.

Griffin, M.L., BJ. Barfield, and R.C. Warner. 1985. Laboratory studies of dead storage in
   sediment ponds. Transactions of the American Society of Agricultural Engineers 28(3):799-
   804.

Haan, C.T., BJ. Barfield, and J.C. Hayes. 1994. Design Hydrology and Sedimentology for Small
   Catchments. Academic Press, San Diego, CA.

Harding, M.V. 1990. Erosion control effectiveness: comparative studies of alternative mulching
   techniques. Environmental Restoration, pp. 149-156, as cited in USEPA 1993. Guidance
   specifying management measures for sources of nonpoint pollution in coastal waters. EPA
   840-B-92-002. U.S. Environmental Protection Agency, Office of Water, Washington, DC.

Harper, H. 1988. Effects of Stormwater Management Systems on Groundwater Quality. Final
   Report. Prepared for Florida Department of Environmental Regulation, by Environmental
   Research and Design, Inc. Also in: Runoff and groundwater dynamics of two swales in
   Florida. Article No. 113 in The Practice of Watershed Protection. Center for Watershed
   Protection, Ellicott City, MD. .

Hayes, J.C., BJ. Barfield, and R.I. Barnhisel. 1984. Performance of grass filters under laboratory
   and field conditions. Transactions of the American Society of Agricultural Engineers
   27(5):1321-1331.

Hayes, J.C., and B J. Barfield. 1995. Engineering aids and design guidelines for control of
   sediment in South Carolina. In South Carolina Stormwater and Sediment Control Handbook
   for Land Disturbance Activities. South Carolina Department of Health and Environmental
   Control, Columbia,  SC.

Hayes, J.C., A.L. Akridge, B J. Barfield, and K.F. Holbrook. 2001. Simplifying design of
   sediment controls in Jefferson County, Kentucky. In Proceedings soil erosion research for
   the 21st Century: An international symposium and exhibition, Honolulu, HI, January 2001.
   Sponsored by American Society of Agricultural Engineers,  St. Joseph MI.

Herrera, N.M., and G.K. Felton.  1991. Hydraulics of flow through a rockfill dam using sediment-
   free water. Transactions of the American Society of Agricultural Engineers 34(3):871-875.

Heyer, T. No Date. Vegetative Measures for Streambank Stabilization. U.S. Department of
   Agriculture, U.S. Forest Service, St, Paul, MN.

Hirschi, M.C. 1981. Efficiency of Small Sediment Controls. Unpublished Agricultural
   Engineering File Report. University of Kentucky, Agricultural Engineering, Lexington, KY.
   (Details of the report are given in Chapter 9 of Haan et al, 1994).
                                         7-100

-------
                                                            Section 7: Technology Assessment
Holbrook, K.F., J.C. Hayes, BJ. Barfield, and A.W. Fogle. 1998. Engineering aids and design
   guidelines for control of sediment in South Carolina. In Proceedings of Conference 29.
   International Erosion Control Association. Reno, NV, February 16-20, 1998, pp. 129-140.

Honnigford, L.L. 2002. ECTC Offers a Website Tool to Help Landscape Architects with Erosion
   Problems. Erosion Control Technology Council, St. Paul, MN.

IDNR (Indiana Department of Natural Resources). 1992. Indiana Handbook for Erosion
   Controls in Developing Areas, Guidelines for Protecting Water Quality through the Control
   of Soil and Erosion and Sedimentation on Construction Sites. Indiana Department of Natural
   Resources, Division of Soil  Conservation, Indianapolis, IN.

Jarrett, A. 1999. Designing sedimentation basins for better sediment capture. In Proceedings of
   Conference 30. International Erosion Control Association. Nashville, TN, February 22-26,
   1999, pp. 218-233.

Kay-Shoemake, J, M. Watwood, R. Sojka, and R. Lentz. 2000. Soil amidase activity in
   polyacrylamide (PAM) treated soils and potential activity toward common amide containing
   agrochemicals. Biology and Fertility of Soils 31(2): 183-186.

Kercher, W.C., J.C. Landon, and R. Massarelli. 1983. Grassy swales prove cost-effective for
   water pollution control. Public Works 16:53-55.

Knisel, W.G., ed. 1980. CREAMS: A Field-Scale Model for Chemicals, Runoff, and Erosion
   from Agricultural Management Systems. Conservation Research Report No. 26.  U.S.
   Department of Agriculture, Washington, DC.

Koon, J. 1995. Evaluation of Water Quality Ponds and Swales in the Issaquah/East Lake
   Sammamish Basins. King County Surface Water Management and Washington Department
   of Ecology,  Seattle, WA.

Kouwen, N. 1990. Silt Fences to Control Sediment on Construction Sites. Technical Publication
   MAT-90-03. Ontario Ministry of Transportation, Research and Development Branch,
   Downsview, Ontario, Canada.

Lancaster, T., D. Lutyens, and D. Austin. 2002. Flexible Channel Lining System: The Benefits of
   Geosynthetically Reinforced Vegetation Over Rock Riprap. Erosion Control Technology
   Council, St.  Paul, MN.

Lane, E.W. 1955. The  importance of fluvial geomorphology in hydraulic engineering. In
   Proceedings, American Society of Civil Engineers. Vol. 81, Paper 745, July 1955.
   . Accessed November 2008.

Lindley, M., B.J. Barfield, J. Ascough, B.N. Wilson, and E. Stevens.  1998. The surface
   impoundment element for WEPP. Transactions of the American Society of Agricultural
   Engineers. 41(3):555-564.
                                         7-101

-------
                                                            Section 7: Technology Assessment
Masters, A., K.A. Flahive, S. Mostaghimi, D. Vaughan, A. Mendez, M. Peterie, and S. Radke.
   2000. A Comparative Investigation of the Effectiveness of Polyacrylamide (PAM) for Erosion
   Control in Urban Areas. Paper No. 002176. American Society of Agricultural Engineers, St.
   Joseph, MI.

Markusic, M.S., and R. A McLaughlin. 2008. Effects of Design Changes on Sediment Retention
   Basin Efficiency. In Proceedings of the 39th Annual Conference and Expo of the
   International Erosion Control Association 2008. February 18-22, 2008, Orlando, FL.

McBurnie, J.C., BJ. Barfield, M.L. Clar, and E. Shaver. 1990. Maryland sediment detention
   pond design criteria and performance. Applied Engineering in Agriculture 6(2): 167-173.

McLaughlin, R. A. No Date a. Target Turbidity Limits for Passive Treatment Systems. North
   Carolina State University, Raleigh, NC. Docket Number EPA-HQ-OW-2008-0465.0984.6.

McLaughlin, R.A. No Date b. Water Quality Improvements Using Modified Sediment Control
   Systems on Construction Sites. Project HWY-2002-04. North Carolina Department of
   Transportation. Raleigh, NC. Docket Number EPA-HQ-OW-2008-0465.0984.3.

McLaughlin, R.A. 2006. Polyacrylamide Blocks for Turbidity Control on Construction Sites.
   American Society of Agricultural and Biological Engineers. St. Joseph, MI. Docket Numbers
   EPA-HQ-OW-2008-0465.0984.7 and 0984.10.

McLaughlin, R.A., and M.S. Markusic. 2007. Evaluating Sediment Capture Rates for Different
   Sediment Basin Designs. North Carolina Department of Transportation Project Authorization
   No. HWY-2006-17. August 23, 2007.

McLaughlin, R.A., and A. Zimmerman. 2008. Best Management Practices for Chemical
   Treatment Systems for Construction Stormwater andDewatering. FHWA-WFL/TD-09-001.
   Federal Highway Administration, Western Federal Lands Highway Division, Vancouver,
   WA. (DCN43124)

McLaughlin, R.A., S.E. King, and G.D. Jennings. 2009. Improving construction site runoff
   quality with fiber check dams and polyacrylamide. Journal of Soil and Water Conservation.
   March/April 2009-Vol. 64, No. 2.  (DCN 43116)

MDE (Maryland Department of the Environment). 1994. Maryland Standards and Specifications
   for Soil Erosion and Sediment Control. Maryland Department of the Environment, Water
   Management Administration, Soil Conservation Service, and State Soil Conservation
   Committee, Baltimore, MD.

MacPherson, J. 2006a. Comparative Analysis of Characteristics of Stormwater Treatment Agents
   Used on Construction Sites. Natural Site Solutions, LLC. October 2, 2006. See Rain for Rent
   DCN 43007 in Section 4 of the Public Record.

MacPherson, J. 2006b. Natural Site Solutions, LLC Original Developer of the Chitosan
   Enhanced Sand Filtration Technology. Natural Site Solutions, LLC.  May 2006. See Rain for
   Rent DCN 43007 in Section 4 of the Public Record.
                                        7-102

-------
                                                           Section 7: Technology Assessment
McCoy, S. 2005. Filter Sock Presentation provided at Erosion, Sediment Control and
   Stormwater Management with Compost BMPs Workshop, U.S. Composting Council 13th
   Annual Conference and Trade Show, January 2005, San Antonio, Texas.

Napper, C. 2006. Burned Area Emergency Response (BAER) Treatment Catalog. National
   Technology and Development Program Technical Report 0625 1801-SDTDC. U.S.
   Department of Agriculture, Forest Service, San Dimas Technology and Development Center.
   San Dimas, CA.

NAHB (National Association of Home Builders) Research Center. No Date. Storm Water Runoff
   & Nonpoint Source Pollution  Control Guide for Builders and Developers. National
   Association of Home Builders, Research Center, Washington, DC.

NCDNR (North Carolina Department of Natural Resources). 1988. North Carolina Erosion and
   Sediment Control Planning and Design Manual. North Carolina Sedimentation Control
   Commission, the North Carolina Department of Natural Resources and Community
   Development, and the North Carolina Agricultural Extension Service, Raleigh, NC.

Nichols, E. 1997. Chitosan: chemistry and use in water clarification. In Proceedings of the 2nd
   Annual Chemistry Symposium. National Spa and Pool Institute, November 1997, pp. 19-25.
   See Halosource, Inc. DCN 43005 in Section 4 of the Public Record.

Oakland, P.H. 1983.  An evaluation of stormwater pollutant removal through grassed swale
   treatment. In Proceedings of the International Symposium of Urban Hydrology, Hydraulics
   and Sediment Control.  HJ. Sterling, Lexington, KY, pp. 173-182.

ODEQ (Oregon Department of Environmental Quality). 2004. Best Management Practices for
   Stormwater Discharges Associated with Construction Activity,  Guidance for Eliminating or
   Reducing Pollutants in Stormwater Discharges,  Oregon Department of Environmental
   Quality, Northwest Region, Portland, OR.

Pitt, R., and J. McLean. 1986. Toronto Area Watershed Management Strategy Study: Humber
   River Pilot Water shed Project. Ontario Ministry of Environment and Energy, Toronto,
   Canada.

ProTech GCS. 2004. Technical Report: TR01.1 Polymer Coagulants and Flocculants for
   Stormwater Applications. ProTech GCS, Fairfield, CA. (DCN 43006)

Ray, H. 2001. Storm-Klear™Liqui-Floc™ Material Safety Data Sheet, Revision No. 2. Vanson.
   August 31, 2001. Natural Site Solutions, Redmond, WA. (DCN 43006)

Renard, K.G., G.R. Foster, G.A. Weesies, K.D., K. McCool, and D.C. Yoder. 1994. Predicting
   Soil Erosion by Water—A Guide to Conservation Planning with the Revised Universal Soil
   Loss Equation (RUSLE). U.S. Department of Agriculture, Agricultural Research Service,
   Washington, DC.
                                        7-103

-------
                                                            Section 7: Technology Assessment
Roa-Espinosa, A., G.D. Bubenzer, andE.S. Miyashita. 2000. Sediment and Runoff Control on
   Construction Sites using Four Application Methods ofPolyacrylamideMix. National
   Conference on Tools for Urban Water Resource Management and Protection, Chicago,
   February 7-10, 2000.

R.S. Means. 2000. Building Construction Cost Data. 58th ed. R.S. Means, Co., Kingston, MA.

R.S. Means. 2009. Building Construction Cost Data. 67th ed. R.S. Means, Co., Kingston, MA.

Seattle Metro and Washington Department of Ecology. 1992. Biofiltration Swale Performance:
   Recommendations and Design Considerations. Publication No. 657. Water Pollution Control
   Department, Seattle, Washington. Also in Watershed Protection Techniques 1(3): 117-119.

SCDHEC (South Carolina Department of Health and Environmental Control).  1995. South
   Carolina Stormwater Management and Sediment Reductions Regulations. In South Carolina
   Stormwater Management and Sediment Control Sedimentology Resource. South Carolina
   Department of Health and Environmental Control, Columbia, SC.

Schumm, S.A. 1977. The Fluvial System. John Wiley and Sons, New York.

Simons, D.B., and M.L. Albertson. 1960. Uniform water conveyance channels in alluvial
   material. In Proceedings of American Society of Civil Engineers 86(HY5):33-71.

Smolen, M.D., D.W. Miller, L.C. Wyall, J. Lichthardt, and A.L. Lanier. 1988. Erosion and
   Sediment Control Planning and Design Manual. North Carolina Sedimentation Control
   Commission and North Carolina Department of Natural Resources and Community
   Development, Raleigh, NC.

Sojka, R.E., and R.D. Lentz. 1994. Poly aery lamide (PAM): A New Weapon in  the Fight Against
   Irrigation-induced Erosion. Note #01-94. U.S. Department of Agriculture, Agricultural
   Research Service, Northwest Irrigation and Soils Research Lab Station, Kimberly, ID.

Sprague, J.  1999. Assuring the Effectiveness of Silt Fences and Other Sediment Barriers.
   TRI/Environmental, Inc., Greenville, SC. Also in Proceedings of Conference 30.
   International Erosion Control Association. Nashville, TN, February 22-26, 1999.

Stahre, P., and B. Urbonas. 1990. Stormwater Detention for Drainage, Water Quality, andCSO
   Management. Prentice Hall, Englewood Cliffs, NJ.

Stevens, L.E., TJ. Ayers, J.B. Bennett, K. Christensen, M.J.C. Kearsley, VJ. Meretsky, A.M.
   Phillips III, R.A. Parnell, J. Spence, M.K. Sogge, A.E. Springer, and D.L. Wegner. 2001.
   Planned flooding and Colorado River riparian trade-offs downstream from  Glen Canyon
   Dam, Arizona. Ecological Applications 11 (3): 701-710.

Sturm, T.W., and R.E. Kirby.  1991. Sediment Reduction in Urban Stormwater Runoff from
   Construction Sites.  Georgia Institute of Technology, Atlanta, GA.
                                         7-104

-------
                                                            Section 7: Technology Assessment
Sutherland, D. 1999. Washington Aqueduct Sediment Discharges Report of Panel
   Recommendations. U.S. Fish and Wildlife Service, Washington, DC. (DCN 43078)

SWRPC (Southeastern Wisconsin Regional Planning Commission). 1991. Costs of Urban
   Nonpoint Source Water Pollution Control Measures. Technical Report no. 31. Southeastern
   Wisconsin Regional Planning Commission, Waukesha, WI.

Tobiason, S., D. Jenkins, E. Molash, and S. Rush. 2000. Polymer use and testing for erosion and
   sediment control on construction sites: Recent experience in the Pacific Northwest. In
   Proceedings of Conference 31. International Erosion Control Association. Palm Spring, CA,
   February 21-25, 2000, pp. 41-52.

Tapp, J.S., and BJ. Barfield. 1986. Modeling the flocculation process in sediment ponds.
   Transactions of the American Society of Agricultural Engineers 29(3):741-747.

Tapp, J.S., BJ.Barfield, andM.L.  Griffin. 1981. Predicting suspended solids removal in pilot
   scale sediment ponds utilizing chemical flocculation. Research Report IMMR 81/063.
   University of Kentucky, Institute for Mining and Minerals Research, Lexington, KY.

UNEP (United Nations Environment Programme). 1994. Guidelines for sediment control
   practices in the insular Caribbean. CEP Technical Report No. 32. United Nations
   Environment Programme, Caribbean Environment  Programme, Kingston, Jamaica.
   .

USAF (U.S. Air Force). 1998. USAF landscape design: Erosion control measures. U.S. Air
   Force, St. Paul, MN.
   .

USDA (U.S. Department of Agriculture). 1979. Engineering Field Handbook. U.S. Department
   of Agriculture, Soil Conservation Service, Washington, DC.

USDOT (U.S. Department of Transportation). 1995. Best management practices for erosion and
   sediment control. Report No. FHWA-FLP-94-005.  U.S Department of Transportation,
   Eastern Federal Lands Highway Design, Sterling, VA.

USEPA (U.S. Environmental Protection Agency). 1971. Control of sediment resulting from
   construction of highways and land development. U.S. Environmental Protection Agency,
   Washington, DC.

USEPA (U.S. Environmental Protection Agency). 1983. Results of the Nationwide Urban Runoff
   Program. NTIS PB84-185552. U.S. Environmental Protection Agency, Water Planning
   Division, Washington, DC.

USEPA (U.S. Environmental Protection Agency). 1992. Storm Water Management for
   Construction Activities: Developing Pollution Prevention Plans and Best Management
   Practices. EPA 832-R-92-005. U.S. Environmental Protection Agency, Office of Water,
   Washington, DC.
                                         7-105

-------
                                                            Section 7: Technology Assessment
USEPA (U.S. Environmental Protection Agency). 1993. Guidance Specifying Management
   Measures for Sources of Nonpoint Pollution in Coastal Waters. EPA 840-B-92-002. U.S.
   Environmental Protection Agency, Office of Water, Washington, DC.

USEPA (U.S. Environmental Protection Agency). 2000. Urban Nonpoint Source Management
   Measure Guidance -Draft. U.S. Environmental Protection Agency, Office of Water,
   Washington, DC.

USEPA (U.S. Environmental Protection Agency). 2003. Documentation of Unit Cost for Erosion
   and Sediment Controls. (DCN 75008) Docket Numbers OW-2002-0030-0071 and OW-2002-
   0030-0072.

USEPA (U.S. Environmental Protection Agency). 2008. NationalMenu of Best Management
   Practices for Stormwater Phase II. Washington, DC.
   . Updated January 09, 2008.

Vanderwel, D., and S. Abday. 1998. An introduction to water erosion control. Alberta,
   Agriculture, Food and Rural Development, Alberta, Canada.

VDCR (Virginia Department of Conservation and Recreation).  1995. Virginia Erosion &
   Sediment Control Fie Id Manual. 2nd ed. Virginia Department of Conservation and
   Recreation, Division of Soil and Water Conservation, Richmond, VA.

VDCR (Virginia Department of Conservation and Recreation). 2001. Virginia Erosion and
   Sediment Control Law, Regulations, and Certification Regulations. Virginia Department of
   Conservation and Recreation, Division of Soil and Water Conservation, Richmond, VA.

Wang, T., D. Spyridakis, B. Mar, and R. Horner. 1981. Transport, Deposition and Control of
   Heavy Metals in Highway Runoff. FHWA-WA-RD-39-10. University of Washington,
   Department of Civil Engineering, Seattle, WA.

Ward, A.D., C.T. Haan, and BJ. Barfield. 1977. Simulation of the Sedimentology of Sediment
   Detention Basins. Research Report No. 103. University of Kentucky, Water Resources
   Research Institute, Lexington, KY.

Ward, A.D., C.T. Haan, and IS. Tapp 1979. The DEPOSITS Sedimentation Pond Design
   Manual. University of Kentucky, OISTL, University of Kentucky Institute for Mining and
   Minerals Research, Lexington, KY.

Warner, R.C., P.J. Schwab, and DJ. Marshall. 1999. SEDCAD 4 for Windows 95 & NT-Design
   Manual and User's Guide. Civil Software Design, Ames, IA.

WDEC (Washington State Department of Ecology).  1992. Stormwater Management Manual for
   the Puget Sound Basin. Washington State Department of Ecology, Olympia, WA.

Welborn, C., and J. Veenhuis. 1987. Effects of Runoff Controls on the Quantity and Quality of
   Urban Runoff in Two Locations in Austin, TX. U.S. Geological Survey  Water Resources
   Investigations Report. 87-4004, pp. 88. U.S. Geological Survey, Reston, VA.
                                        7-106

-------
                                                            Section 7: Technology Assessment
Williams, J.R. No Date. Sediment yield prediction with Universal Equation using runoff energy
   factor. Present and Prospective Technology for Predicting Sediment Yields and Sources.
   Publication ARS-S-40. U.S. Department of Agriculture, Agricultural Research Service,
   Washington, DC.

Wilson, B.N., and BJ. Barfield. 1984. A sediment detention pond model using CSTRS mixing
   theory. Transactions of American Society of Agricultural Engineers 27(5): 1339-1344.

Wilson, B.N., BJ. Barfield, and ID. Moore. 1982. A hydrology andsedimentology watershed
   model. University of Kentucky, Department of Agricultural Engineering, Lexington, KY.

Wischmeier, W.H., andD.D. Smith. 1965. Predicting Rainfall-Erosion Losses from Cropland
   East of the Rocky Mountains—Guide for Selection of Practices for Soil and Water
   Conservation. Agricultural Handbook No. 282. U.S. Department of Agriculture, Washington,
   DC.

Wishowski, J.M., M. Mamo, and G.D. Bubenzer. 1998. Trap Efficiencies of Filter Fabric Fence.
   Paper No 982158, American Society of Agricultural Engineers, St. Joseph, MI.

Wyant, D.C. 1980. Evaluation of Filter Fabric for Use as Silt Fences. Virginia Highway and
   Transportation Research Council, Richmond, VA.

WYDEQ (Wyoming Department of Environmental Quality). 1999. Urban Best Management
   Practices for Nonpoint Source Pollution. Wyoming Department of Environmental Quality,
   Water Quality Division, Point and Nonpoint Source Programs, Cheyenne,  WY.

Yousef, Y., M. Wanielista, H. Harper, D. Pearce, and R.  Tolbert. 1985. Best Management
   Practices: Removal of Highway Contaminants by Roadside Swales. Final report. University
   of Central Florida, Florida Department of Transportation, Orlando, FL. Also in Pollutant
   Removal Pathways in Florida Swales. Watershed Protection Techniques 2(1):299-301.

Yu, S., S. Barnes, and V. Gerde. 1993. Testing of Best Management Practices for Controlling
   Highway Runoff. FHWA/VA-93-R16. Virginia Transportation Research Council. Also in
   Performance  of Grassed Swales Along East Coast Highways. Article No. 114 in The Practice
   of Water shed Protection. Center for Watershed Protection, Ellicott City, MD. 2000.
   .
                                         7-107

-------
                                                 Section 8: Regulatory Development and Rationale
8.   BCT COST-REASONABLENESS ASSESSMENT

This section presents a summary of the Best Conventional Pollutant Control Technology (BCT)
methodology and the results of the two-part cost-reasonableness test. In considering whether to
promulgate BCT limits more stringent than the requirements being promulgated for Best
Practicable Control Technology (BPT), the U.S. Environmental Protection Agency (EPA)
considered whether technologies are available that would achieve greater removals of
conventional pollutants than the BPT effluent limitations guidelines. EPA also considered
whether those technologies are cost-reasonable according to the BCT cost test, which compares
the incremental removals and costs associated with BCT limitations to benchmarks associated
with BPT and publicly owned treatment works (POTWs).

 8.1.    BACKGROUND ON THE BCT COST TEST

In 1977 Congress amended the Clean Water Act to include section 304(b)(4)(B). This provision
specifies that, among other factors, the assessment of BCT effluent limitations must include
consideration of

       ...the reasonableness of the relationship between the costs of attaining a reduction
       in effluents and the effluent reduction benefits derived, and the comparison of the
       cost and level of reduction of such pollutants from the discharge of publicly
       owned treatment works to the cost and level of reduction of such pollutants from a
       class or category of industrial sources...

Accordingly, EPA developed the BCT methodology to determine whether it is cost-reasonable
for an industry category or subcategory to control conventional pollutants at a level more
stringent than would be achieved by BPT effluent limitations.

The BCT methodology was originally published on August 29, 1979, along with the
promulgation of BCT effluent limitations guidelines for a number of industry sectors  (44 Federal
Register [FR] 50732). The crux of the methodology was a comparison of the costs of removing
conventional pollutants for an average-sized POTW. The Fourth Circuit of the U.S. Court of
Appeals remanded the BCT regulation and directed EPA to develop an industry cost test in
addition to the POTW test. EPA subsequently proposed a revised BCT methodology in 1982 that
addressed the industry cost test (47 FR 49176; October 29, 1982). In 1984 EPA again addressed
the BCT methodology and proposed to base the POTW benchmark on model plant costs (49 FR
37046; September 20, 1984). The final BCT methodology was published in 1986, maintaining
the basic approach of the  1982 proposed BCT methodology and adopting the use of the new
model POTW data (51 FR 24974; July 9, 1986). These guidelines state that the BCT cost
analysis "...answers the question of whether it is 'cost reasonable' for industry to control
conventional pollutants at a level more stringent than BPT effluent limitations already require."
See 51 FR at 24974. Conventional pollutants are biochemical oxygen demand (BOD), total
suspended solids (TSS), oil and grease, fecal  coliform, and pH.

The final BCT methodology incorporates two cost tests to establish cost reasonableness: the
POTW test and the industry cost-effectiveness test. Each of these tests is compared with
                                         8-1

-------
                                                 Section 8: Regulatory Development and Rationale
established benchmarks, the derivation of which is described in detail in the 1986 FR notice. The
BCT cost methodology is described in more detail in the following section.

 8.2.    OPTIONS EVALUATED FOR BCT

   8.2.1.    OPTION 1

Option 1 contains requirements for implementing a variety or erosion and sediment controls on
all construction sites that are required to obtain a permit.

   8.2.2.    OPTION 2

Option 2 contains the same requirements as Option 1. In addition, construction sites of 30 or
more acres of disturbed land would be required to meet a numeric turbidity limit in stormwater
discharges from the site based on the application of ATS.

   8.2.3.    OPTION 3
Option 3 contains the same requirements as Option 1. Option 3 also requires all sites with 10 or
more acres of disturbed land to meet a numeric turbidity standard based on the application of
ATS.

   8.2.4.    OPTION 4

Option 4 contains the same requirements as Option 1. Option 4 also requires all sites with 10 or
more acres of disturbed land to meet a numeric turbidity standard based on the application of
passive treatment systems.

 8.3.    CALCULATION OF THE  BCT COST TEST

POTW Test
The first part of the BCT cost test is the POTW test. The POTW test compares the cost per
pound of conventional pollutants removed  by industrial dischargers in upgrading from BPT to
BCT candidate technologies, to the cost per pound of removing conventional pollutants in
upgrading POTWs from secondary treatment to advanced secondary treatment.

To pass the POTW test,  the cost per pound of conventional pollutant discharges removed in
upgrading from BPT to the candidate BCT must be less than the POTW benchmark. The POTW
benchmark presented in the 1986 Federal Register notice is $0.25 per pound (in 1976 dollars) for
industries in  which the cost per pound of pollutant reduction is based  on long-term performance
data. EPA used cost index data from R.S. Means Historical Cost Indices to update this POTW
benchmark to 2008 dollars according to the following equation:
             Index for 2008
             	x  Cost in 1976$ = Cost in 2008$
             Index for 1976
                                         8-2

-------
                                                   Section 8: Regulatory Development and Rationale
                     173.0
                     46.9
x $0.25 = $0.92
Using estimated reductions for TSS, EPA then calculated the incremental costs per pound of
conventional pollutant removed ($/lb) for each candidate BCT technology option. If any candidate
technology option passes the first part of the BCT cost test (i.e., is less than the inflation-adjusted
value of $0.92 in 2008 dollars), the technology is further evaluated in the second part of the test.
EPA used only TSS pollutant reductions for the cost test calculations, because of the limited data
available. However, EPA expects that discharges of oil and grease and fecal coliform would be
minimal from construction sites. EPA also expects that BOD, where present, would be removed
along with TSS. The results of the POTW test are presented in Table 8-1.

                             Table 8-1. POTW cost test results
BCT
option
1
2
3
4
Total annual costs and
conventional pollutant
removals
Cost
(million $)
(2008$)
176
4,863
9,081
959
Pollutant
removals
(million Ibs)
1,743
3,616
4,507
3,971
Incremental costs and conventional
pollutant removals, relative to BPTa
Cost
(million $)
(2008$)
Pollutant
removals
(million Ibs)
Cost per
pound
($/lb)
POTW cost
test result
(< $0.92/lb)

4,687
8,905
783
1,873
2,764
2,228
2.50
3.22
0.35
Fail
Fail
Pass
a Option 1 is equal to the BPT effluent limitations. Therefore, all incremental values are calculated relative to Option 1.

Industry Cost-Effectiveness Test
The second part of the BCT cost test is the industry cost-effectiveness test, which computes the
ratio of two incremental costs. The first of these incremental costs is the cost per pound of
conventional pollutants removed in upgrading from BPT to the BCT candidate technology. This
value serves as the numerator of the ratio. The second incremental cost, which serves as the
denominator of the ratio, is the cost per pound of conventional pollutants removed by BPT
relative to no treatment (i.e., this value compares raw wasteload to pollutant load after
application of BPT). This ratio is compared to an industry cost benchmark, which is based on
POTW cost and pollutant removal data. The industry cost benchmark is also a ratio of two
incremental costs: the cost per pound to upgrade a POTW from secondary treatment to advanced
secondary treatment, divided by the cost per pound to initially achieve secondary treatment. If
the industry  cost-effectiveness test is lower than the industry cost benchmark of 1.29 (i.e., the
normalized cost increase must be less than 29 percent), the candidate BCT technology passes this
part of the cost test. The calculation and results of the industry cost-effectiveness test are
presented in Tables 8-2 and 8-3. Because Options 2, 3, and 4 fail the second part of the BCT cost
test, BCT is  set equal to BPT, which is Option 1.
                                           8-3

-------
                                                   Section 8: Regulatory Development and Rationale
                       Table 8-2. Cost and pollutant removals for BPT

Baseline
Option 1 Incremental
Total BPT
Total annual costs
(million $)
(2008$)
2,804
176
2,980
Conventional pollutant
removals
(million Ibs)
44,620
1,743
46,363
BPT cost per pound
($/lb)

0.064
                      Table 8-3. Industry cost-effectiveness test results
BCT option
1
2
3
4
Incremental cost per pound to
upgrade from BPT to BCT
($/lb)
0
2.50
3.22
0.35
Calculated
ratio
0
38.92
50.12
5.47
Industry cost-effectiveness
test result
(<1.29)
Pass
Fail
Fail
Fail
 8.4.    REFERENCES

R.S. Means. 2008. Building Construction Cost Data. 66th ed. R.S. Means, Co., Kingston, MA.

Earth System Science Center. 2006. Pennsylvania State University, College of Earth and Mineral
    Sciences, Earth System Science Center, State College, PA.
    .
                                           8-4

-------
                                       Section 9: Estimating Incremental Costs for the Final Regulation
9.   ESTIMATING  INCREMENTAL COSTS FOR THE FINAL
     REGULATION

 9.1.    OVERVIEW

This section presents the approach used for estimating the incremental costs associated with the
regulatory options considered. This section also includes discussion on selecting and developing
cost model inputs; the components of cost; and the methodology for estimating costs, including an
overview of the C&D Cost Spreadsheet Models. The economic analyses conducted for the industry
are described in the document Economic Analysis for Final Action for Effluent Guidelines and
Standards for the Construction and Development Category (USEPA 2009).

The U.S. Environmental Protection Agency's (EPA's) first step in estimating national costs for
each option was developing an array of model construction sites. EPA then estimated unit
compliance costs for the regulatory options. The most significant input parameter in estimating
the size and cost of treatment equipment for model construction sites is the volume of rainfall
requiring treatment. EPA estimated the amount of stormwater runoff requiring treatment using
the drainage area of the  model site,  state-specific rainfall estimates, and runoff coefficients to
estimate the volume of rainfall converted to runoff and requiring subsequent treatment. The
model project costs were then scaled to the state and national level on the basis of the national
project distribution described in Table 4-3.

Costs for the U.S. territories were not estimated because EPA lacked data on the annual amount
of new construction acreage in these areas. However, assuming a small amount of construction
occurs in those areas, EPA expects that the values would be low in comparison to the national
costs.

The total costs of the options considered are presented in Table 9-1.
              Table 9-1. Estimated total annual social costs of regulatory options
                                  for the C&D industry
Regulatory option
Option 1
Option 2
Option 3
Option 4
Annual cost
(millions 2008 dollars)
$176
$4,863
$9,081
$959
                                          9-1

-------
                                        Section 9: Estimating Incremental Costs for the Final Regulation
 9.2.    DEVELOPMENT OF MODEL CONSTRUCTION SITES AND ESTIMATING
         TREATMENT VOLUMES

   9.2.1.   MODEL CONSTRUCTION SITES

As discussed in Section 4.2.2 and in Appendix C, EPA developed a series of model projects from
an analysis of NOT data. This matrix consisted of 12 model project sizes and 12 model project
duration, yielding a total of 144 individual model projects. By analyzing the NOT data, EPA was
able to develop a distribution of model projects for each model project size and duration category
for three project types (residential construction, nonresidential construction, and transportation).
For each state and the District of Columbia, EPA was able to estimate the number of model
projects in those site size categories. Table 9-2 shows the model project matrix. Table 9-3 shows
the estimated number of acres developed per year for the model projects. (Tables 9-2 and 9-3
appear later in this section grouped with other tables of similar size.)

EPA based its costing on the size of the model site, the volume of runoff being treated, and the
duration of land disturbance. EPA assumes, for costing purposes, that the duration of
construction activities (and hence the duration of time needed to meet the turbidity limits) under
Options 2, 3, and 4 for some projects are shorter than the NOT project durations. That is because
projects are likely to transition from major land disturbing activities into the vertical construction
phase. For the final months of a project,  the majority of soil disturbance would likely be
complete as structures are constructed. Final stabilization of remaining disturbed areas around
the building footprint would be complete once the majority of the exterior construction work is
complete. Final vegetation is usually one of the last steps in the construction sequence, and the
Notice of Termination (NOT) is usually  filed after final stabilization. Because the turbidity limits
under Options 2, 3, and 4 are tied to the disturbed area of the site, EPA expects that disturbed
land  on the majority of construction sites subject to the turbidity limit would fall below the
associated thresholds several months before the end of the project. Table 9-4 presents the
original project durations based on NOIs and the duration used for costing purposes. For projects
shorter than 7 months, the duration of the project for costing purposes was not changed. For
projects longer than 7 months, the duration of the project was reduced by 1 month (for a 7-month
project) up to 6 months (for a 36-month project).

                             Table 9-4. Model project durations
NOI Project Duration (months)
Duration for Costing (months)
1
1
2
2
4
4
7
6
10
8
13
10

NOI Project Duration (months)
Duration for Costing (months)
16
13
19
15
22
16
27
21
32
26
36
30
(Tables 9-2 and 9-3 appear later in this section grouped with other tables of similar size.)

For costing ATS, EPA assumed that 100 percent of each construction site would be producing
stormwater runoff that would require treatment. That is a conservative assumption, because some
portion of sites will likely discharge through perimeter controls because diffuse runoff and would
not require sampling and compliance with the numeric limit. EPA notes that these assumptions
for project duration and amount of the site requiring treatment, while useful for modeling, likely
                                           9-2

-------
                                        Section 9: Estimating Incremental Costs for the Final Regulation
vary considerably among actual construction projects. However, because EPA lacks data on the
typical duration of disturbed soils at construction projects and the duration that treatment would
need to be in place, this modeling approach provides a reasonable, if somewhat conservative,
means of estimating runoff volumes requiring treatment and incremental compliance costs of the
options.

   9.2.2.   ESTIMATION OF RAINFALL DEPTHS AND STORAGE VOLUMES

To calculate basin sizing for storage for ATS under Options 2 and 3 and to  determine runoff
volumes for costing treatment systems, EPA evaluated several references to determine rainfall
depths for a series of design storm return periods for one indicator city in each state (for a
discussion of that analysis, see Section 3.5.3 and Appendix D). The storm depths for each
indicator city were used as point estimates for rainfall depths in each respective state. Using the
storm depths, EPA estimated runoff coefficients for each state using the process described in
TR-55 (USDA 1986). EPA estimated a runoff curve number using the Soil  Conservation Service
runoff curve number equation:
where
       Q = runoff (in)
       P = rainfall (in)
       S = potential maximum retention after runoff begins (in)
       Ia = initial abstraction (in) = 0.2S

         (P-0.2S)2
          (P + 0.8S)
         CN

Using the values contained in TR-55 for Curve Numbers for Developing Urban Areas (from TR-
55, Table 2-2a), along with the 2-year, 24-hour storm depths in Table 3-3, EPA calculated runoff
coefficients for the four hydrologic soil groups in each state for the 2-year, 24-hour storm event
(see Table 9-5). Using data on the prevalence of soils by hydrologic soil groups obtained from
STATSGO for each state (see  Table 3-4), EPA then calculated a weighted runoff coefficient for
each state for the 2-year, 24-hour storm event. These results are summarized in Table 9-5 and
were used to determine the required basin sizes for capturing runoff from the 2-year, 24-hour
storm event in each state, which is a requirement for Options 2 and 3 (see Table 9-6). Table 9-6
also contains baseline sediment basin sizes for states based on a review of current state permit
requirements. The rainfall analysis data is DCN 43095, and the STATSGO soils data evaluation
is DCN 43096  in the Administrative Record.
                                          9-3

-------
                                        Section 9: Estimating Incremental Costs for the Final Regulation
   9.2.3.   ESTIMATION OF ATS TREATMENT VOLUMES

For estimating average runoff volumes for each model site in each state, EPA multiplied the
average annual precipitation in each state by the site size and a runoff coefficient of 0.4. This
value of 0.4 was chosen as a reasonable estimate of the percent of average rainfall that would be
converted to runoff. EPA acknowledges that this approach likely overestimates runoff volumes
(and hence, volumes requiring treatment and associated treatment costs) because many smaller
storm events would not produce any, or very little, runoff. This approach also does not account
for precipitation that falls as snow. It also assumes that the entire site is in a newly graded areas
state, as defined by TR-55, for the entire duration of the project. In reality, site areas during
various states of the construction project would be in various states, ranging from bare soil, to
temporarily stabilized with mulch, to vegetated. As a result, curve numbers could be much lower
and associated runoff volumes could be much lower. However, this approach does allow for a
reasonable, albeit conservative, estimate of treatment volumes for determining compliance costs.
EPA multiplied the average annual runoff volumes for each model project by the project duration
(in years) account for length of time stormwater would be produced. This analysis does not
account for seasonal variations in rainfall—EPA assumed that precipitation would be evenly
distributed over the year. So, for a project that is less than one year in duration, the treatment
volume was based on the associated fraction of the year, irrespective of when construction would
actually occur.

Table 9-7 summarizes the monthly treatment volumes for each state for each of the 14 model
project size categories. Treatment volumes for each model project were determined by
multiplying the monthly treatment volumes by the project durations (in months). For each state
and each model project size category, EPA determined the treatment system flowrate needed to
treat runoff from the 2-year, 24-hour storm event within 72 hours. Table 9-8 shows the system
flowrate required. For selecting a design flowrate for costing, if the treatment flowrate was  100
gallons per minute (gpm) or less, a design flowrate of 100 gpm was selected. If the treatment
flowrate was greater than 100 gpm, the design flowarate was selected by  rounding up in 500-
gpm increments. Although ATS filtration systems are available in various flowrates depending
on the vendor, 500  gpm is a typical sand filter flowrate. Table 9-9 shows  the design flowrates
selected for costing purposes. EPA notes that rounding up to 500-gpm increments is a
conservative assumption, because a 500-gpm system could operate at a higher flowrate. So, for
example, EPA estimated that a 46-acre site in Illinois would require 525 gpm of treatment. A
500-gpm sand filter would likely be able to operate at 525 gpm, but EPA selected a 1,000-gpm
system for this model project. The 500-gpm system would have a lower rental cost than the
1,000-gpm system, but treatment times (and hence operator labor requirements) would be longer.
                                          9-4

-------
Table 9-2. Model project matrix
RESIDENTIAL
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
34
46
69
85
145
Total
Residential
Duration (days)
0-46
124
68
44
22
23
18
1
1
11
—
-
—
312
47-91
147
100
28
26
14
2
28
-
—
—
-
—
345
92-182
632
351
242
187
195
84
107
48
33
1
-
8
1,888
183-274
571
341
168
172
218
174
133
59
84
38
2
11
1,971
275-365
1,111
703
444
318
506
209
239
182
126
38
17
45
3,938
366-456
926
657
301
309
388
219
303
143
114
54
36
50
3,500
457-547
283
312
81
52
139
95
73
23
11
17
-
15
1,101
549-639
293
213
110
70
78
107
90
42
24
19
7
12
1,065
640-730
222
249
169
109
72
74
103
18
70
18
7
37
1,148
731-912
450
397
220
180
430
360
325
214
155
60
43
50
2,884
913-1,095
98
125
76
68
75
77
101
69
101
11
27
39
867
> 1 ,096
57
177
109
167
283
137
307
185
192
117
103
77
1,911
Total by
site size
4,914
3,693
1,992
1,680
2,421
1,556
1,810
984
921
373
242
344
20,930
NONRESIDENTIAL
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
34
46
69
85
145
Total
Nonresidential
Duration (days)
0-46
558
219
150
55
77
13
38
3
-
2
—
-
1,115
47-91
1,359
547
206
97
73
82
59
27
17
2
—
8
2,477
92-182
4,910
1,990
996
578
493
246
166
78
49
29
10
25
9,570
183-274
5,334
2,973
1,368
736
660
261
264
164
65
50
1
46
1 1 ,922
275-365
4,806
2,643
1,516
616
950
505
542
176
129
33
86
124
12,126
366-456
3,276
1,715
1,059
617
741
419
250
126
150
30
22
25
8,430
457-547
940
910
513
165
347
203
215
48
57
30
11
10
3,449
549-639
685
461
285
152
178
60
208
72
152
87
21
11
2,372
640-730
442
199
240
144
162
114
135
29
12
16
—
5
1,498
731-912
592
465
188
274
291
131
142
101
131
71
51
58
2,495
913-1,095
180
140
89
49
41
29
25
5
51
2
—
29
640
> 1 ,096
155
148
99
96
71
39
34
36
34
14
11
15
752
Total by
site size
23,237
12,410
6,709
3,579
4,084
2,102
2,078
865
847
366
213
356
56,846

-------
CD
05
TRANSPORTATION
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
34
46
69
85
145
Total
Transportation
NATIONAL
TOTAL
Duration (days)
0-46
82
25
16
7
8
-
5
—
-
-
—
—
143
1,570
47-91
210
87
23
15
21
3
2
—
1
-
—
—
362
3,184
92-182
629
277
184
70
70
63
31
1
2
-
7
—
1,334
12,792
183-274
418
255
138
73
109
49
36
26
19
-
—
1
1,124
15,017
275-365
308
246
170
78
39
13
6
17
9
3
—
3
892
16,956
366-456
323
146
53
78
95
15
76
6
25
9
3
4
833
12,763
457-547
136
81
45
33
52
21
40
16
21
10
—
1
456
5,006
549-639
55
60
26
89
43
2
27
4
15
4
—
9
334
3,771
640-730
132
34
-
17
17
8
14
9
9
-
2
11
253
2,899
731-912
117
58
78
7
36
64
95
5
24
5
4
24
517
5,896
913-1,095
7
22
15
21
13
6
16
42
4
21
12
35
214
1,721
> 1 ,096
—
7
22
6
45
28
15
2
51
-
28
30
234
2,897
Total by
site size
2,417
1,298
770
494
548
272
363
128
180
52
56
118
6,696
84,472

-------
Table 9-3. Acreage developed matrix
RESIDENTIAL
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
34
46
69
85
145
Total
Residential
Duration (days)
0-46
236
258
264
187
276
306
23
34
506
—
-
—
2,090
47-91
279
380
168
221
168
34
644
-
—
—
-
—
1,894
92-182
1,201
1,334
1,452
1,590
2,340
1,428
2,461
1,632
1,518
69
-
1,160
16,184
183-274
1,085
1,296
1,008
1,462
2,616
2,958
3,059
2,006
3,864
2,622
170
1,595
23,741
275-365
2,111
2,671
2,664
2,703
6,072
3,553
5,497
6,188
5,796
2,622
1,447
6,525
47,849
366-456
1,759
2,497
1,806
2,627
4,656
3,723
6,969
4,862
5,244
3,726
3,064
7,250
48,182
457-547
538
1,186
486
442
1,668
1,615
1,679
782
506
1,173
-
2,175
12,249
549-639
557
809
660
595
936
1,819
2,070
1,428
1,104
1,311
596
1,740
13,625
640-730
422
946
1,014
927
864
1,258
2,369
612
3,220
1,242
596
5,365
18,834
731-912
855
1,509
1,320
1,530
5,160
6,120
7,475
7,276
7,130
4,140
3,659
7,250
53,424
913-1,095
186
475
456
578
900
1,309
2,323
2,346
4,646
759
2,298
5,655
21,931
> 1 ,096
108
673
654
1,420
3,396
2,329
7,061
6,290
8,832
8,073
8,765
11,165
58,766
Total by
site size
9,337
14,033
1 1 ,952
14,280
29,052
26,452
41,630
33,456
42,366
25,737
20,594
49,880
318,769
NONRESIDENTIAL
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
34
46
69
85
145
Total
Nonresidential
Duration (days)
0-46
1,060
832
900
468
924
221
874
102
-
138
—
-
5,519
47-91
2,582
2,079
1,236
825
876
1,394
1,357
918
782
138
—
1,160
13,346
92-182
9,329
7,562
5,976
4,913
5,916
4,182
3,818
2,652
2,254
2,001
851
3,625
53,079
183-274
10,135
1 1 ,297
8,208
6,256
7,920
4,437
6,072
5,576
2,990
3,450
85
6,670
73,096
275-365
9,131
10,043
9,096
5,236
1 1 ,400
8,585
12,466
5,984
5,934
2,277
7,319
17,980
105,451
366-456
6,224
6,517
6,354
5,245
8,892
7,123
5,750
4,284
6,900
2,070
1,872
3,625
64,856
457-547
1,786
3,458
3,078
1,403
4,164
3,451
4,945
1,632
2,622
2,070
936
1,450
30,995
549-639
1,302
1,752
1,710
1,292
2,136
1,020
4,784
2,448
6,992
6,003
1,787
1,595
32,820
640-730
840
756
1,440
1,224
1,944
1,938
3,105
986
552
1,104
—
725
14,614
731-912
1,125
1,767
1,128
2,329
3,492
2,227
3,266
3,434
6,026
4,899
4,340
8,410
42,443
913-1,095
342
532
534
417
492
493
575
170
2,346
138
—
4,205
10,244
> 1 ,096
295
562
594
816
852
663
782
1,224
1,564
966
936
2,175
1 1 ,429
Total by
site size
44,150
47,158
40,254
30,422
49,008
35,734
47,794
29,410
38,962
25,254
18,126
51,620
457,892

-------
CD
CO
TRANSPORTATION
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
34
46
69
85
145
Total
Transportation
NATIONAL
TOTAL
Duration (days)
0-46
156
95
96
60
96
—
115
-
-
—
—
-
617
8,226
47-91
399
331
138
128
252
51
46
-
46
—
—
-
1,390
16,631
92-182
1,195
1,053
1,104
595
840
1,071
713
34
92
—
596
-
7,292
76,556
183-274
794
969
828
621
1,308
833
828
884
874
—
—
145
8,084
104,921
275-365
585
935
1,020
663
468
221
138
578
414
207
—
435
5,664
158,964
366-456
614
555
318
663
1,140
255
1,748
204
1,150
621
255
580
8,103
121,141
457-547
258
308
270
281
624
357
920
544
966
690
—
145
5,363
48,607
549-639
105
228
156
757
516
34
621
136
690
276
—
1,305
4,823
51,268
640-730
251
129
—
145
204
136
322
306
414
—
170
1,595
3,672
37,120
731-
912
222
220
468
60
432
1,088
2,185
170
1,104
345
340
3,480
10,115
105,981
913-1,095
13
84
90
179
156
102
368
1,428
184
1,449
1,021
5,075
10,149
42,323
> 1 ,096
-
27
132
51
540
476
345
68
2,346
—
2,383
4,350
10,717
80,912
Total by
site size
4,592
4,932
4,620
4,199
6,576
4,624
8,349
4,352
8,280
3,588
4,766
17,110
75,988
852,650

-------
Table 9-5. State runoff coefficients for 2-year, 24-hour storm events

Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
2-year,
24-hour
storm depth
(inches)
4.50
1.40
4.10
2.00
2.00
3.10
3.26
4.75
3.70
1.20
2.85
2.95
3.25
3.50
3.00
5.25
2.80
3.16
3.10
2.40
2.75
4.45
3.45
1.30
3.00
1.00
2.80
3.31
1.54
2.90
Runoff
coefficient
for A soils
0.49
0.12
0.46
0.22
0.22
0.37
0.38
0.51
0.43
0.08
0.34
0.35
0.38
0.41
0.36
0.54
0.33
0.37
0.37
0.28
0.33
0.49
0.40
0.10
0.36
0.05
0.33
0.39
0.15
0.35
Runoff
coefficient
for B soils
0.67
0.31
0.64
0.42
0.42
0.56
0.58
0.68
0.62
0.25
0.54
0.55
0.58
0.60
0.55
0.70
0.53
0.57
0.56
0.48
0.53
0.66
0.60
0.28
0.55
0.20
0.53
0.58
0.34
0.54
Runoff
coefficient
for C soils
0.78
0.47
0.76
0.58
0.58
0.70
0.71
0.79
0.74
0.42
0.68
0.69
0.71
0.73
0.69
0.80
0.67
0.70
0.70
0.63
0.67
0.78
0.72
0.45
0.69
0.36
0.67
0.71
0.50
0.68
Runoff
coefficient
for D soils
0.85
0.61
0.83
0.70
0.70
0.79
0.80
0.86
0.82
0.56
0.77
0.78
0.80
0.81
0.78
0.87
0.77
0.79
0.79
0.74
0.77
0.85
0.81
0.58
0.78
0.50
0.77
0.80
0.63
0.78
%A
soils
8.7%
4.7%
0.6%
10.9%
7.2%
9.1%
20.8%
18.1%
6.6%
4.4%
1 .4%
3.5%
0.9%
3.8%
0.1%
1 .7%
7.7%
10.0%
23.9%
29.0%
8.3%
2.3%
1 .0%
2.9%
31 .9%
5.6%
17.1%
12.5%
5.6%
9.6%
%B
soils
41 .2%
38.6%
28.3%
32.2%
46.7%
41.1%
30.9%
6.3%
53.1%
46.8%
44.5%
32.6%
66.0%
58.0%
42.7%
14.4%
12.9%
38.6%
16.6%
28.7%
37.4%
32.3%
40.1%
39.5%
53.6%
26.4%
24.8%
32.8%
41 .9%
18.5%
%C
soils
28.8%
17.2%
35.9%
18.4%
24.6%
35.9%
13.4%
8.6%
16.9%
23.1%
27.0%
41 .8%
1 1 .6%
19.5%
44.9%
28.9%
43.9%
26.4%
34.4%
12.9%
15.4%
38.6%
39.8%
27.2%
3.0%
17.7%
41 .4%
25.1%
16.5%
51.1%
%D
soils
21 .3%
39.5%
35.1%
38.5%
21 .4%
13.9%
34.9%
67.0%
23.5%
25.7%
27.1%
22.1%
21 .5%
18.7%
12.3%
55.1%
35.5%
25.0%
25.2%
29.4%
38.9%
26.9%
19.0%
30.4%
1 1 .5%
50.3%
16.6%
29.6%
36.0%
20.7%
Weighted runoff
coefficient 2-year,
24-hour storm
0.72
0.44
0.75
0.54
0.51
0.63
0.63
0.78
0.67
0.36
0.64
0.65
0.64
0.66
0.64
0.82
0.66
0.64
0.62
0.52
0.63
0.75
0.68
0.41
0.52
0.37
0.60
0.66
0.46
0.64

-------
CD




O

North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
District of
Columbia
2-year,
24-hour
storm depth
(inches)
3.34
1.90
2.62
3.70
2.50
3.23
3.20
3.62
2.25
3.37
3.90
1.40
2.40
3.11
2.00
2.56
2.80
1.60
3.16
Runoff
coefficient
for A soils
0.39
0.21
0.31
0.43
0.30
0.38
0.38
0.42
0.26
0.40
0.44
0.12
0.28
0.37
0.22
0.30
0.33
0.16
0.37
Runoff
coefficient
for B soils
0.59
0.41
0.51
0.62
0.50
0.58
0.57
0.61
0.46
0.59
0.63
0.31
0.48
0.56
0.42
0.50
0.53
0.35
0.57
Runoff
coefficient
for C soils
0.72
0.57
0.66
0.74
0.64
0.71
0.71
0.73
0.62
0.72
0.75
0.47
0.63
0.70
0.58
0.65
0.67
0.51
0.70
Runoff
coefficient
for D soils
0.80
0.69
0.76
0.82
0.75
0.80
0.79
0.82
0.73
0.80
0.83
0.61
0.74
0.79
0.70
0.75
0.77
0.64
0.79
%A
soils
7.9%
4.7%
0.6%
6.8%
5.2%
6.0%
15.3%
1 1 .9%
2.9%
0.1%
5.1%
5.3%
4.9%
1 .7%
6.6%
7.3%
14.4%
4.5%
10.0%
%B
soils
48.8%
56.1%
16.8%
44.5%
32.1%
28.4%
35.7%
41 .8%
45.2%
53.6%
27.2%
36.2%
18.0%
53.7%
53.4%
21 .5%
46.8%
40.5%
38.6%
%C
soils
16.5%
16.6%
54.6%
22.3%
37.1%
54.2%
32.4%
19.5%
1 1 .5%
30.4%
24.5%
16.2%
54.3%
32.3%
24.2%
54.2%
18.1%
19.5%
26.4%
%D
soils
26.8%
22.6%
28.0%
26.4%
25.6%
1 1 .5%
16.5%
26.8%
40.4%
15.9%
43.2%
42.3%
22.8%
12.3%
15.8%
17.0%
20.7%
35.5%
25.0%
Weighted runoff
coefficient 2-year,
24-hour storm
0.65
0.49
0.66
0.68
0.61
0.66
0.62
0.67
0.58
0.66
0.73
0.45
0.61
0.63
0.49
0.61
0.58
0.48
0.64

-------
                   Section 9: Estimating Incremental Costs for the Final Regulation
Table 9-6. ATS storage requirements for states

Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Baseline sediment
basin size
(cf/acre)
1,800
2,254
3,600
3,389
1,800
3,600
3,600
3,600
1,800
3,600
1,585
3,600
1,800
3,600
3,600
3,600
3,600
3,600
3,600
3,600
3,600
3,600
3,600
3,600
1,800
1,800
1,350
3,600
1,800
2,570
3,600
1,800
3,361
1,800
3,600
3,600
5,000
1,800
3,600
3,600
3,600
3,600
3,099
1,800
2-year, 24-hour runoff
basin size
(cf/acre)
11,798
2,254
11,179
3,898
3,685
7,043
7,487
13,371
9,019
1,585
6,591
6,968
7,536
8,337
7,004
15,638
6,744
7,353
6,977
4,530
6,247
12,147
8,574
1,948
5,680
1,350
6,064
7,886
2,570
6,776
7,884
3,361
6,261
9,184
5,495
7,754
7,239
8,756
4,745
8,095
10,403
2,284
5,345
7,141
                     9-11

-------
Section 9: Estimating Incremental Costs for the Final Regulation

Washington
West Virginia
Wisconsin
Wyoming
District of Columbia
Baseline sediment
basin size
(cf/acre)
3,600
512
3,600
1,800
1,800
2-year, 24-hour runoff
basin size
(cf/acre)
3,575
5,679
5,884
2,770
7,353
  9-12

-------
Table 9-7. Monthly ATS treatment volumes (gallons)
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
Model site size (acres)
1.9
84,331
13,309
81,888
31,628
22,885
76,378
73,984
106,642
87,456
31,452
19,491
56,824
68,246
54,449
63,087
77,407
101,070
71,652
72,714
72,921
52,076
49,451
90,244
63,161
20,256
48,692
6,567
69,179
80,817
26,580
63,472
74,271
3.8
168,662
26,617
163,775
63,256
45,770
152,755
147,968
213,284
174,911
62,905
38,983
113,648
136,492
108,899
126,174
154,815
202,140
143,304
145,427
145,842
104,152
98,902
180,489
126,323
40,513
97,383
13,134
138,357
161,633
53,160
126,943
148,541
6
266,308
42,028
258,593
99,878
72,269
241,193
233,634
336,764
276,176
99,323
61,552
179,444
215,514
171,945
199,222
244,445
319,168
226,269
229,622
230,277
164,451
156,161
284,982
199,457
63,968
153,763
20,738
218,459
255,210
83,937
200,437
234,539
8.5
377,270
59,539
366,340
141,494
102,381
341,690
330,981
477,083
391,249
140,708
87,198
254,213
305,312
243,589
282,231
346,297
452,155
320,548
325,298
326,226
232,972
221 ,228
403,724
282,564
90,621
217,831
29,378
309,483
361,548
118,911
283,953
332,263
12
532,617
84,055
517,185
199,756
144,537
482,385
467,267
673,528
552,352
198,647
123,104
358,888
431,028
343,891
398,443
488,889
638,336
452,538
459,245
460,554
328,902
312,322
569,964
398,914
127,935
307,526
41,475
436,917
510,421
167,875
400,874
469,077
17
754,540
119,078
732,679
282,988
204,761
683,379
661 ,962
954,165
782,499
281,417
174,397
508,425
610,623
487,178
564,462
692,593
904,309
641,095
650,597
652,451
465,944
442,456
807,449
565,129
181,242
435,661
58,757
618,966
723,096
237,823
567,905
664,526
23
1,020,849
161,106
991,272
382,866
277,030
924,572
895,596
1,290,930
1,058,674
380,740
235,949
687,869
826,137
659,124
763,683
937,038
1,223,478
867,364
880,219
882,728
630,395
598,617
1,092,431
764,586
245,209
589,424
79,495
837,425
978,307
321,760
768,342
899,065
34
1,509,081
238,156
1,465,359
565,976
409,522
1,366,758
1,323,924
1,908,331
1,564,997
562,833
348,794
1,016,850
1 ,221 ,247
974,357
1,128,923
1,385,186
1,808,619
1,282,190
1,301,193
1,304,902
931,888
884,912
1,614,898
1,130,257
362,484
871,323
117,514
1,237,932
1,446,193
475,645
1,135,810
1,329,052
46
2,041,697
322,212
1,982,544
765,732
554,060
1,849,143
1,791,191
2,581,859
2,117,349
761,480
471,897
1,375,739
1,652,275
1,318,248
1,527,367
1,874,076
2,446,955
1,734,728
1,760,438
1,765,456
1,260,789
1,197,234
2,184,862
1,529,172
490,419
1,178,848
158,989
1,674,849
1,956,613
643,520
1,536,684
1,798,129
69
3,062,546
483,317
2,973,816
1,148,598
831,089
2,773,715
2,686,787
3,872,789
3,176,023
1,142,220
707,846
2,063,608
2,478,412
1,977,371
2,291,050
2,811,113
3,670,433
2,602,092
2,640,657
2,648,184
1,891,184
1,795,850
3,277,293
2,293,758
735,628
1,768,273
238,484
2,512,274
2,934,920
965,281
2,305,027
2,697,194
85
3,777,140
596,091
3,667,707
1,416,604
1,025,010
3,420,915
3,313,704
4,776,440
3,917,096
1,408,738
873,010
2,545,117
3,056,709
2,438,758
2,825,628
3,467,040
4,526,867
3,209,247
3,256,810
3,266,093
2,332,460
2,214,882
4,041,994
2,828,968
907,275
2,180,870
294,130
3,098,471
3,619,735
1,190,513
2,842,866
3,326,539
145
6,435,785
1,015,667
6,249,324
2,413,720
1 ,746,492
5,828,821
5,646,147
8,138,469
6,674,252
2,400,317
1,487,503
4,336,568
5,208,258
4,155,346
4,814,525
5,907,412
7,713,228
5,468,165
5,549,206
5,565,024
3,974,228
3,773,888
6,887,064
4,820,216
1,545,886
3,715,935
501,161
5,279,416
6,167,586
2,028,488
4,843,896
5,668,016

-------
CD
State
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Puerto Rico
District of Columbia
Model site size (acres)
1.9
27,663
64,983
56,430
70,026
71,565
76,941
78,161
27,679
79,292
55,927
25,034
58,475
69,392
60,638
73,641
54,119
25,340
87,290
71,780
3.8
55,327
129,967
112,861
140,052
143,130
153,883
156,322
55,358
158,584
111,855
50,068
116,950
138,784
121,275
147,283
108,239
50,679
174,579
143,560
6
87,358
205,211
178,201
221,135
225,994
242,973
246,824
87,407
250,396
176,613
79,055
184,657
219,132
191,487
232,552
170,903
80,020
275,651
226,673
8.5
123,757
290,715
252,451
313,274
320,158
344,211
349,668
123,826
354,727
250,201
1 1 1 ,994
261,598
310,438
271,273
329,449
242,113
113,361
390,506
321,120
12
174,715
410,421
356,402
442,270
451,988
485,945
493,649
174,813
500,791
353,225
158,110
369,315
438,265
382,974
465,104
341,807
160,040
551,302
453,346
17
247,513
581,430
504,903
626,549
640,316
688,423
699,336
247,652
709,454
500,403
223,989
523,196
620,875
542,546
658,897
484,226
226,723
781,012
642,240
23
334,871
786,641
683,104
847,684
866,310
931,395
946,160
335,059
959,850
677,015
303,043
707,854
840,007
734,033
891,449
655,130
306,743
1,056,663
868,913
34
495,027
1,162,860
1,009,805
1,253,098
1,280,633
1,376,845
1,398,672
495,305
1,418,908
1,000,805
447,977
1,046,392
1,241,750
1,085,092
1,317,794
968,453
453,446
1,562,023
1,284,481
46
669,742
1,573,282
1,366,207
1,695,368
1,732,621
1,862,790
1,892,321
670,118
1,919,700
1,354,030
606,087
1,415,707
1,680,015
1 ,468,066
1,782,898
1,310,260
613,485
2,113,326
1,737,827
69
1,004,613
2,359,922
2,049,311
2,543,051
2,598,931
2,794,185
2,838,481
1,005,177
2,879,549
2,031,046
909,130
2,123,561
2,520,022
2,202,099
2,674,347
1,965,390
920,228
3,169,988
2,606,740
85
1,239,023
2,910,571
2,527,484
3,136,430
3,205,348
3,446,162
3,500,793
1,239,718
3,551,444
2,504,956
1,121,261
2,619,058
3,108,027
2,715,922
3,298,361
2,423,981
1,134,948
3,909,652
3,214,979
145
2,111,144
4,959,257
4,306,523
5,344,094
5,461,521
5,871,839
5,964,924
2,112,329
6,051,227
4,268,139
1,910,491
4,462,555
5,295,699
4,627,600
5,620,004
4,130,167
1,933,812
6,661,570
5,477,932

-------
Table 9-8. ATS system flowrate required (gpm)
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
Model site size (acres)
1.9
39
7
37
13
12
23
25
44
30
42
5
22
23
25
27
23
51
22
24
23
15
21
40
28
6
19
4
20
26
8
22
26
3.8
78
15
74
26
24
46
49
88
59
85
10
43
46
50
55
46
103
44
48
46
30
41
80
56
13
37
9
40
52
17
45
52
6
123
23
116
40
38
73
78
139
94
134
16
68
72
78
87
73
162
70
76
72
47
65
126
89
20
59
14
63
82
27
70
82
8.5
174
33
165
57
54
104
110
197
133
189
23
97
103
111
123
103
230
99
108
103
67
92
179
126
29
84
20
89
116
38
100
116
12
245
47
232
81
77
146
156
278
187
267
33
137
145
157
173
146
325
140
153
145
94
130
252
178
40
118
28
126
164
53
141
164
17
347
66
329
115
108
207
220
394
265
379
47
194
205
222
245
206
460
199
216
205
133
184
358
252
57
167
40
178
232
76
199
232
23
470
90
445
155
147
280
298
532
359
512
63
262
277
300
332
279
623
269
293
278
180
249
484
341
78
226
54
241
314
102
270
314
34
695
133
658
229
217
415
441
787
531
758
93
388
410
444
491
412
921
397
433
411
267
368
715
505
115
334
79
357
464
151
399
464
46
940
180
890
310
294
561
596
1,065
718
1,025
126
525
555
600
664
558
1,246
537
586
556
361
498
967
683
155
452
108
483
628
205
540
628
69
1,410
269
1,336
466
440
841
894
1,597
1,078
1,537
189
787
832
900
996
837
1,868
806
878
834
541
746
1,451
1,024
233
679
161
724
942
307
810
942
85
1,738
332
1,647
574
543
1,038
1,103
1,970
1,329
1,896
234
971
1,027
1,110
1,228
1,032
2,304
994
1,083
1,028
667
920
1,790
1,263
287
837
199
894
1,162
379
998
1,162
145
2,962
566
2,807
979
925
1,768
1,880
3,357
2,264
3,231
398
1,655
1,749
1,892
2,093
1,758
3,926
1,693
1,846
1,752
1,137
1,568
3,050
2,153
489
1,426
339
1,522
1,980
645
1,701
1,979

-------
CD
State
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Puerto Rico
District of Columbia
Model site size (acres)
1.9
11
21
30
18
26
24
29
16
27
34
8
18
23
12
19
19
9
37
24
3.8
22
41
60
36
51
48
58
31
53
68
15
35
47
24
37
39
18
74
48
6
35
65
95
57
81
75
91
49
84
108
24
56
74
37
59
61
29
116
76
8.5
49
92
135
81
114
107
129
70
119
153
34
79
105
53
84
87
41
165
108
12
70
130
191
114
161
150
182
99
168
216
47
111
148
74
118
122
58
233
153
17
99
184
270
162
228
213
258
140
238
306
67
157
210
105
167
173
82
329
216
23
134
249
366
219
309
288
349
189
322
414
91
213
284
142
226
234
110
446
293
34
198
369
541
323
456
426
515
279
477
612
134
315
420
210
334
346
163
659
433
46
268
499
731
438
618
577
697
378
645
829
182
426
569
285
452
469
221
891
586
69
402
748
1,097
656
926
865
1,046
567
967
1,243
273
639
853
427
678
703
331
1,337
878
85
495
923
1,353
810
1,143
1,067
1,290
699
1,193
1,533
337
788
1,052
527
837
867
408
1,649
1,083
145
844
1,572
2,306
1,380
1,947
1,817
2,198
1,191
2,032
2,612
573
1,342
1,793
898
1,426
1,477
695
2,809
1,846

-------
Table 9-9. ATS system flowrate selected for costing (gpm)
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
Model site size (acres)
1.9
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
3.8
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
500
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
6
500
100
500
100
100
100
100
500
100
500
100
100
100
100
100
100
500
100
100
100
100
100
500
100
100
100
100
100
100
100
100
100
8.5
500
100
500
100
100
500
500
500
500
500
100
100
500
500
500
500
500
100
500
500
100
100
500
500
100
100
100
100
500
100
100
500
12
500
100
500
100
100
500
500
500
500
500
100
500
500
500
500
500
500
500
500
500
100
500
500
500
100
500
100
500
500
100
500
500
17
500
100
500
500
500
500
500
500
500
500
100
500
500
500
500
500
500
500
500
500
500
500
500
500
100
500
100
500
500
100
500
500
23
500
100
500
500
500
500
500
1,000
500
1,000
100
500
500
500
500
500
1,000
500
500
500
500
500
500
500
100
500
100
500
500
500
500
500
34
1,000
500
1,000
500
500
500
500
1,000
1,000
1,000
100
500
500
500
500
500
1,000
500
500
500
500
500
1,000
1,000
500
500
100
500
500
500
500
500
46
1,000
500
1,000
500
500
1,000
1,000
1,500
1,000
1,500
500
1,000
1,000
1,000
1,000
1,000
1,500
1,000
1,000
1,000
500
500
1,000
1,000
500
500
500
500
1,000
500
1,000
1,000
69
1,500
500
1,500
500
500
1,000
1,000
2,000
1,500
2,000
500
1,000
1,000
1,000
1,000
1,000
2,000
1,000
1,000
1,000
1,000
1,000
1,500
1,500
500
1,000
500
1,000
1,000
500
1,000
1,000
85
2,000
500
2,000
1,000
1,000
1,500
1,500
2,000
1,500
2,000
500
1,000
1,500
1,500
1,500
1,500
2,500
1,000
1,500
1,500
1,000
1,000
2,000
1,500
500
1,000
500
1,000
1,500
500
1,000
1,500
145
3,000
1,000
3,000
1,000
1,000
2,000
2,000
3,500
2,500
3,500
500
2,000
2,000
2,000
2,500
2,000
4,000
2,000
2,000
2,000
1,500
2,000
3,500
2,500
500
1,500
500
2,000
2,000
1,000
2,000
2,000

-------
CD




OO
State
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Puerto Rico
District of Columbia
Model site size (acres)
1.9
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
3.8
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
6
100
100
100
100
100
100
100
100
100
500
100
100
100
100
100
100
100
500
100
8.5
100
100
500
100
500
500
500
100
500
500
100
100
500
100
100
100
100
500
500
12
100
500
500
500
500
500
500
100
500
500
100
500
500
100
500
500
100
500
500
17
100
500
500
500
500
500
500
500
500
500
100
500
500
500
500
500
100
500
500
23
500
500
500
500
500
500
500
500
500
500
100
500
500
500
500
500
500
500
500
34
500
500
1,000
500
500
500
1,000
500
500
1,000
500
500
500
500
500
500
500
1,000
500
46
500
500
1,000
500
1,000
1,000
1,000
500
1,000
1,000
500
500
1,000
500
500
500
500
1,000
1,000
69
500
1,000
1,500
1,000
1,000
1,000
1,500
1,000
1,000
1,500
500
1,000
1,000
500
1,000
1,000
500
1,500
1,000
85
500
1,000
1,500
1,000
1,500
1,500
1,500
1,000
1,500
2,000
500
1,000
1,500
1,000
1,000
1,000
500
2,000
1,500
145
1,000
2,000
2,500
1,500
2,000
2,000
2,500
1,500
2,500
3,000
1,000
1,500
2,000
1,000
1,500
1,500
1,000
3,000
2,000

-------
                                        Section 9: Estimating Incremental Costs for the Final Regulation
 9.3.    ESTIMATION OF COSTS

EPA estimated costs for the regulatory options using three categories of costs: erosion and
sediment controls (ESCs), ATS storage and treatment costs, and passive treatment costs. The
components of the estimated costs are discussed below.

   9.3.1.   EROSION AND SEDIMENT CONTROL COSTS

EPA estimated costs for ESCs under baseline conditions as well as incremental costs as a result
of the regulatory options. Estimating baseline costs was necessary to conduct the BCT cost test.
While a variety of controls are likely to be used on individual construction sites, EPA does not
have any comprehensive data on current practices because a survey of the industry was not
conducted.  Therefore, EPA made assumptions about controls employed for each of the model
projects as a way of estimating baseline industry costs. EPA assumes that baseline controls
would consist of sediment basins (for sites greater than 10 acres), silt fence, and soil cover.

For estimating basin unit costs, EPA determined existing state sediment basin sizing
requirements on the basis of a review of state permits. Table  9-6 shows the baseline sediment
basins sizing requirements  for each state. Table 9-11 shows the corresponding baseline basin size
for each the model projecs  greater than 10 acres. Descriptions of sediment traps and basins,
including design criteria, performance, and costs are described in detail in Section 7.2.3.

EPA (1993) references the  Southeastern Wisconsin Regional Planning Commission's 1991 Costs
of Urban Nonpoint Source  Water Pollution Control Measures (SWRPC  1991) for estimating
costs of temporary sediment basins. Costs include site preparation (e.g., grading, excavation,
place and compact fill), site development (e.g., riprap, temporary basin inlet and outlet
structures), and contingencies. Temporary basins are a generally less expensive  option as
compared to permanent basins (e.g., SWRPC assumes temporary basin inlet and outlet costs to
be one-half of permanent detention basin inlet and outlet costs). Table 9-10 summarizes the cost
data EPA used for estimating sediment basin costs. EPA used the average value of $0.30 per
cubic foot of storage for calculating incremental costs for sediment basins and for storage
volumes required for ATS. EPA adjusted all sediment basin costs from 1989 dollars to 2008
dollars.
                     Table 9-10. Sediment basin construction cost data
Cost data source
USEPA(1993), original reference is
SWRPC (1991). Numerous sources
reference this data. Many of these
sources adjusted USEPA (1 993) to
other basis years.
Cost
$0.10 to $0.40 per cubic foot of
storage, average of $0.30.
Basin size range of
validity
12,000 ft3 to 195,000
ft3 SWRPC (1991)
Basis year
1989
                                          9-19

-------
                                        Section 9: Estimating Incremental Costs for the Final Regulation
EPA (1993) estimates annual operation and maintenance costs for temporary sediment basins
(associated with runoff from active construction sites) as 25 percent of construction costs. EPA
used this value to estimate costs for sediment basins.

Capital and annual cost data for basins were standardized to 2008 dollars on the basis of the
Engineering News-Record (ENR) Construction Cost Index (CCI). The sediment basin cost
reference used by EPA was  based on a study completed in  1989. The ENR CCI in 1989 was
4,615, and the value in February of 2008 was 8,084. EPA adjusted sediment basin costs obtained
in 1989 dollars to 2008 dollars by increasing costs by 57 percent (8,084 / 4,615 = 1.57). All other
data was obtained in 2008 and 2009, so no standardization  was necessary to arrive at year 2008
costs. Table 9-11 shows the baseline sediment basin size, and Table 9-12 shows the baseline
sediment basin costs for each state and the District of Columbia for each of the model site sizes
greater than 10 acres.

                     Table 9-11. Baseline sediment basin  size (cubic feet)
State
Alabama
Arizona
Alaska
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New
Hampshire
New Jersey
New Mexico
Site size (acres)
12
21,600
27,048
6,144
43,200
40,662
21,600
43,200
43,200
43,200
21,600
43,200
19,020
43,200
21,600
43,200
43,200
43,200
43,200
43,200
43,200
43,200
43,200
43,200
43,200
43,200
21,600
21,600
16,200
43,200
21,600
30,840
17
30,600
38,318
8,704
61,200
57,605
30,600
61,200
61,200
61,200
30,600
61,200
26,945
61,200
30,600
61,200
61,200
61,200
61,200
61,200
61,200
61,200
61,200
61,200
61,200
61,200
30,600
30,600
22,950
61,200
30,600
43,690
23
41,400
51,842
1 1 ,776
82,800
77,936
41,400
82,800
82,800
82,800
41,400
82,800
36,455
82,800
41,400
82,800
82,800
82,800
82,800
82,800
82,800
82,800
82,800
82,800
82,800
82,800
41,400
41,400
31,050
82,800
41,400
59,110
34
61,200
76,636
17,408
122,400
115,209
61,200
122,400
122,400
122,400
61,200
122,400
53,890
122,400
61,200
122,400
122,400
122,400
122,400
122,400
122,400
122,400
122,400
122,400
122,400
122,400
61,200
61,200
45,900
122,400
61,200
87,380
46
82,800
103,684
23,552
165,600
155,872
82,800
165,600
165,600
165,600
82,800
165,600
72,910
165,600
82,800
165,600
165,600
165,600
165,600
165,600
165,600
165,600
165,600
165,600
165,600
165,600
82,800
82,800
62,100
165,600
82,800
118,220
69
124,200
155,526
35,328
248,400
233,807
124,200
248,400
248,400
248,400
124,200
248,400
109,365
248,400
124,200
248,400
248,400
248,400
248,400
248,400
248,400
248,400
248,400
248,400
248,400
248,400
124,200
124,200
93,150
248,400
124,200
177,330
85
153,180
191,815
43,571
306,360
288,362
153,180
306,360
306,360
306,360
153,180
306,360
134,884
306,360
153,180
306,360
306,360
306,360
306,360
306,360
306,360
306,360
306,360
306,360
306,360
306,360
153,180
153,180
114,885
306,360
153,180
218,707
145
261,000
326,830
74,240
522,000
491,334
261,000
522,000
522,000
522,000
261,000
522,000
229,825
522,000
261,000
522,000
522,000
522,000
522,000
522,000
522,000
522,000
522,000
522,000
522,000
522,000
261,000
261,000
195,750
522,000
261,000
372,650
                                          9-20

-------
                 Section 9: Estimating Incremental Costs for the Final Regulation
State
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
District of
Columbia
Site size (acres)
12
43,200
21,600
40,332
21,600
43,200
43,200
60,000
21,600
43,200
43,200
43,200
43,200
37,188
21,600
43,200
6,144
43,200
21,600
21,600
21,600
17
61,200
30,600
57,137
30,600
61,200
61,200
85,000
30,600
61,200
61,200
61,200
61,200
52,683
30,600
61,200
8,704
61,200
30,600
30,600
30,600
23
82,800
41,400
77,303
41,400
82,800
82,800
115,000
41,400
82,800
82,800
82,800
82,800
71,277
41,400
82,800
1 1 ,776
82,800
41,400
41,400
41,400
34
122,400
61,200
114,274
61,200
122,400
122,400
170,000
61,200
122,400
122,400
122,400
122,400
105,366
61,200
122,400
17,408
122,400
61,200
61,200
61,200
46
165,600
82,800
154,606
82,800
165,600
165,600
230,000
82,800
165,600
165,600
165,600
165,600
142,554
82,800
165,600
23,552
165,600
82,800
82,800
82,800
69
248,400
124,200
231,909
124,200
248,400
248,400
345,000
124,200
248,400
248,400
248,400
248,400
213,831
124,200
248,400
35,328
248,400
124,200
124,200
124,200
85
306,360
153,180
286,021
153,180
306,360
306,360
425,500
153,180
306,360
306,360
306,360
306,360
263,725
153,180
306,360
43,571
306,360
153,180
153,180
153,180
145
522,000
261,000
487,345
261,000
522,000
522,000
725,000
261,000
522,000
522,000
522,000
522,000
449,355
261,000
522,000
74,240
522,000
261,000
261,000
261,000
Table 9-12. Baseline sediment basins costs
State
Alabama
Arizona
Alaska
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Site size (acres)
12
$14,207
$17,790
$4,041
$28,413
$26,744
$14,207
$28,413
$28,413
$28,413
$14,207
$28,413
$12,510
$28,413
$14,207
$28,413
$28,413
$28,413
$28,413
$28,413
$28,413
$28,413
$28,413
17
$20,126
$25,202
$5,725
$40,252
$37,887
$20,126
$40,252
$40,252
$40,252
$20,126
$40,252
$17,722
$40,252
$20,126
$40,252
$40,252
$40,252
$40,252
$40,252
$40,252
$40,252
$40,252
23
$27,229
$34,097
$7,745
$54,459
$51,260
$27,229
$54,459
$54,459
$54,459
$27,229
$54,459
$23,977
$54,459
$27,229
$54,459
$54,459
$54,459
$54,459
$54,459
$54,459
$54,459
$54,459
34
$40,252
$50,405
$11,450
$80,504
$75,775
$40,252
$80,504
$80,504
$80,504
$40,252
$80,504
$35,444
$80,504
$40,252
$80,504
$80,504
$80,504
$80,504
$80,504
$80,504
$80,504
$80,504
46
$54,459
$68,195
$15,491
$108,918
$102,519
$54,459
$108,918
$108,918
$108,918
$54,459
$108,918
$47,954
$108,918
$54,459
$108,918
$108,918
$108,918
$108,918
$108,918
$108,918
$108,918
$108,918
69
$81,688
$102,292
$23,236
$163,376
$153,779
$81,688
$163,376
$163,376
$163,376
$81,688
$163,376
$71,931
$163,376
$81,688
$163,376
$163,376
$163,376
$163,376
$163,376
$163,376
$163,376
$163,376
85
$100,749
$126,160
$28,657
$201,498
$189,660
$100,749
$201,498
$201,498
$201,498
$100,749
$201,498
$88,715
$201,498
$100,749
$201,498
$201,498
$201,498
$201,498
$201,498
$201,498
$201,498
$201,498
145
$171,664
$214,961
$48,829
$343,327
$323,158
$171,664
$343,327
$343,327
$343,327
$171,664
$343,327
$151,159
$343,327
$171,664
$343,327
$343,327
$343,327
$343,327
$343,327
$343,327
$343,327
$343,327
                   9-21

-------
                                         Section 9: Estimating Incremental Costs for the Final Regulation
State
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New
Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
District of
Columbia
Site size (acres)
12
$28,413
$28,413
$28,413
$14,207
$14,207
$10,655
$28,413
$14,207
$20,284
$28,413
$14,207
$26,527
$14,207
$28,413
$28,413
$39,463
$14,207
$28,413
$28,413
$28,413
$28,413
$24,459
$14,207
$28,413
$4,041
$28,413
$14,207
$14,207
$28,413
17
$40,252
$40,252
$40,252
$20,126
$20,126
$15,095
$40,252
$20,126
$28,736
$40,252
$20,126
$37,580
$20,126
$40,252
$40,252
$55,906
$20,126
$40,252
$40,252
$40,252
$40,252
$34,650
$20,126
$40,252
$5,725
$40,252
$20,126
$20,126
$40,252
23
$54,459
$54,459
$54,459
$27,229
$27,229
$20,422
$54,459
$27,229
$38,878
$54,459
$27,229
$50,843
$27,229
$54,459
$54,459
$75,637
$27,229
$54,459
$54,459
$54,459
$54,459
$46,880
$27,229
$54,459
$7,745
$54,459
$27,229
$27,229
$54,459
34
$80,504
$80,504
$80,504
$40,252
$40,252
$30,189
$80,504
$40,252
$57,471
$80,504
$40,252
$75,160
$40,252
$80,504
$80,504
$111,812
$40,252
$80,504
$80,504
$80,504
$80,504
$69,301
$40,252
$80,504
$11,450
$80,504
$40,252
$40,252
$80,504
46
$108,918
$108,918
$108,918
$54,459
$54,459
$40,844
$108,918
$54,459
$77,755
$108,918
$54,459
$101,687
$54,459
$108,918
$108,918
$151,274
$54,459
$108,918
$108,918
$108,918
$108,918
$93,760
$54,459
$108,918
$15,491
$108,918
$54,459
$54,459
$108,918
69
$163,376
$163,376
$163,376
$81,688
$81,688
$61 ,266
$163,376
$81,688
$116,633
$163,376
$81,688
$152,530
$81,688
$163,376
$163,376
$226,912
$81,688
$163,376
$163,376
$163,376
$163,376
$140,640
$81,688
$163,376
$23,236
$163,376
$81,688
$81,688
$163,376
85
$201,498
$201,498
$201,498
$100,749
$100,749
$75,562
$201,498
$100,749
$143,847
$201,498
$100,749
$188,120
$100,749
$201,498
$201,498
$279,858
$100,749
$201,498
$201,498
$201,498
$201,498
$173,456
$100,749
$201,498
$28,657
$201,498
$100,749
$100,749
$201,498
145
$343,327
$343,327
$343,327
$171,664
$171,664
$128,748
$343,327
$171,664
$245,098
$343,327
$171,664
$320,534
$171,664
$343,327
$343,327
$476,843
$171,664
$343,327
$343,327
$343,327
$343,327
$295,548
$171,664
$343,327
$48,829
$343,327
$171,664
$171,664
$343,327
Under baseline conditions, EPA also calculated costs for installing silt fence and costs for
installing temporary soil cover. For silt fence, EPA assumes that silt fence would be installed
around the entire perimeter of the site. The perimeter was calculated assuming the sites were
square. For temporary cover, EPA assumes that a percentage of the project would require cover,
ranging from 10 percent for the 1.9-acre model project to 20 percent for the 145-acre model
project. Costs for silt fence and temporary cover do not vary geographically because the controls
are only a function of site size. R.S. Means (2000) reports that a 3-foot-tall silt fence installation
costs between $0.68 and $0.92 per linear foot (for favorable and challenging installations). EPA
used a cost of $0.92 per linear foot for estimating baseline silt fence costs. For temporary cover,
EPA used an average cost based on a range of several types of temporary cover (see Table 9-13).
Table 9-14 shows the baseline costs assumptions for silt fence and temporary cover.
                                           9-22

-------
                                         Section 9: Estimating Incremental Costs for the Final Regulation
                             Table 9-13. Temporary cover costs
Type
Jute mesh
Curled wood fiber
Straw
Wood fiber
Coconut fiber
Straw coconut fiber
AVERAGE
Installed cost
(per acre)
$6,500
$10,500
$8,900
$8,900
$13,000
$10,900
$9,783
           Table 9-14. Baseline silt fence and temporary cover assumptions and costs
Site size
(acres)
1.9
3.8
6.0
8.5
12.0
17.0
23.0
34.0
46.0
69.0
85.1
145.0
% Cover
10%
10.1%
10.3%
10.5%
10.7%
11.1%
1 1 .5%
12.2%
13.1%
14.7%
15.8%
20%
Cost cover
$1,859
$3,767
$6,038
$8,699
$12,569
$18,387
$25,820
$40,727
$58,876
$99,167
$131,675
$283,717
Length silt fence
(ft)
288
407
511
608
723
861
1,001
1,217
1,416
1,734
1,925
2,513
Cost silt fence
$265
$374
$470
$559
$665
$792
$921
$1,120
$1,303
$1,595
$1,771
$2,312
EPA also estimated costs for installing surface outlets on basins. EPA developed these costs on
the basis of an assumed 3-day drain time for sediment basins and assumptions about the number
of basins contained on projects of different sizes in Table 9-15. Surface outlet costs are a
function of the outlet size, which is a function of the flowrate needed to drain the basin in the
specified period. Table 9-16 shows the assumptions used to develop costs for outlets of various
size. Table 9-17 shows the per-state costs for surface outlets for the various model project sizes.
When calculating costs for linear projects, the per-project costs in Table 9-16 were multiplied by
3 to account for the larger number of basins that might be present on linear projects.
                               Table 9-15. Basin assumptions
Site size
(acres)
145
85.1
69
46
# of basins
4
3
3
2
Site size
(acres)
34
23
17
12
# of basins
2
2
1
1
                                           9-23

-------
                                                Section 9: Estimating Incremental Costs for the Final Regulation
                             Table 9-16. Surface outlet cost assumptions
Skimmer
size
(inches)
1.5
2
2.5
3
4
5
6
8
Price per
skimmer
$435
$535
$660
$795
$1,135
$1,655
$2,470
$3,900
Shipping
$20
$20
$22
$25
$32
$155
$280
$500
Ancillary
equipment
$30
$30
$30
$30
$30
$30
$30
$30
Installation
labor
$600
$600
$600
$600
$600
$600
$600
$600
Cubic feet
drained
per day
1,728
3,283
6,234
9,774
20,109
32,832
51,840
97,978
Cubic feet
drained
over 3
days
5,184
9,849
18,702
29,322
60,327
98,496
155,520
293,934
Total cost
$1,085
$1,185
$1,312
$1,450
$1,797
$2,440
$3,380
$5,030
Notes:
- Faircloth recommends 3 days to drain. EPA used 3 days for all states for cost estimates.
- Ancillary costs (pipe, glue) estimated as $30 for all sizes.
- Assumed 8 hours assembly time x $75/hour for installation.

                                  Table 9-17. Costs for surface outlets
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Site size (acres)
12
$1,797
$1,797
$1,450
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,450
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,450
$1,797
17
$2,440
$2,440
$1,797
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$1,450
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$1,797
$2,440
23
$3,594
$3,594
$2,900
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$2,624
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$2,900
$3,594
34
$4,880
$4,880
$3,594
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$2,900
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$3,594
$4,880
46
$4,880
$4,880
$3,594
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$3,594
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$3,594
$4,880
69
$7,320
$7,320
$5,391
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$5,391
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$5,391
$7,320
85.1
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$5,391
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$5,391
$7,320
145
$13,520
$13,520
$9,760
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$7,188
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$9,760
$13,520
                                                   9-24

-------
                                       Section 9: Estimating Incremental Costs for the Final Regulation
State
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Site size (acres)
12
$1,312
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
$1,450
$1,797
$1,797
$1,797
$1,797
$1,797
$1,797
17
$1,450
$2,440
$2,440
$1,797
$2,440
$2,440
$1,797
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$2,440
$1,797
$2,440
$2,440
$2,440
$2,440
$2,440
$1,797
23
$2,624
$3,594
$3,594
$2,900
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
$2,900
$3,594
$3,594
$3,594
$3,594
$3,594
$3,594
34
$2,900
$4,880
$4,880
$3,594
$4,880
$4,880
$3,594
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$3,594
$4,880
$4,880
$4,880
$4,880
$4,880
$3,594
46
$3,594
$4,880
$4,880
$3,594
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
$3,594
$4,880
$4,880
$4,880
$4,880
$4,880
$4,880
69
$5,391
$7,320
$7,320
$5,391
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$5,391
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
85.1
$5,391
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
$7,320
145
$7,188
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
$9,760
$13,520
$13,520
$13,520
$13,520
$13,520
$13,520
   9.3.2.   PASSIVE TREATM ENT COSTS

EPA estimated costs for passive treatment on the basis of implementing a liquid polymer dosing
system for sediment basins and compost filter berms. For a liquid polymer dosing system, a
study conducted in Auckland, New Zealand, indicates that the cost for a rainfall-driven system
ranges from $2,400 NZ to $12,000 NZ ($1,600 to $8,000 U.S.). EPA used the average of the two
values, $4,800, as an estimate of the cost for implementing such a system at each sediment basin.
EPA also estimated using best professional judgment (BPJ) that $500 per month, per system,
would be required for operation and maintenance. Chemical requirements were based on an
estimate of $1,000 per million gallons treated. EPA made various assumptions about the number
of laborers and the hours of labor per storm event required for sampling.  Those assumptions, as
well  as the costs for labor (assuming a labor rate of $30 per hour) and the polymer dosing system
costs are presented in Table 9-18 for the various model project sizes.

For filter berms, EPA used the same assumptions used for estimating silt fence length under
baseline conditions (berms installed around the entire perimeter). EPA used a cost of $1.70 per
linear foot installed. The costs assumptions for filter berms are in Table 9-19.

Tables 9-20 through 9-22 show monthly treatment volumes and associated monthly costs for
passive treatment. Appendix I shows the per-project costs for all options.
                                         9-25

-------
                               Section 9: Estimating Incremental Costs for the Final Regulation
Table 9-18. Passive treatment unit costs: polymer dosing system and labor
Site size
(acres)
1.9
3.8
6.0
8.5
12.0
17.0
23.0
34.0
46.0
69.0
85.1
145.0
# Basins
0
1
1
1
1
1
2
2
2
3
3
4
Cost
polymer
dosing
system
$-
$4,800
$4,800
$4,800
$4,800
$4,800
$9,600
$9,600
$9,600
$14,400
$14,400
$19,200
Monthly
O&M
$-
$500
$500
$500
$500
$500
$1,000
$1,000
$1,000
$1,500
$1,500
$2,000
Sampling
laborers
1
1
1
1
1
1
1
1
2
2
2
3
# Events
per month
monitored
2
2
2
2
2
2
2
2
2
2
2
2
Labor
hours
per
event
4
4
4
4
4
4
4
4
4
4
4
4
Labor
cost
per
event
$120
$120
$120
$120
$120
$120
$120
$120
$240
$240
$240
$360
Monthly
labor cost
$240
$240
$240
$240
$240
$240
$240
$240
$480
$480
$480
$720
          Table 9-19. Passive treatment unit costs: filter berms
Site size
(acres)
1.9
3.8
6.0
8.5
12.0
17.0
23.0
34.0
46.0
69.0
85.1
145.0
Project
length
(ft)
72
102
128
152
181
215
250
304
354
434
481
628
Length filter
berms
(ft)
288
407
511
608
723
861
1,001
1,217
1,416
1,734
1,925
2,513
Cost filter
berms
$490
$692
$869
$1,034
$1,229
$1 ,464
$1,702
$2,069
$2,407
$2,948
$3,273
$4,272
                                 9-26

-------
Table 9-20. Monthly treatment volumes for passive treatment (gallons)
Site size
(acres)
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
1.9
84,331
60,638
13,309
81,888
31,628
22,885
76,378
73,984
72,714
106,642
87,456
31,452
19,491
56,824
68,246
54,449
63,087
77,407
101,070
71,652
72,714
72,921
52,076
49,451
90,244
63,161
20,256
48,692
6,567
69,179
80,817
3.8
168,662
121,275
26,617
163,775
63,256
45,770
152,755
147,968
145,427
213,284
174,911
62,905
38,983
113,648
136,492
108,899
126,174
154,815
202,140
143,304
145,427
145,842
104,152
98,902
180,489
126,323
40,513
97,383
13,134
138,357
161,633
6.0
266,308
191,487
42,028
258,593
99,878
72,269
241,193
233,634
229,622
336,764
276,176
99,323
61,552
179,444
215,514
171,945
199,222
244,445
319,168
226,269
229,622
230,277
164,451
156,161
284,982
199,457
63,968
153,763
20,738
218,459
255,210
8.5
377,270
271,273
59,539
366,340
141,494
102,381
341,690
330,981
325,298
477,083
391,249
140,708
87,198
254,213
305,312
243,589
282,231
346,297
452,155
320,548
325,298
326,226
232,972
221 ,228
403,724
282,564
90,621
217,831
29,378
309,483
361,548
12.0
532,617
382,974
84,055
517,185
199,756
144,537
482,385
467,267
459,245
673,528
552,352
198,647
123,104
358,888
431,028
343,891
398,443
488,889
638,336
452,538
459,245
460,554
328,902
312,322
569,964
398,914
127,935
307,526
41,475
436,917
510,421
17.0
754,540
542,546
119,078
732,679
282,988
204,761
683,379
661 ,962
650,597
954,165
782,499
281,417
174,397
508,425
610,623
487,178
564,462
692,593
904,309
641,095
650,597
652,451
465,944
442,456
807,449
565,129
181,242
435,661
58,757
618,966
723,096
23.0
1,020,849
734,033
161,106
991,272
382,866
277,030
924,572
895,596
880,219
1,290,930
1,058,674
380,740
235,949
687,869
826,137
659,124
763,683
937,038
1,223,478
867,364
880,219
882,728
630,395
598,617
1,092,431
764,586
245,209
589,424
79,495
837,425
978,307
34.0
1,509,081
1,085,092
238,156
1,465,359
565,976
409,522
1,366,758
1,323,924
1,301,193
1,908,331
1,564,997
562,833
348,794
1,016,850
1 ,221 ,247
974,357
1,128,923
1,385,186
1,808,619
1,282,190
1,301,193
1,304,902
931,888
884,912
1,614,898
1,130,257
362,484
871,323
117,514
1,237,932
1,446,193
46.0
2,041,697
1 ,468,066
322,212
1,982,544
765,732
554,060
1,849,143
1,791,191
1,760,438
2,581,859
2,117,349
761,480
471,897
1,375,739
1,652,275
1,318,248
1,527,367
1,874,076
2,446,955
1,734,728
1,760,438
1,765,456
1,260,789
1,197,234
2,184,862
1,529,172
490,419
1,178,848
158,989
1,674,849
1,956,613
69.0
3,062,546
2,202,099
483,317
2,973,816
1,148,598
831,089
2,773,715
2,686,787
2,640,657
3,872,789
3,176,023
1,142,220
707,846
2,063,608
2,478,412
1,977,371
2,291,050
2,811,113
3,670,433
2,602,092
2,640,657
2,648,184
1,891,184
1,795,850
3,277,293
2,293,758
735,628
1,768,273
238,484
2,512,274
2,934,920
85.1
3,777,140
2,715,922
596,091
3,667,707
1,416,604
1,025,010
3,420,915
3,313,704
3,256,810
4,776,440
3,917,096
1,408,738
873,010
2,545,117
3,056,709
2,438,758
2,825,628
3,467,040
4,526,867
3,209,247
3,256,810
3,266,093
2,332,460
2,214,882
4,041,994
2,828,968
907,275
2,180,870
294,130
3,098,471
3,619,735
145.0
6,435,785
4,627,600
1,015,667
6,249,324
2,413,720
1 ,746,492
5,828,821
5,646,147
5,549,206
8,138,469
6,674,252
2,400,317
1,487,503
4,336,568
5,208,258
4,155,346
4,814,525
5,907,412
7,713,228
5,468,165
5,549,206
5,565,024
3,974,228
3,773,888
6,887,064
4,820,216
1,545,886
3,715,935
501,161
5,279,416
6,167,586

-------
CD

ro
oo
Site size
(acres)
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
1.9
26,580
63,472
74,271
27,663
64,983
56,430
70,026
71,565
76,941
78,161
27,679
79,292
55,927
25,034
58,475
69,392
60,638
73,641
54,119
25,340
3.8
53,160
126,943
148,541
55,327
129,967
112,861
140,052
143,130
153,883
156,322
55,358
158,584
111,855
50,068
116,950
138,784
121,275
147,283
108,239
50,679
6.0
83,937
200,437
234,539
87,358
205,211
178,201
221,135
225,994
242,973
246,824
87,407
250,396
176,613
79,055
184,657
219,132
191,487
232,552
170,903
80,020
8.5
118,911
283,953
332,263
123,757
290,715
252,451
313,274
320,158
344,211
349,668
123,826
354,727
250,201
1 1 1 ,994
261,598
310,438
271,273
329,449
242,113
113,361
12.0
167,875
400,874
469,077
174,715
410,421
356,402
442,270
451,988
485,945
493,649
174,813
500,791
353,225
158,110
369,315
438,265
382,974
465,104
341,807
160,040
17.0
237,823
567,905
664,526
247,513
581,430
504,903
626,549
640,316
688,423
699,336
247,652
709,454
500,403
223,989
523,196
620,875
542,546
658,897
484,226
226,723
23.0
321,760
768,342
899,065
334,871
786,641
683,104
847,684
866,310
931,395
946,160
335,059
959,850
677,015
303,043
707,854
840,007
734,033
891,449
655,130
306,743
34.0
475,645
1,135,810
1,329,052
495,027
1,162,860
1,009,805
1,253,098
1,280,633
1,376,845
1,398,672
495,305
1,418,908
1,000,805
447,977
1,046,392
1,241,750
1,085,092
1,317,794
968,453
453,446
46.0
643,520
1,536,684
1,798,129
669,742
1,573,282
1,366,207
1,695,368
1,732,621
1,862,790
1,892,321
670,118
1,919,700
1,354,030
606,087
1,415,707
1,680,015
1 ,468,066
1,782,898
1,310,260
613,485
69.0
965,281
2,305,027
2,697,194
1,004,613
2,359,922
2,049,311
2,543,051
2,598,931
2,794,185
2,838,481
1,005,177
2,879,549
2,031,046
909,130
2,123,561
2,520,022
2,202,099
2,674,347
1,965,390
920,228
85.1
1,190,513
2,842,866
3,326,539
1,239,023
2,910,571
2,527,484
3,136,430
3,205,348
3,446,162
3,500,793
1,239,718
3,551,444
2,504,956
1,121,261
2,619,058
3,108,027
2,715,922
3,298,361
2,423,981
1,134,948
145.0
2,028,488
4,843,896
5,668,016
2,111,144
4,959,257
4,306,523
5,344,094
5,461,521
5,871,839
5,964,924
2,112,329
6,051,227
4,268,139
1,910,491
4,462,555
5,295,699
4,627,600
5,620,004
4,130,167
1,933,812

-------
Table 9-21. Monthly chemical cost for passive treatment
Site size
(acres)
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
1.9
$84
$61
$13
$82
$32
$23
$76
$74
$73
$107
$87
$31
$19
$57
$68
$54
$63
$77
$101
$72
$73
$73
$52
$49
$90
$63
$20
$49
$7
$69
$81
$27
3.8
$169
$121
$27
$164
$63
$46
$153
$148
$145
$213
$175
$63
$39
$114
$136
$109
$126
$155
$202
$143
$145
$146
$104
$99
$180
$126
$41
$97
$13
$138
$162
$53
6.0
$266
$191
$42
$259
$100
$72
$241
$234
$230
$337
$276
$99
$62
$179
$216
$172
$199
$244
$319
$226
$230
$230
$164
$156
$285
$199
$64
$154
$21
$218
$255
$84
8.5
$377
$271
$60
$366
$141
$102
$342
$331
$325
$477
$391
$141
$87
$254
$305
$244
$282
$346
$452
$321
$325
$326
$233
$221
$404
$283
$91
$218
$29
$309
$362
$119
12.0
$533
$383
$84
$517
$200
$145
$482
$467
$459
$674
$552
$199
$123
$359
$431
$344
$398
$489
$638
$453
$459
$461
$329
$312
$570
$399
$128
$308
$41
$437
$510
$168
17.0
$755
$543
$119
$733
$283
$205
$683
$662
$651
$954
$782
$281
$174
$508
$611
$487
$564
$693
$904
$641
$651
$652
$466
$442
$807
$565
$181
$436
$59
$619
$723
$238
23.0
$1,021
$734
$161
$991
$383
$277
$925
$896
$880
$1,291
$1,059
$381
$236
$688
$826
$659
$764
$937
$1,223
$867
$880
$883
$630
$599
$1,092
$765
$245
$589
$79
$837
$978
$322
34.0
$1,509
$1,085
$238
$1,465
$566
$410
$1,367
$1,324
$1,301
$1,908
$1,565
$563
$349
$1,017
$1,221
$974
$1,129
$1,385
$1,809
$1,282
$1,301
$1,305
$932
$885
$1,615
$1,130
$362
$871
$118
$1,238
$1 ,446
$476
46.0
$2,042
$1,468
$322
$1,983
$766
$554
$1,849
$1,791
$1,760
$2,582
$2,117
$761
$472
$1,376
$1,652
$1,318
$1,527
$1,874
$2,447
$1,735
$1,760
$1,765
$1,261
$1,197
$2,185
$1,529
$490
$1,179
$159
$1,675
$1,957
$644
69.0
$3,063
$2,202
$483
$2,974
$1,149
$831
$2,774
$2,687
$2,641
$3,873
$3,176
$1,142
$708
$2,064
$2,478
$1,977
$2,291
$2,811
$3,670
$2,602
$2,641
$2,648
$1,891
$1,796
$3,277
$2,294
$736
$1,768
$238
$2,512
$2,935
$965
85.1
$3,777
$2,716
$596
$3,668
$1,417
$1,025
$3,421
$3,314
$3,257
$4,776
$3,917
$1,409
$873
$2,545
$3,057
$2,439
$2,826
$3,467
$4,527
$3,209
$3,257
$3,266
$2,332
$2,215
$4,042
$2,829
$907
$2,181
$294
$3,098
$3,620
$1,191
145.0
$6,436
$4,628
$1,016
$6,249
$2,414
$1,746
$5,829
$5,646
$5,549
$8,138
$6,674
$2,400
$1,488
$4,337
$5,208
$4,155
$4,815
$5,907
$7,713
$5,468
$5,549
$5,565
$3,974
$3,774
$6,887
$4,820
$1,546
$3,716
$501
$5,279
$6,168
$2,028

-------
CD

CO
O
Site size
(acres)
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
1.9
$63
$74
$28
$65
$56
$70
$72
$77
$78
$28
$79
$56
$25
$58
$69
$61
$74
$54
$25
3.8
$127
$149
$55
$130
$113
$140
$143
$154
$156
$55
$159
$112
$50
$117
$139
$121
$147
$108
$51
6.0
$200
$235
$87
$205
$178
$221
$226
$243
$247
$87
$250
$177
$79
$185
$219
$191
$233
$171
$80
8.5
$284
$332
$124
$291
$252
$313
$320
$344
$350
$124
$355
$250
$112
$262
$310
$271
$329
$242
$113
12.0
$401
$469
$175
$410
$356
$442
$452
$486
$494
$175
$501
$353
$158
$369
$438
$383
$465
$342
$160
17.0
$568
$665
$248
$581
$505
$627
$640
$688
$699
$248
$709
$500
$224
$523
$621
$543
$659
$484
$227
23.0
$768
$899
$335
$787
$683
$848
$866
$931
$946
$335
$960
$677
$303
$708
$840
$734
$891
$655
$307
34.0
$1,136
$1,329
$495
$1,163
$1,010
$1,253
$1,281
$1,377
$1,399
$495
$1,419
$1,001
$448
$1,046
$1 ,242
$1,085
$1,318
$968
$453
46.0
$1,537
$1,798
$670
$1,573
$1,366
$1,695
$1,733
$1,863
$1,892
$670
$1,920
$1,354
$606
$1,416
$1,680
$1 ,468
$1,783
$1,310
$613
69.0
$2,305
$2,697
$1,005
$2,360
$2,049
$2,543
$2,599
$2,794
$2,838
$1,005
$2,880
$2,031
$909
$2,124
$2,520
$2,202
$2,674
$1,965
$920
85.1
$2,843
$3,327
$1,239
$2,911
$2,527
$3,136
$3,205
$3,446
$3,501
$1,240
$3,551
$2,505
$1,121
$2,619
$3,108
$2,716
$3,298
$2,424
$1,135
145.0
$4,844
$5,668
$2,111
$4,959
$4,307
$5,344
$5,462
$5,872
$5,965
$2,112
$6,051
$4,268
$1,910
$4,463
$5,296
$4,628
$5,620
$4,130
$1,934

-------
Table 9-22. Total monthly cost for passive treatment
Site Size
(acres)
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
1.9
$324
$301
$253
$322
$272
$263
$316
$314
$313
$347
$327
$271
$259
$297
$308
$294
$303
$317
$341
$312
$313
$313
$292
$289
$330
$303
$260
$289
$247
$309
$321
$267
3.8
$909
$861
$767
$904
$803
$786
$893
$888
$885
$953
$915
$803
$779
$854
$876
$849
$866
$895
$942
$883
$885
$886
$844
$839
$920
$866
$781
$837
$753
$878
$902
$793
6.0
$1,006
$931
$782
$999
$840
$812
$981
$974
$970
$1,077
$1,016
$839
$802
$919
$956
$912
$939
$984
$1,059
$966
$970
$970
$904
$896
$1,025
$939
$804
$894
$761
$958
$995
$824
8.5
$1,117
$1,011
$800
$1,106
$881
$842
$1,082
$1,071
$1,065
$1,217
$1,131
$881
$827
$994
$1,045
$984
$1,022
$1,086
$1,192
$1,061
$1,065
$1,066
$973
$961
$1,144
$1,023
$831
$958
$769
$1,049
$1,102
$859
12.0
$1,273
$1,123
$824
$1,257
$940
$885
$1 ,222
$1,207
$1,199
$1,414
$1,292
$939
$863
$1,099
$1,171
$1,084
$1,138
$1 ,229
$1,378
$1,193
$1,199
$1,201
$1,069
$1,052
$1,310
$1,139
$868
$1,048
$781
$1,177
$1,250
$908
17.0
$1,495
$1,283
$859
$1,473
$1,023
$945
$1,423
$1,402
$1,391
$1,694
$1,522
$1,021
$914
$1,248
$1,351
$1,227
$1,304
$1,433
$1 ,644
$1,381
$1,391
$1,392
$1,206
$1,182
$1,547
$1,305
$921
$1,176
$799
$1,359
$1,463
$978
23.0
$2,261
$1,974
$1,401
$2,231
$1,623
$1,517
$2,165
$2,136
$2,120
$2,531
$2,299
$1,621
$1,476
$1,928
$2,066
$1,899
$2,004
$2,177
$2,463
$2,107
$2,120
$2,123
$1,870
$1,839
$2,332
$2,005
$1,485
$1,829
$1,319
$2,077
$2,218
$1,562
34.0
$2,749
$2,325
$1,478
$2,705
$1,806
$1,650
$2,607
$2,564
$2,541
$3,148
$2,805
$1,803
$1,589
$2,257
$2,461
$2,214
$2,369
$2,625
$3,049
$2,522
$2,541
$2,545
$2,172
$2,125
$2,855
$2,370
$1,602
$2,111
$1,358
$2,478
$2,686
$1,716
46.0
$3,522
$2,948
$1,802
$3,463
$2,246
$2,034
$3,329
$3,271
$3,240
$4,062
$3,597
$2,241
$1,952
$2,856
$3,132
$2,798
$3,007
$3,354
$3,927
$3,215
$3,240
$3,245
$2,741
$2,677
$3,665
$3,009
$1,970
$2,659
$1,639
$3,155
$3,437
$2,124
69.0
$5,043
$4,182
$2,463
$4,954
$3,129
$2,811
$4,754
$4,667
$4,621
$5,853
$5,156
$3,122
$2,688
$4,044
$4,458
$3,957
$4,271
$4,791
$5,650
$4,582
$4,621
$4,628
$3,871
$3,776
$5,257
$4,274
$2,716
$3,748
$2,218
$4,492
$4,915
$2,945
85.1
$5,757
$4,696
$2,576
$5,648
$3,397
$3,005
$5,401
$5,294
$5,237
$6,756
$5,897
$3,389
$2,853
$4,525
$5,037
$4,419
$4,806
$5,447
$6,507
$5,189
$5,237
$5,246
$4,312
$4,195
$6,022
$4,809
$2,887
$4,161
$2,274
$5,078
$5,600
$3,171
145.0
$9,156
$7,348
$3,736
$8,969
$5,134
$4,466
$8,549
$8,366
$8,269
$10,858
$9,394
$5,120
$4,208
$7,057
$7,928
$6,875
$7,535
$8,627
$10,433
$8,188
$8,269
$8,285
$6,694
$6,494
$9,607
$7,540
$4,266
$6,436
$3,221
$7,999
$8,888
$4,748

-------
CD

CO
IV)
Site Size
(acres)
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
1.9
$303
$314
$268
$305
$296
$310
$312
$317
$318
$268
$319
$296
$265
$298
$309
$301
$314
$294
$265
3.8
$867
$889
$795
$870
$853
$880
$883
$894
$896
$795
$899
$852
$790
$857
$879
$861
$887
$848
$791
6.0
$940
$975
$827
$945
$918
$961
$966
$983
$987
$827
$990
$917
$819
$925
$959
$931
$973
$911
$820
8.5
$1,024
$1,072
$864
$1,031
$992
$1,053
$1,060
$1,084
$1,090
$864
$1,095
$990
$852
$1,002
$1,050
$1,011
$1,069
$982
$853
12.0
$1,141
$1,209
$915
$1,150
$1,096
$1,182
$1,192
$1 ,226
$1,234
$915
$1,241
$1,093
$898
$1,109
$1,178
$1,123
$1,205
$1,082
$900
17.0
$1,308
$1,405
$988
$1,321
$1,245
$1,367
$1,380
$1,428
$1,439
$988
$1,449
$1,240
$964
$1,263
$1,361
$1,283
$1,399
$1 ,224
$967
23.0
$2,008
$2,139
$1,575
$2,027
$1,923
$2,088
$2,106
$2,171
$2,186
$1,575
$2,200
$1,917
$1,543
$1,948
$2,080
$1,974
$2,131
$1,895
$1,547
34.0
$2,376
$2,569
$1,735
$2,403
$2,250
$2,493
$2,521
$2,617
$2,639
$1,735
$2,659
$2,241
$1,688
$2,286
$2,482
$2,325
$2,558
$2,208
$1,693
46.0
$3,017
$3,278
$2,150
$3,053
$2,846
$3,175
$3,213
$3,343
$3,372
$2,150
$3,400
$2,834
$2,086
$2,896
$3,160
$2,948
$3,263
$2,790
$2,093
69.0
$4,285
$4,677
$2,985
$4,340
$4,029
$4,523
$4,579
$4,774
$4,818
$2,985
$4,860
$4,011
$2,889
$4,104
$4,500
$4,182
$4,654
$3,945
$2,900
85.1
$4,823
$5,307
$3,219
$4,891
$4,507
$5,116
$5,185
$5,426
$5,481
$3,220
$5,531
$4,485
$3,101
$4,599
$5,088
$4,696
$5,278
$4,404
$3,115
145.0
$7,564
$8,388
$4,831
$7,679
$7,027
$8,064
$8,182
$8,592
$8,685
$4,832
$8,771
$6,988
$4,630
$7,183
$8,016
$7,348
$8,340
$6,850
$4,654

-------
                                       Section 9: Estimating Incremental Costs for the Final Regulation
   9.3.3.   ATS COSTS

EPA estimated costs for ATS using a combination of one-time costs, which accounts for items
such as site preparation and installing storage, as well as recurring costs that account for items
such as equipment rental, operator labor, and treatment chemicals. EPA estimates the size of the
ATS system needed in each state for each model projects size on the basis of the system design
flowrates in Table 9-9. On the basis of system flowrate, EPA estimated monthly equipment
rental costs provided in vendor quotes (see Table 7-26) by scaling up the unit costs for a 500-
gpm system to the desired system flowrate. EPA notes that this is a conservative assumption,
because there are economies of scale that are not captured in this approach. For example, for a
1,000-gpm system, it might be more economical to rent one larger pump instead of two smaller
pumps. However, the approach does provide reasonable, albeit conservative, estimates of costs.
For labor and treatment chemical, the system run-time was estimated on the basis of the monthly
treatment volumes (including an additional 10 percent for system startup and shutdown) and the
associated chemical dosage rate and pump and generator fuel consumption.  Operator labor was
also included. For storage, EPA estimated basin storage volumes to impound runoff from the 2-
year, 24-hour storm. If that volume was larger than the incremental sediment basin storage
requirements contained in existing state permits, additional costs for storage were also calculated
using the same methodology for estimating sediment basin costs described above. Costs were
also estimated for periodic equipment servicing and for providing a stabilized pad consisting of
crushed stone.

Table 9-23 shows the cost assumptions used for an ATS  system for a flowrate of 500 gpm. For
larger systems, costs were estimated as multiples of the values found in Table 9-24.  For the 100-
gpm system, it was assumed that a trailer-mounted system would be used with a rental cost of
$10,000 per month.

Tables 9-24 and 9-25 show the one-time ATS and monthly ATS costs for the 12 model project
sizes. Additional details of the costing approach are in the C&D Cost Spreadsheet Models. The
ATS Cost Spreadsheet Model (DCN 43119) was used to estimate ATS treatment costs and
storage costs, while the Unit Cost Spreadsheet Model (DCN 43120) was used to estimate ESC
and passive treatment costs. The per-project costs under  each of the regulatory options are in
Appendix I. Model project costs were scaled to  the national level using the distribution of
projects described earlier. National compliance  costs are described in the Economic Analysis for
Final Effluent Guidelines and Standards for the Construction and Development Category
(USEPA 2009).
                                         9-33

-------
               Section 9: Estimating Incremental Costs for the Final Regulation
Table 9-23. ATS costs: 500-gpm system
Equipment rental
Filter
System
Pump
Hose
Elbow
Genset
Tanks
$/month
$4,000
$4,000
$1,650
$98
$36
$1,050
$1,260
Quantity
1
1
2
4
2
1
2
Total Monthly Costs
Total $/mo
$4,000
$4,000
$3,300
$392
$72
$1,050
$2,520
$15,334

One-Time Costs
Delivery
Pick-up
Stabilized pad (1600 SF @ $0.50/SF)
Calibration
Pipes, valves, electrical
Installation
Misc. lab equipment & supplies
Sand and gravel
Total One-Time Costs
$1,485
$1,485
$800
$1,300
$6,000
$7,500
$4,250
$1 ,222
$24,042

Labor, O&M and Materials
# Laborers
Labor rate ($/hr)
Fuel ($/hr)
Total labor + fuel ($/hr)


Equipment servicing (every 250 hrs)
Chitosan ($/Mgal)
1
$75
$30
3 units® 10GPH per unit
$105

$/service
$300
$982

# units
3


Total
$900

                 9-34

-------
Table 9-24. One-time ATS costs model projects
Site size
(acres)
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New
Hampshire
1.9
$21,886
$13,220
$9,392
$18,863
$10,029
$11,748
$13,695
$14,249
$14,082
$21 ,602
$18,413
$20,974
$9,392
$13,130
$15,850
$14,311
$15,312
$13,646
$24,435
$13,321
$14,082
$13,612
$10,554
$12,700
$20,073
$15,608
$9,577
$14,241
$9,392
$12,471
3.8
$34,380
$17,047
$9,392
$28,334
$10,665
$14,103
$17,997
$19,107
$18,772
$33,813
$27,435
$32,556
$9,392
$16,867
$22,308
$19,229
$21,231
$17,900
$54,129
$17,250
$18,772
$17,832
$11,716
$16,008
$30,754
$21 ,824
$9,762
$19,089
$9,392
$15,550
6.0
$63,497
$21,479
$9,392
$53,951
$11,403
$16,831
$22,979
$24,731
$24,202
$62,601
$37,880
$60,616
$9,392
$21,195
$29,786
$24,925
$28,086
$22,825
$71,547
$21,799
$24,202
$22,719
$13,062
$19,838
$57,771
$29,021
$9,976
$24,704
$9,392
$19,116
8.5
$79,937
$26,516
$9,392
$66,413
$12,240
$19,930
$43,290
$45,773
$45,023
$78,668
$64,400
$75,855
$9,392
$26,113
$52,934
$46,047
$50,525
$43,072
$91,341
$26,969
$45,023
$42,921
$14,591
$24,190
$71,825
$51,850
$10,219
$31,083
$9,392
$23,167
12.0
$102,952
$33,567
$9,392
$83,860
$13,413
$24,270
$51,216
$54,720
$53,663
$101,160
$81,019
$97,190
$9,392
$47,649
$64,831
$55,107
$61 ,429
$50,908
$119,053
$48,856
$53,663
$50,695
$16,732
$44,934
$91,500
$63,300
$10,560
$54,665
$9,392
$43,489
17.0
$135,831
$58,290
$9,392
$108,784
$29,739
$45,118
$62,539
$67,503
$66,005
$133,293
$104,759
$127,669
$9,392
$57,485
$81,826
$68,051
$77,007
$62,103
$158,641
$59,196
$66,005
$61,801
$34,440
$53,639
$119,607
$79,657
$11,047
$67,425
$9,392
$51,592
23.0
$175,286
$70,377
$9,392
$138,693
$31,749
$52,557
$76,126
$82,842
$80,815
$191,644
$133,247
$184,035
$9,392
$69,288
$102,221
$83,584
$95,701
$75,536
$225,938
$71,603
$80,815
$75,127
$38,111
$64,084
$153,336
$99,286
$11,631
$82,736
$9,392
$61,316
34.0
$267,412
$92,538
$24,042
$213,318
$35,435
$66,195
$101,035
$110,964
$107,968
$262,336
$205,268
$251,088
$9,392
$90,928
$139,610
$112,060
$129,972
$100,163
$313,032
$94,349
$107,968
$99,560
$44,839
$83,235
$234,965
$155,064
$27,352
$110,808
$9,392
$79,143
46.0
$346,322
$116,713
$24,042
$273,136
$39,457
$81,072
$148,002
$161,435
$157,381
$359,247
$262,244
$344,028
$24,042
$134,326
$200,191
$162,917
$187,151
$146,822
$427,834
$138,955
$157,381
$146,005
$52,179
$104,127
$302,423
$194,322
$28,520
$141,431
$24,042
$98,590
69.0
$517,359
$163,048
$24,042
$407,579
$47,164
$109,588
$200,085
$220,235
$214,154
$526,849
$391,241
$504,022
$24,042
$179,573
$278,370
$222,459
$258,810
$198,315
$629,731
$186,516
$214,154
$197,090
$86,040
$163,961
$451,509
$289,358
$30,759
$219,917
$24,042
$155,656
85.1
$643,022
$215,275
$24,042
$507,626
$72,351
$149,340
$256,336
$281,187
$273,687
$630,316
$467,685
$602,162
$24,042
$21 1 ,245
$352,887
$283,930
$328,763
$254,153
$776,996
$219,809
$273,687
$252,642
$95,888
$191,991
$561,807
$342,028
$32,326
$261 ,004
$24,042
$181,748
145.0
$1,080,748
$340,198
$48,084
$850,052
$96,674
$227,854
$416,022
$458,366
$445,586
$1,098,684
$815,718
$1,050,712
$48,084
$372,916
$580,534
$463,040
$579,014
$412,302
$1,314,884
$387,508
$445,586
$409,728
$176,360
$340,108
$981,952
$601,616
$62,198
$457,698
$48,084
$322,656

-------
CD
CO
Site size
(acres)
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
1.9
$16,997
$9,392
$13,361
$16,995
$9,392
$14,967
$16,370
$11,760
$12,834
$16,189
$15,835
$10,823
$15,009
$17,893
$9,392
$13,822
$13,817
$13,220
$11,990
$14,496
$10,604
3.8
$24,603
$9,392
$17,330
$24,598
$9,392
$20,541
$23,348
$14,128
$16,275
$22,986
$22,278
$12,254
$20,626
$26,395
$9,392
$18,252
$18,242
$17,047
$14,588
$19,599
$11,816
6.0
$33,409
$9,392
$21,925
$33,401
$9,392
$26,996
$31,428
$16,870
$20,260
$30,856
$29,739
$13,911
$27,131
$50,889
$9,392
$23,382
$23,366
$21,479
$17,596
$25,509
$13,220
8.5
$58,066
$9,392
$27,148
$58,055
$9,392
$34,332
$55,260
$19,986
$39,438
$54,449
$52,867
$15,793
$49,172
$62,075
$9,392
$29,211
$43,838
$26,516
$21,015
$32,224
$14,815
12.0
$72,076
$9,392
$49,109
$72,060
$9,392
$59,251
$68,114
$38,998
$45,778
$66,970
$64,736
$18,429
$59,519
$77,735
$9,392
$52,021
$51,990
$33,567
$40,451
$56,275
$17,048
17.0
$92,091
$9,392
$59,553
$92,068
$9,392
$73,921
$86,478
$45,230
$54,835
$84,856
$81,692
$36,844
$74,301
$100,107
$9,392
$63,679
$63,634
$58,290
$47,288
$69,706
$20,238
23.0
$116,108
$24,042
$72,087
$116,077
$24,042
$91,526
$108,514
$52,709
$65,703
$106,320
$102,039
$41,363
$92,040
$126,954
$9,392
$77,669
$77,608
$70,377
$55,492
$85,822
$38,716
34.0
$160,139
$24,042
$95,065
$160,094
$24,042
$123,800
$168,705
$66,419
$85,628
$145,671
$159,134
$49,647
$124,561
$195,965
$24,042
$103,316
$103,227
$92,538
$70,533
$115,370
$45,733
46.0
$227,965
$24,042
$139,924
$227,905
$24,042
$159,009
$212,777
$81,375
$127,156
$208,390
$199,828
$58,684
$179,830
$249,658
$24,042
$131,296
$150,967
$116,713
$86,942
$147,603
$53,389
69.0
$320,031
$24,042
$187,968
$319,940
$24,042
$246,285
$317,041
$129,834
$168,817
$290,669
$297,617
$95,797
$247,828
$372,362
$24,042
$204,714
$204,533
$163,048
$138,184
$229,175
$68,063
85.1
$404,269
$24,042
$221 ,600
$404,157
$24,042
$293,523
$376,171
$149,900
$217,772
$368,055
$352,215
$107,921
$315,218
$464,192
$24,042
$242,253
$261,821
$215,275
$160,199
$272,422
$78,334
145.0
$668,082
$48,084
$390,558
$667,892
$48,084
$513,108
$659,790
$268,392
$350,314
$606,378
$618,972
$196,866
$555,934
$776,046
$48,084
$425,750
$425,368
$340,198
$285,940
$477,154
$140,592

-------
Table 9-25. Monthly ATS costs for model projects
Site size
(acres)
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New
Hampshire
New Jersey
1.9
$11,503
$11,045
$10,229
$11,461
$10,545
$10,394
$11,366
$11,325
$11,287
$11,888
$11,557
$10,542
$10,336
$10,979
$1 1 ,226
$10,938
$11,137
$11,384
$11,792
$11,285
$11,303
$11,307
$10,897
$10,852
$11,605
$11,138
$10,349
$10,839
$10,113
$1 1 ,242
$11,443
3.8
$12,956
$12,140
$10,459
$12,872
$11,140
$10,789
$12,682
$12,600
$12,524
$13,775
$13,064
$11,134
$10,672
$12,008
$12,402
$11,927
$12,224
$12,718
$16,291
$12,519
$12,556
$12,563
$11,845
$11,754
$13,160
$12,227
$10,698
$11,728
$10,226
$12,434
$12,835
6.0
$16,594
$13,400
$10,724
$16,558
$11,771
$11,295
$14,256
$14,126
$14,006
$17,003
$14,859
$15,804
$11,061
$13,142
$13,814
$13,013
$13,533
$14,312
$16,919
$13,999
$14,057
$14,068
$12,884
$12,741
$16,683
$13,537
$11,152
$12,700
$10,357
$13,864
$14,498
8.5
$17,194
$14,775
$11,026
$17,143
$12,488
$11,814
$17,026
$16,975
$16,929
$17,667
$17,260
$16,000
$11,553
$14,481
$16,779
$16,487
$16,670
$17,048
$17,549
$15,674
$16,948
$16,953
$14,115
$13,912
$17,319
$16,671
$11,612
$13,854
$10,506
$15,433
$17,120
12.0
$17,929
$16,749
$11,498
$17,856
$13,542
$12,541
$17,692
$17,620
$17,554
$18,596
$18,023
$16,274
$12,171
$17,107
$17,449
$17,036
$17,294
$17,722
$18,430
$17,550
$17,582
$17,588
$15,818
$16,812
$18,106
$17,297
$12,255
$16,789
$10,715
$17,476
$17,824
17.0
$18,979
$17,976
$12,102
$18,876
$16,673
$16,303
$18,643
$18,541
$18,448
$19,999
$19,112
$16,666
$13,055
$17,815
$18,298
$17,714
$18,080
$18,686
$19,688
$18,443
$18,488
$18,496
$17,614
$17,503
$19,230
$18,083
$13,173
$17,471
$11,012
$18,338
$18,831
23.0
$20,315
$18,882
$12,826
$20,175
$17,221
$16,645
$19,784
$19,647
$19,521
$33,477
$20,494
$30,296
$14,166
$18,664
$19,318
$18,528
$19,023
$19,843
$33,250
$19,513
$19,574
$19,586
$18,392
$18,242
$20,653
$19,027
$14,325
$18,198
$11,420
$19,372
$20,113
34.0
$34,209
$20,619
$16,461
$34,062
$18,087
$17,347
$21,951
$21,749
$21 ,562
$35,674
$34,397
$30,907
$16,160
$20,296
$21 ,263
$20,095
$20,826
$22,039
$35,215
$21,551
$21,641
$21,659
$19,819
$19,596
$34,564
$32,937
$17,124
$19,532
$12,075
$21 ,342
$22,327
46.0
$36,122
$22,431
$16,934
$35,923
$19,032
$18,031
$35,351
$35,156
$34,977
$52,728
$36,376
$45,607
$17,642
$33,761
$34,690
$33,568
$34,270
$35,434
$52,213
$34,966
$35,053
$35,070
$21,450
$21,149
$36,603
$34,276
$17,730
$21 ,062
$16,086
$23,484
$35,836
69.0
$54,737
$26,054
$17,696
$54,398
$20,919
$19,342
$38,579
$38,288
$38,019
$69,837
$55,170
$60,220
$18,759
$36,196
$37,588
$35,906
$36,959
$38,705
$68,933
$38,003
$38,133
$38,158
$35,617
$35,172
$55,556
$51,628
$18,890
$35,079
$16,463
$37,702
$39,121
85.1
$69,516
$38,385
$18,230
$68,923
$33,899
$32,584
$56,104
$55,695
$55,318
$72,871
$57,997
$61,340
$19,540
$37,812
$54,714
$52,182
$53,833
$56,280
$86,787
$40,166
$55,478
$55,513
$37,098
$36,703
$70,405
$53,845
$19,702
$36,589
$16,726
$39,670
$56,862
145.0
$110,658
$53,166
$35,624
$109,948
$42,390
$39,082
$78,104
$77,490
$76,926
$140,992
$111,568
$120,830
$37,856
$73,094
$76,020
$72,486
$104,124
$78,368
$139,566
$76,892
$77,164
$77,218
$71,878
$71,204
$136,342
$104,144
$38,134
$70,760
$33,040
$76,258
$79,240

-------
CD

CO
OO
Site size
(acres)
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
1.9
$10,458
$11,144
$11,330
$10,477
$11,170
$10,972
$11,257
$11,283
$11,376
$11,397
$10,477
$11,416
$10,964
$10,431
$11,008
$1 1 ,246
$11,045
$11,319
$10,933
$10,437
3.8
$10,916
$12,237
$12,610
$10,953
$12,290
$11,995
$12,463
$12,516
$12,702
$12,744
$10,954
$12,783
$11,977
$10,863
$12,065
$12,442
$12,140
$12,588
$11,915
$10,873
6.0
$11,496
$13,554
$14,142
$11,555
$13,636
$13,121
$13,911
$13,994
$14,287
$14,353
$11,556
$14,415
$16,170
$11,412
$13,232
$13,876
$13,400
$14,107
$12,995
$11,429
8.5
$12,099
$14,993
$16,981
$12,183
$15,110
$16,529
$15,548
$16,924
$17,038
$17,064
$12,184
$17,088
$16,518
$11,980
$14,608
$16,803
$14,775
$15,827
$14,272
$12,003
12.0
$12,943
$17,306
$17,629
$13,061
$17,351
$17,095
$17,502
$17,548
$17,708
$17,745
$13,062
$17,779
$17,080
$12,775
$17,157
$17,483
$16,749
$17,610
$17,026
$12,808
17.0
$14,198
$18,096
$18,554
$14,365
$18,160
$17,798
$18,374
$18,439
$18,667
$18,718
$16,506
$18,766
$17,777
$13,960
$17,885
$18,347
$17,976
$18,527
$17,700
$14,007
23.0
$16,932
$19,045
$19,663
$16,994
$19,131
$18,641
$19,420
$19,508
$19,816
$19,961
$16,994
$20,026
$18,613
$15,322
$18,759
$19,384
$18,882
$19,627
$18,509
$16,786
34.0
$17,660
$20,859
$21,773
$17,751
$20,987
$32,533
$21,414
$21 ,544
$21,999
$33,838
$17,753
$22,198
$32,503
$17,529
$20,436
$21,360
$20,619
$21,720
$20,067
$17,555
46.0
$18,454
$34,302
$35,179
$18,578
$23,004
$33,729
$23,581
$34,959
$35,396
$35,621
$18,580
$35,712
$33,688
$18,277
$22,183
$34,783
$22,431
$23,996
$21 ,684
$18,312
69.0
$20,052
$37,006
$38,322
$20,238
$37,190
$50,696
$37,805
$37,993
$38,648
$53,882
$32,517
$38,935
$50,626
$19,711
$36,397
$37,728
$26,054
$38,246
$35,866
$19,764
85.1
$21,118
$38,812
$55,744
$21,347
$39,039
$52,520
$39,922
$55,281
$56,200
$56,409
$33,305
$56,602
$65,020
$20,790
$38,060
$54,910
$38,385
$40,466
$37,405
$20,855
145.0
$40,566
$74,796
$77,564
$40,958
$75,184
$102,184
$76,476
$76,870
$78,248
$108,862
$65,378
$109,192
$102,038
$40,008
$73,516
$76,314
$53,166
$77,402
$72,400
$40,118

-------
                                       Section 9: Estimating Incremental Costs for the Final Regulation
 9.4.    REFERENCES

CSQA (California Stormwater Quality Association). 2003. Stormwater BMP Handbook:
   Construction. California Stormwater Quality Association, Menlo Park, CA.
   .

SWRPC (Southeastern Wisconsin Regional Planning Commission). 1991. Costs of Urban
   Nonpoint Source Water Pollution Control Measures. Technical Report No. 31. Southeastern
   Wisconsin Regional Planning Commission, Waukesha, WI.

USDA (U.S. Department of Agriculture). 1986. Urban Hydrology for Small Watersheds.
   Technical Release 55 (TR55). U.S. Department of Agriculture, Washington, DC.

USEPA (U.S. Environmental Protection Agency). 2009. Economic Analysis for Final Effluent
   Guidelines and Standards for the Construction and Development Category (EPA-821-R-09-
   011). U.S. Environmental Protection Agency, Office of Water, Washington, DC.

USEPA (U.S. Environmental Protection Agency). 1993. Guidance Specifying Management
   Measures for Sources of Nonpoint Pollution in Coastal Waters. EPA-840-B-92-002. U.S.
   Environmental Protection Agency, Office of Water, Washington, DC.
                                         9-39

-------
                                                  Section 10: Estimating Pollutant Load Reductions
10.  ESTIMATING  POLLUTANT LOAD REDUCTIONS

 10.1.   OVERVIEW  OF APPROACH

Estimating the performance of the variety of erosion and sediment controls (ESCs) likely to be
employed at construction sites, given the array of site conditions, site geometries, soil types and
rainfall conditions nationally is an extremely complicated undertaking. Models do exist (such as
SEDCAD and SEDIMOT III) that can be used to estimate, for a given set of site conditions and
for a given storm event,  the sediment generation, sediment transport, and sediment removals
through BMPs. However, a significant amount of data regarding site conditions, watershed
parameters and design features of various control structures is needed. Given the range of
possible conditions nationally, U.S. Environmental Protection Agency (EPA) determined that
basing a national loading reductions estimate on an input parameter-intensive model such as
SEDCAD was not feasible for this analysis. Therefore, EPA developed a relatively
straightforward approach to estimate loading reductions estimates for the regulatory options
considered.

EPA used a model site approach to estimate baseline sediment loads and to estimate loading
reductions for the Construction and Development (C&D) industry under the regulatory options
evaluated. EPA used the Revised Universal Soil Loss Equation (RUSLE) to estimate loads and
load reductions at the Reach File Version 1.0 (RF1) scale. This approach consisted of the
following steps:
   1. Developing a series of model projects of differing sizes, durations, and project types on
      the basis of analysis of Notice of Intent (NOI) data.
   2. Determining RF1-level estimates for RUSLE and hydrologic parameters using national
      geographic information system (GIS) data layers, supplemented with best professional
      judgment (BPJ) estimates for parameters for which data were not available.
   3. Estimating baseline and option-specific estimates of sediment loads for each RF1. For
      Option 1, estimates were developed according to changes in the RUSLE P- and C factors
      from baseline. For Options 2, 3, and 4, estimates were developed first using the change in
      RUSLE P- and C-factors for all sites to account for the effects of the enhanced erosion
      and sediment control requirements, and second a concentration approach for acres subject
      to turbidity limits.
   4.  Summing RF1 loads to the national level.

The following sections describe these steps in detail.

 10.2.   MODEL PROJECT ANALYSIS

EPA evaluated NOI data from four states and developed a distribution of projects by site size,
project duration, and project type (see Appendix C). EPA categorized NOI data into three main
project types:  residential, nonresidential, and transportation. On the basis of the NOI data
evaluation, EPA developed the distribution shown in Table 10-1. For the loads analysis, it was
not necessary  to maintain the breakout by project durations because longer duration projects are
                                         10-1

-------
                                                   Section 10: Estimating Pollutant Load Reductions
calculated using the same methodology as shorter duration projects. Therefore, one project
duration was calculated for each of the 12 model project sizes within each of the three project
type categories (residential, nonresidential, and transportation), yielding a total of 36 individual
model projects (12 site size categories times 3 project types). As with the cost analysis, it was
assumed that the duration of land disturbance would be less than the project duration according
to the NOIs. Therefore, the duration of each  model project was determined using BPJ, and the
duration for each of the 36 site size categories were determined by collapsing the model project
matrix. Table 10-2 shows the collapsed distribution and the duration of both the NOIs as well as
the calculated project duration for each of the 36 model project size categories used for the
loading estimates.

  10.3.   MODEL PARAMETER AND LOADS ESTIMATION

Sediment loads were estimated using the RUSLE. RUSLE is an empirical relationship that can
be used to estimate soil erosion rates from various land uses.  RUSLE calculates soil loss, A (in
tons/acre/year), on the basis of six parameters using the following relationship:

A=RxKxLxSxCxP

The parameters in RUSLE are

A =    Average annual soil loss (tons/acre/year)
R =    rainfall-runoff erosivity factor
K =    soil erodibility factor
L =    slope length factor
S =    slope steepness factor
C =    cover-management factor
P =    support practice factor

EPA used a combination of data sources as well as BPJ in selecting RUSLE parameters. EPA's
load estimation methodology calculates soil  loss at the RF1 scale. Therefore, EPA used national
databases, where available, to determine some parameters that are geographically based. Other
parameters are site-specific. Therefore, EPA estimated the parameters by applying BPJ to
various data sources. EPA assumed a delivery ration of 1 (i.e., all estimates by RUSLE were
assumed to be discharged from the construction site). That assumption does not account for
losses that could occur if, for example, sediment were to be deposited between the construction
site and the storm drain or receiving water. This is a reasonable assumption, however, because
discharges from construction sites (particularly larger sites that comprise the bulk of the acres
affected) are typically discharged via a pipe or channel directly to a storm drain  network or to
receiving water. In addition, the SPARROW model accounts for some potential losses as
sediment is delivered to the RF1 stream network.
                                          10-2

-------
Table 10-1. Model project matrix
RESIDENTIAL
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
3
46
69
85
145
Total
Residential
Duration (days)
0-46
124
68
44
22
23
18
1
1
11
-
—
—
312
47-91
147
100
28
26
14
2
28
—
-
-
—
—
345
92-182
632
351
242
187
195
84
107
48
33
1
—
8
1,888
183-274
571
341
168
172
218
174
133
59
84
38
2
11
1,971
275-365
1,111
703
444
318
506
209
239
182
126
38
17
45
3,938
366-456
926
657
301
309
388
219
303
143
114
54
36
50
3,500
457-547
283
312
81
52
139
95
73
23
11
17
—
15
1,101
549-639
293
213
110
70
78
107
90
42
24
19
7
12
1,065
640-730
222
249
169
109
72
74
103
18
70
18
7
37
1,148
731-
912
450
397
220
180
430
360
325
214
155
60
43
50
2,884
913-
1,095
98
125
76
68
75
77
101
69
101
11
27
39
867
> 1 ,096
57
177
109
167
283
137
307
185
192
117
103
77
1,911
Total by
site size
4,914
3,693
1,992
1,680
2,421
1,556
1,810
984
921
373
242
344
20,930
NONRESIDENTIAL
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
3
46
69
85
145
Total
Nonresidential
Duration (days)
0-46
558
219
150
55
77
13
38
3
—
2
-
—
1,115
47-91
1,359
547
206
97
73
82
59
27
17
2
-
8
2,477
92-182
4,910
1,990
996
578
493
246
166
78
49
29
10
25
9,570
183-274
5,334
2,973
1,368
736
660
261
264
164
65
50
1
46
1 1 ,922
275-365
4,806
2,643
1,516
616
950
505
542
176
129
33
86
124
12,126
366-456
3,276
1,715
1,059
617
741
419
250
126
150
30
22
25
8,430
457-547
940
910
513
165
347
203
215
48
57
30
11
10
3,449
549-639
685
461
285
152
178
60
208
72
152
87
21
11
2,372
640-730
442
199
240
144
162
114
135
29
12
16
-
5
1,498
731-
912
592
465
188
274
291
131
142
101
131
71
51
58
2,495
913-
1,095
180
140
89
49
41
29
25
5
51
2
-
29
640
> 1 ,096
155
148
99
96
71
39
34
36
34
14
11
15
752
Total by
site size
23,237
12,410
6,709
3,579
4,084
2,102
2,078
865
847
366
213
356
56,846

-------
TRANSPORTATION
Project size
(acres)
1.9
3.8
6
8.5
12
17
23
3
46
69
85
145
Total
Transportation
NATIONAL
TOTAL
Duration (days)
0-46
82
25
16
7
8
—
5
-
—
—
-
-
143
1,570
47-91
210
87
23
15
21
3
2
-
1
—
-
-
362
3,184
92-182
629
277
184
70
70
63
31
1
2
—
7
-
1,334
12,792
183-274
418
255
138
73
109
49
36
26
19
—
-
1
1,124
15,017
275-365
308
246
170
78
39
13
6
17
9
3
-
3
892
16,956
366-456
323
146
53
78
95
15
76
6
25
9
3
4
833
12,763
457-547
136
81
45
33
52
21
40
16
21
10
-
1
456
5,006
549-639
55
60
26
89
43
2
27
4
15
4
-
9
334
3,771
640-730
132
34
-
17
17
8
14
9
9
-
2
11
253
2,899
731-912
117
58
78
7
36
64
95
5
24
5
4
24
517
5,896
913-
1,095
7
22
15
21
13
6
16
42
4
21
12
35
214
1,721
> 1 ,096
—
7
22
6
45
28
15
2
51
—
28
30
234
2,897
Total by
site size
2,417
1,298
770
494
548
272
363
128
180
52
56
118
6,696
84,472

-------
                                                  Section 10: Estimating Pollutant Load Reductions
                        Table 10-2. Project matrix for loads analysis

Project size
(acres)
1.9
3.8
6
8.5
12
17
23
34
46
69
85.1
145
Average
Residential
NOI
duration
(months)
13
15
15
16
17
18
20
21
22
23
28
23
16
Duration for
loads
estimation
(months)
10
12
12
13
14
15
16
17
17
19
22
18
13
Non residential
NOI
duration
(months)
9
10
11
12
12
12
13
14
17
17
17
15
11
Duration for
loads
estimation
(months)
8
8
9
10
10
10
11
11
14
13
13
12
9
Transportation
NOI
duration
(months)
9
10
11
13
14
16
18
20
22
23
29
29
12
Duration for
loads
estimation
(months)
8
8
9
10
11
13
14
16
18
18
24
23
10
Table 10-3 summarizes sources used by EPA for each RUSLE factor. As discussed in Section
3.5.4, EPA used CONUS-SOIL soil database (Miller and White 1998) to evaluate soil and
physical parameter values at the RF1 level. The loads analysis evaluated only RF1 watersheds
that contained new developed land between 1992 and 2001 as indicated by the National Land
Cover Dataset (NLCD). EPA developed a procedure (called masking) that would use only the
parameter values for the geographic areas that were developed between 1992 and 2001.
                   Table 10-3. Data sources used to obtain RUSLE factors
RUSLE
term
C
P
K
R
LS
Definition
Cover-
management
Factor
Support
Practice Factor
Soil Erodibility
Factor
Rainfall-Runoff
Erosivity Factor
Slope Length
Factor
Source of
information
Literature review
and BPJ
Literature review
and BPJ
CONUS-SOIL
Database
CIS layer prepared
for the EPA LEW
Calculator
BPJ
Method for determining RF1 -level values
EPA calculated an average annual value that is based on
assumptions about how cover, and associated C factors,
are likely to be employed over the duration of a typical
project. These values were determined by applying BPJ to
various sources in the literature and did not vary
geographically or across regulatory option.
EPA assigned values on the basis of a literature review and
BPJ. Values varied by assumptions about how practices
were likely to change under the regulatory options.
EPA determined the spatially averaged value from soils
data for each RF1 watershed, weighted toward areas where
development occurred between 1992 and 2001 using NLCD
data.
EPA determined the spatially averaged value from soils
data for each RF1 watershed, weighted toward areas where
development occurred between 1992 and 2001 using NLCD
data.
EPA assumed an average slope of 4% and a slope length
of 80 feet across all model projects.
                                         10-5

-------
                                                   Section 10: Estimating Pollutant Load Reductions
The array of surface conditions at constructions sites varies as construction advances from
clearing/grubbing, to earth moving/contouring, followed by the installation of structures and
final landscaping. Given the large number of construction sites commencing annually and the
large variation in construction sequences and construction activities occurring across the country,
determining site-specific RUSLE parameters for C, P, L, and S is extremely challenging.
Therefore, EPA applied BPJ to estimate values for these parameters. EPA selected values that
are expected to be typical across the entire country and, on average, could be considered typical
of conditions over the duration of the project. To evaluate the sensitivity of the loading estimates
under the primary analysis (which were used to estimate national water quality changes in the
SPARROW model and subsequent environmental benefits), EPA chose to vary the assumptions
for one parameter, P. RF1 and national loads were estimated using three assumptions for P, with
the middle (called average) value used for water quality modeling. The results for the other two
scenarios (called low and high) are included in the discussion below; however, SPARROW
model runs were not conducted for those scenarios.

   10.3.1.  LS FACTOR

As suggested by its name, the RUSLE slope length parameter is composed of a slope component
and a length component. In combination, the two components express the influence of the path
taken by runoff as it travels across a construction site to the point of discharge. The steeper and
longer the pathway, the greater the amount of erosion the RUSLE predicts .

While the LS factor is one of the most important RUSLE parameters, it is also one of the more
difficult values to determine at the national  scale. The STATSGO data used by EPA reports land
slope only in terms of a low value and a high value. Therefore, determining a representative
slope for specific geographic areas is not possible using that data set. While more detailed data
sets are available, the resources required to analyze preexisting slopes at the national scale made
the use of more detailed data sets infeasible for this analysis.

Table 10-4  indicates per-state, spatially averaged, lowest-reported, and highest-reported slope
data  extracted from the STATSGO data. The values generated through this process are  relatively
high (e.g., greater than 5 percent). Therefore, EPA elected not to use these data for modeling
purposes, but instead assumed a 4 percent land slope across all model sites, acknowledging that
slopes present on actual construction sites can be much lower (such as  if a corn field is  converted
to a big-box store) or much higher (such as  on a highway road cut). The value of 4 percent falls
toward the bottom of the ranges reported in Table 10-4.
                     Table 10-4. Slope ranges from STATSGO (percent)
State
AL
AR
AZ
CA
CO
CT
DC
Average of lowest
reported slopes
4.71
4.71
1.66
6.83
1.50
5.21
6.69
Average of highest
reported slopes
11.96
11.91
8.24
17.37
11.33
11.89
14.05
State
NC
ND
NE
NH
NJ
NM
NV
Average of lowest
reported slopes
5.37
2.49
2.95
6.25
2.16
1.60
2.14
Average of highest
reported slopes
11.53
6.98
7.29
14.93
6.42
10.99
8.99
                                          10-6

-------
                                                   Section 10: Estimating Pollutant Load Reductions
State
DE
FL
GA
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
Ml
MN
MO
MS
MT
Average of lowest
reported slopes
0.93
0.32
4.49
3.60
2.88
2.79
2.12
1.71
11.04
1.36
4.82
4.89
4.04
2.11
2.24
5.25
5.15
3.07
Average of highest
reported slopes
3.66
3.24
10.40
6.95
10.83
6.80
5.96
5.31
24.63
4.78
12.55
11.66
11.48
7.80
6.60
12.71
12.52
13.50
State
NY
OH
OK
OR
PA
Rl
SC
SD
TN
TX
UT
VA
VT
WA
Wl
WV
WY
National
Average of lowest
reported slopes
4.37
3.99
2.25
9.82
6.86
3.69
2.72
1.99
8.00
0.89
3.19
7.51
9.19
8.13
3.25
15.79
1.32
3.65
Average of highest
reported slopes
11.61
9.90
7.08
24.08
17.27
10.64
6.40
6.83
17.73
5.05
14.80
17.13
21.79
21.46
9.01
31.27
11.00
10.18
EPA assumed a relatively short distance of 80 feet across all its model sites for the slope length.
That distance is intended to represent the typical density of channels for a 4 percent slope used to
facilitate drainage on the site before the permanent drainage network is installed, or the length
before some sort of slope break would be provided. In general, the permanent drainage
infrastructure is installed early in the construction process, and it is available to drain the
construction footprint around individual structures at some point during construction.

For its analysis of construction site erosion, EPA computed LS on the basis of a high rill-to-
interrill  erosion ratio. That assumption is generally considered the most appropriate assumption
for construction site conditions (USDA  1997).

    10.3.2.  P FACTOR

To represent the influence of various ESC technologies, EPA used the RUSLE practice factor
(P). Examination of the literature indicates that P factors for sediment controls at construction
sites can vary from between 0.1 to 0.9 (see Table 10-5). This means that sediment controls can
vary from between 10 percent to 90 percent effective in removing sediment. In reality, the
performance of a given sediment control is dependent on a number of factors, including design,
size, frequency, and duration of rainfall  and runoff events; particle size distribution of sediment
particles; the presence of particle surface charge; influent sediment concentration; the degree  of
sediment accumulation;  and the extent to which maintenance has been performed. To estimate
loads for the national estimates, EPA chose to assign a P factor of 0.4 to characterize baseline
sediment control performance. This means that 60 percent of the sediment estimated to be
produced at the site is removed through sediment controls such as sediment traps, sediment
basins, silt fences, and check dams. The value is intended to be typical of the range of practices
used at sites nationwide, recognizing that different sites employ a mix  of sediment control
practices. To evaluate the influence of the ESC requirements under Options 1 through 4, EPA
                                           10-7

-------
                                                   Section 10: Estimating Pollutant Load Reductions
reduced the P factor to 0.3. That accounts for using surface outlets on all basins, which is the
major change to sediment controls from existing requirements contained in the nonnumeric
effluent limitations of the regulatory options. EPA also assumed that filter berms will be used in
place of silt fence, which has an associated improvement in sediment removal. As a sensitivity
analysis, EPA varied the baseline P factor assumption. EPA assumed a baseline P factor of 0.3 as
the low-value and 0.5 as the high-value in the sensitivity analysis. As with the primary case, EPA
assumed that the regulatory options would reduce the P factor by 0.1 units. Table 10-6
summarizes the P factor assumptions.
                     Table 10-5. P factors for construction site practices
Practice
Sediment Containment Systems
Bale or Sandbag Barriers
Rock Barriers at Sump Locations
Silt Fence Barrier
P factor
0.1-0.9
0.9
0.8
0.6
Grass Buffer Strips
0% to 10% Slope
11% to 24% Slope
0.6
0.8
                        Table 10-6. P factors used for load estimation
Scenario
Baseline
Regulatory Options
Low value
0.3
0.2
Average value
0.4
0.3
High value
0.5
0.4
    10.3.3.  C FACTOR

EPA used the RUSLE cover factor (C) to represent various cover conditions present on the
model sites under baseline conditions and under the regulatory options. Table 10-7 shows C
factors for various construction site controls. Permittees are likely to implement various types of
cover practices on the site. During clearing and grading, substantial portions of the site can be
disturbed and bare soil can be prevalent. As portions of the site reach final grade, permittees
usually install some sort of temporary cover, such as straw mulch or temporary seeding. As
construction progresses, temporarily stabilized areas could be exposed again during excavation
activities,  and eventually the site is stabilized at the end of construction. However,  such practices
vary widely nationwide and even within states.

To determine an appropriate C factor for use in the national modeling, EPA developed an
average annual C factor for sites nationwide. The average C factor was determined by making
assumptions about the types of cover present on construction sites during different periods of
construction. For a 1-year duration project under baseline conditions, EPA assumed that the site
would have bare soil that is loose for a period of 1 month during initial clearing and grading,
followed by a 1-month period where soil is bare and compacted. Straw mulch would then be
applied at a rate of 2 tons per acre for a period of 9 months during the vertical construction
                                           10-8

-------
                                                   Section 10: Estimating Pollutant Load Reductions
phase. The remaining 1 month would be seeded and mulched with grasses established. That
gives an average C factor of 0.26 for the year.

Under Options 1, 2, 3, and 4, EPA assumed that the period of compacted, bare soil would be
reduced from 1 month to 2 weeks and that the period when the site is covered in straw mulch
would increase from 9 months to 9.5 months. That change is intended to reflect the effect of the
enhanced soil cover requirement of the options, meaning that permittees would install temporary
cover on average 2 weeks earlier than under baseline conditions. That gives an average C factor
for the year of 0.23.

Table 10-8 shows the assumptions used under baseline conditions and under the regulatory
options and the value of the average annual C factors.
                      Table 10-7. C factors for construction site controls
Treatment
Bare soil conditions













Freshly disked to 6-8 inches
After one rain
Loose to 12 inches, smooth
Loose to 12 inches, rough
Compacted root rake
Compacted bulldozer scraped across slope
Same except root raked across
Rough irregular tracked all directions
Seed and fertilized, fresh unprepared seedbed
Same except after 6 months
Seed, fertilized after 12 months
Undisturbed except scraped
Scarified only
Asphalt/Concrete Pavement
Asphalt emulsion


1,210 gal/acre
605 gal/acre
Gravel (diameter = 25-50 mm) at 90 tons/acre
Dust binder


605 gal/acre
1,210 gal/acre
Other chemicals




Aquatain
Aerospray 70, 10% cover
PVA
Tera-Tack
Straw Mulch


1 ton/acre (slopes less than 1 0%)
1 .5 ton/acre (slopes less than 1 0%)
Seeding



Temporary, 0-60 days
Temporary, after 60 days
Permanent, 2 to 12 months
C factor value

1
0.89
0.9
0.8
1.2
1.2
0.9
0.9
0.64
0.54
0.38
0.66-1.30
0.76-1.31
0.01

0.01-0.019
0.14-0.57
0.05

1.05
0.29-0.78

0.68
0.94
0.71-0.90
0.66

0.2
0.12

0.4
0.05
0.05
                                          10-9

-------
                                                    Section 10: Estimating Pollutant Load Reductions
Treatment
Grass Seeding and Mulch



20% coverage by treatment
40% coverage by treatment
60% coverage by treatment
Brush
C factor value

0.20
0.10
0.042
0.35
Wischmeier and Smith 1978; URS 2008
                       Table 10-8. C factors used for loads estimation
Cover
C factor
Duration
(months)
C factor x fraction of
year
Baseline
Bare soil, loose to 12 inches, rough
Bare soil, compacted
Straw mulch, 1 .5 tons/acre, 1-5% slope
Grass with mulch, 60% cover
0.8
1.2
0.12
0.042
1
1
9
1
Total for Year
0.067
0.1
0.09
0.0035
0.26
Options 1, 2, 3, and 4
Bare soil, loose to 12 inches, rough
Bare soil, compacted
Straw mulch, 1 .5 tons/acre, 1-5% slope
Grass with mulch, 60% cover
0.8
1.2
0.12
0.042
0.5
1
9.5
1
Total for Year
0.033
0.1
0.095
0.0035
0.23
    10.3.4.  RUNOFF VOLUME ESTIMATES

EPA estimated runoff volumes within each RF1 watershed to calculate removals for options that
incorporated a numeric discharge standard. EPA computed runoff coefficients on the basis of the
long-term meteorological record for 11 indicator cities. The values were then assigned to RF1
watersheds for those states. For other states that do not contain an indicator city, values from the
nearest state were used as an approximation. Table 10-9 lists the states/commonwealths
represented by each of the indicator cities. Note that the loading analysis does not include
consideration of construction activities in Hawaii and Alaska, and does not include any of the
U.S. territories.
    Table 10-9. Allocation of states/commonwealths/territories to representative indicator city
State
Alabama
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indicator city
Atlanta, GA
Las Vegas, NV
Dallas, TX
Las Vegas, NV
Denver, CO
Manchester, NH
Washington, DC
Atlanta, GA
Atlanta, GA
Boise, Id
Chicago, IL
State
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
Indicator city
Albany, NY
Dallas, TX
Albany, NY
Atlanta, GA
Denver, CO
Chicago, IL
Dallas, TX
Seattle, WA
Washington, DC
Manchester, NH
Atlanta, GA
                                          10-10

-------
                                                   Section 10: Estimating Pollutant Load Reductions
State
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
Indicator city
Chicago, IL
Kansas City, KS
Kansas City, KS
Atlanta, GA
Dallas, TX
Manchester, NH
Washington, DC
Manchester, NH
Chicago, IL
Chicago, IL
Atlanta, GA
Kansas City, KS
Denver, CO
Kansas City, KS
Las Vegas, NV
Manchester, NH
State
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Alaska
Hawaii
Puerto Rico
Virgin Islands
Pacific Islands
District of Columbia
Indicator city
Denver, CO
Atlanta, GA
Dallas, TX
Denver, CO
Manchester, NH
Washington, DC
Seattle, WA
Washington, DC
Chicago, IL
Denver, CO
Not analyzed
Not analyzed
Not analyzed
Not analyzed
Not analyzed
Washington, DC
EPA used the CONUS-SOIL data on hydrologic soil groups (HSG) to estimate runoff volumes
from the model construction sites. HSG is presented in terms of the percent of land area that is
made up of soils characterized as type A, B, C, or D. Those four soil hydrologic classifications
are correlated to the soil Curve Number, used with the SCS Curve Number methods to convert
inches of rainfall into inches of runoff.

Eleven indicator cities were used to evaluate runoff coefficients, and those values were assigned
to surrounding states using the relationships in Table 10-9. The NRCS Curve Number procedure
(TxDOT 2009) was used to estimate runoff coefficients and associated runoff volumes,
considering the Curve Number for each of the four HSGs and the distribution of HSGs within
each geographic area.

The hourly rainfall record was then  evaluated for a single year's meteorological record to
determine runoff amounts for each hour's precipitation, for each HSG. The rainfall year selected
for each indicator city was judged to be typical or a year that did not contain rainfall events with
greater than a 2-year return period.

For simplicity, the total runoff volume from all the individual rainfall events was divided by the
total annual rainfall amount. The result is a runoff coefficient that can be used to convert annual
precipitation into annual runoff.  Table 10-10 indicates the runoff coefficients for each HSG, for
each indicator city.

EPA used values in Table 10-10 to estimate the annual runoff amount for developed acres within
each RF1 watershed. For example, if an RF1 watershed near Albany New York, has equal
amounts of A, B, C, and D soils, its effective runoff coefficient is estimated as the sum of 25
percent of each of the Albany HSG  values. (0.25 x (0.12 + 0.23 + 0.34 + 0.45) or 0.285).
Multiplying the total annual precipitation associated with each RF1 watershed by the customized
per-RFl runoff coefficient, yields the estimated annual runoff amount. Additional information on
EPA's processing hydrologic data and developing Table 10-10 values is in Appendix H.
                                          10-11

-------
                                                   Section 10: Estimating Pollutant Load Reductions
Appendix H also contains information on EPA's assessment of indicator city meteorological data
to establish the number of rainfall events expected in a construction period and the duration of
discharge monitoring for runoff events.

             Table 10-10 Estimated runoff coefficients by HSG for indicator regions
City
Manchester, NH
Albany, NY
Washington, DC
Atlanta, GA
Chicago, IL
Dallas, TX
Kansas City, KS
Denver, CO
Las Vegas, NV
Boise, ID
Seattle, WA
EPA Region*
1
2
3
4
5
6
7
8
9
10a
10b
A soil
0.15
0.12
0.15
0.17
0.14
0.14
0.13
0.04
0.03
0.01
0.13
B soil
0.26
0.23
0.27
0.30
0.26
0.28
0.25
0.10
0.10
0.04
0.22
Csoil
0.36
0.34
0.39
0.41
0.37
0.41
0.37
0.18
0.18
0.09
0.32
D soil
0.46
0.45
0.49
0.52
0.47
0.52
0.47
0.27
0.26
0.16
0.42
* EPA Region 10 was divided into two portions to help account for differences in rainfall patterns.

To obtain annual precipitation for each RF1 watershed, EPA performed a spatial analysis using
the  1-km resolution U.S. Average Monthly or Annual Precipitation (1971-2000) PRISM Group
raster data coverage (PRISM Group 2006). The annual rainfall for the urbanized acres within
each RF1 watershed boundary was averaged and used to estimate the per-RFl annual rainfall
value.

 10.4.   LOAD ESTIMATION

EPA estimated loads under baseline as well as under the regulatory options evaluated. All
calculations were done at the RF1 level. Using NLCD data, EPA estimated the amount of new
development occurring in each RF1 watershed (for a description of this analysis, see the
proposed rule development document). That analysis indicates that approximately 590,545 acres
per year were developed nationally over the period of 1992 to 2001. EPA then scaled these
estimates up to account for growth in the industry  since the 1992-2001 period. EPA's revised
estimate of national developed acreage is 852,650  acres. EPA then scaled up the RF1-level
estimates of developed acres using the ratio between these two values.

Using the RF1-level parameters for K and R, and the C, P, and LS assumptions described above,
EPA calculated the baseline sediment loading for all developed acres with each RF1 watershed
on an annual basis. There were 42,288 unique RF1 watersheds in the analysis (approximately 4
percent of watersheds crossed state boundaries, so those RF1 watersheds were broken into
smaller sections to conform with the RF I/state combinations and were later recombined). All
RF1 watersheds that did not have development between 1992 and 2001 were not analyzed. The
annual values were then used to estimate the loads for each model project category on the basis
of the number of acres within each category and the duration of construction activity for the
entire national model project matrix. For model construction sites with a duration of less than 1
year, the load was calculated using the fraction of the year modeled, with no consideration for
the  actual time of year that construction occurred.  Because parameters such as R vary during the
                                          10-12

-------
                                                   Section 10: Estimating Pollutant Load Reductions
year, the soil loss during any fraction of a year can be calculated if the start and end dates of the
project are known. However, because of the large number of model projects in the analysis, it is
simply assumed that the projects are evenly distributed over the course of the year. In addition,
all loads from projects longer than 1 year were estimated by scaling up the annual values. For
example, if a construction duration was 13 months, the total load from that model construction
site would consist of the annual load from RUSLE (12 months), plus 1/12 the annual load to
account for the 1-month incremental load.

Options 2, 3, and 4 contain site size thresholds whereby specific requirement for meeting a
numeric turbidity limit apply according to site size (10 acres for Options 3 and 4 and 30 acres for
Option 2). As with Option 1, EPA applied the changes in C and P factors to  determine the
influence of the enhanced ESCs. EPA then determined if any additional removals would result
from the turbidity limits using a concentration approach. That was done by dividing the sediment
load by the calculated runoff volume from developed acres  within each RF1 watershed. For
Options 2 and 3, it was assumed that the average total suspended solids (TSS) concentration for
discharges subject to the numeric limit would be 25 milligrams per liter (mg/L). For Option 4, it
was assumed that the average TSS concentration of discharges would be 250 mg/L. For each
RF1 watershed, the discharge load was calculated on the basis of the difference between the
concentration after application of BMPs and either 25 mg/L (Options 2 and 3) or 250 mg/L. If
the baseline concentration was less than either 25 mg/L or 250 mg/L, no removals were
associated with the numeric limit for that RF1 watershed. From the distribution of site sizes and
durations, loads were then calculated on the basis of the quantity of acres within each site size
category and the duration of construction activity for each model project category within the
national model project matrix, using the same procedure described above for Option 1.

Load reductions were summed for each RF1 watershed and then summed to the state and
national level. RFl-level estimates were used for subsequent water quality modeling using the
SPARROW model.

 10.5.   RESULTS

Table 10-11 provides estimates of sediment discharges under baseline conditions and the
regulatory options for the primary analysis case. Tables 10-12 and 10-13 provide estimates of
sediment discharges for the low and high scenarios, respectively. Table  10-14 provides the
estimated sediment removals by regulatory option for the primary analysis case. Tables 10-15
and 10-16 provide estimated removals for the low and high scenarios, respectively. All values
presented in these tables are after full implementation.
                                         10-13

-------
                                  Section 10: Estimating Pollutant Load Reductions
Table 10-11. Discharged loads—primary analysis case
State
AL
AK
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
Ml
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
Rl
SC
SD
TN
TX
UT
VA
VT
WA
Wl
Tons per year
Baseline
120,513
-
78,969
10,813
24,516
12,222
2,878
1,403
3,459
150,699
180,561
—
39,324
1,812
89,672
53,875
101,100
51 ,628
196,969
6,258
29,790
7,376
31,647
17,639
86,647
145,295
3,657
92,858
8,576
14,553
3,246
14,378
2,665
2,225
17,302
62,997
101,097
9,119
53,686
965
79,666
15,388
88,740
460,238
2,631
63,323
1,209
15,752
17,363
Option 1
79,956
-
52,393
7,174
16,266
8,109
1,910
931
2,295
99,983
119,795
—
26,090
1,202
59,494
35,744
67,076
34,253
130,682
4,152
19,765
4,894
20,997
11,703
57,487
96,397
2,426
61,608
5,690
9,655
2,154
9,540
1,768
1,476
1 1 ,479
41,796
67,074
6,050
35,619
640
52,855
10,209
58,876
305,350
1,746
42,012
802
10,451
1 1 ,520
Option 2
36,218
-
23,660
3,225
7,414
3,684
886
422
1,049
46,575
54,635
—
1 1 ,840
547
27,040
16,293
30,325
15,639
58,732
1,965
8,997
2,358
9,833
5,402
25,941
43,421
1,111
28,141
2,566
4,376
1,016
4,383
829
676
5,340
19,176
30,270
2,903
16,418
298
24,163
4,589
26,754
137,702
821
19,199
376
4,996
5,289
Option 3
15,419
-
9,996
1,348
3,204
1,580
399
180
457
21,177
23,648
—
5,063
236
1 1 ,606
7,044
12,849
6,786
24,518
925
3,877
1,153
4,524
2,405
10,940
18,228
486
12,227
1,080
1,865
476
1,930
382
296
2,421
8,419
12,768
1,406
7,288
135
10,519
1,917
1 1 ,478
57,978
381
8,351
173
2,402
2,326
Option 4
25,945
-
15,925
1,970
5,944
2,785
925
311
884
51,377
44,336
—
8,783
427
20,640
13,109
20,906
12,925
35,571
2,588
7,068
3,673
11,515
5,276
17,191
27,665
965
23,428
1,674
3,166
1,303
3,971
957
591
5,789
17,149
20,103
4,251
15,750
327
20,410
2,796
20,350
90,021
996
16,128
446
7,090
4,777
                        10-14

-------
                                     Section 10: Estimating Pollutant Load Reductions
State
WY
WV
NATIONAL
Tons per year
Baseline
1,146
1 1 ,729
2,589,577
Option 1
760
7,782
1,718,085
Option 2
349
3,593
781,433
Option 3
153
1,600
336,018
Option 4
310
3,521
604,009
Table 10-12. Discharged loads—low sensitivity analysis case
State
AL
AK
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
Ml
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
Rl
SC
SD
Tons per year
Baseline
90,385
—
59,227
8,109
18,387
9,166
2,159
1,052
2,594
113,024
135,420
-
29,493
1,359
67,254
40,406
75,825
38,721
147,727
4,694
22,343
5,532
23,735
13,229
64,986
108,971
2,743
69,644
6,432
10,915
2,434
10,784
1,999
1,669
12,976
47,248
75,823
6,839
40,265
723
59,750
11,541
Option 1
53,304
—
34,929
4,783
10,844
5,406
1,273
621
1,530
66,655
79,863
-
17,393
801
39,662
23,829
44,717
22,836
87,121
2,768
13,176
3,263
13,998
7,802
38,325
64,265
1,618
41,072
3,793
6,437
1,436
6,360
1,179
984
7,653
27,864
44,716
4,033
23,746
427
35,237
6,806
Option 2
24,409
—
15,922
2,166
5,011
2,487
604
285
710
31,808
36,942
-
7,987
370
18,253
11,015
20,419
10,580
39,432
1,352
6,078
1,636
6,732
3,673
17,451
29,184
753
19,043
1,725
2,950
698
2,974
568
458
3,645
13,003
20,364
2,009
11,158
203
16,357
3,082
Option 3
10,669
—
6,884
922
2,238
1,098
285
125
320
15,237
16,532
-
3,513
164
8,072
4,921
8,864
4,752
16,755
678
2,703
862
3,277
1,710
7,525
12,501
342
8,567
742
1,292
348
1,363
277
208
1,739
5,936
8,784
1,046
5,172
97
7,379
1,311
Option 4
21,195
—
12,813
1,542
4,954
2,276
812
256
746
44,876
37,219
-
7,233
355
17,105
10,986
16,921
10,890
27,806
2,333
5,892
3,153
10,127
4,562
13,776
21,939
813
19,757
1,336
2,593
1,174
3,381
812
502
5,074
14,666
16,118
3,582
13,631
288
17,271
2,189
                           10-15

-------
                                      Section 10: Estimating Pollutant Load Reductions
State
TN
TX
UT
VA
VT
WA
Wl
WY
WV
NATIONAL
Tons per year
Baseline
66,555
345,178
1,973
47,492
907
11,814
13,023
859
8,797
1,942,181
Option 1
39,251
203,567
1,164
28,008
535
6,967
7,680
507
5,188
1,145,390
Option 2
18,058
92,605
563
12,995
258
3,452
3,587
237
2,443
527,695
Option 3
7,981
39,839
277
5,855
126
1,781
1,641
108
1,138
233,956
Option 4
16,852
71,855
855
13,632
397
5,914
4,086
261
3,056
499,863
Table 10-13. Discharged loads—high sensitivity analysis case
State
AL
AK
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
Ml
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
Tons per year
Baseline
150,642
-
98,711
13,516
30,645
15,278
3,598
1,754
4,324
188,374
225,701
-
49,155
2,265
112,090
67,343
126,376
64,536
246,212
7,823
37,238
9,220
39,559
22,048
108,309
181,618
4,571
116,073
10,720
18,191
4,057
17,973
3,332
2,781
21,627
Option 1
106,608
-
69,857
9,565
21,688
10,812
2,546
1,241
3,060
133,311
159,727
-
34,787
1,603
79,325
47,658
89,435
45,671
174,242
5,536
26,353
6,525
27,996
15,603
76,649
128,530
3,235
82,144
7,586
12,874
2,871
12,719
2,358
1,968
15,305
Option 2
48,026
-
31,397
4,285
9,816
4,882
1,168
560
1,388
61,341
72,327
-
15,693
725
35,826
21,572
40,232
20,697
78,032
2,578
11,916
3,081
12,934
7,130
34,431
57,657
1,470
37,240
3,406
5,802
1,335
5,791
1,090
894
7,035
Option 3
20,168
-
13,108
1,774
4,170
2,062
512
236
593
27,116
30,765
-
6,613
307
15,141
9,167
16,834
8,821
32,281
1,171
5,051
1,443
5,771
3,100
14,355
23,954
630
15,886
1,418
2,439
604
2,497
487
384
3,103
Option 4
30,695
-
19,037
2,397
6,916
3,275
1,039
367
1,021
57,339
51,453
-
10,333
499
24,175
15,232
24,890
14,959
43,335
2,835
8,244
3,967
12,786
5,973
20,606
33,391
1,111
27,093
2,012
3,740
1,431
4,552
1,076
679
6,475
                            10-16

-------
                                  Section 10: Estimating Pollutant Load Reductions
State
OH
OK
OR
PA
Rl
SC
SD
TN
TX
UT
VA
VT
WA
Wl
WY
WV
NATIONAL
Tons per year
Baseline
78,746
126,372
11,398
67,108
1,206
99,583
19,235
110,926
575,297
3,289
79,153
1,512
19,690
21,704
1,432
14,662
3,236,971
Option 1
55,728
89,432
8,067
47,492
853
70,474
13,612
78,501
407,134
2,328
56,016
1,070
13,934
15,360
1,014
10,376
2,290,780
Option 2
25,348
40,176
3,796
21,679
392
31,969
6,097
35,449
182,798
1,079
25,404
495
6,539
6,990
461
4,742
1,035,172
Option 3
10,902
16,753
1,765
9,404
173
13,658
2,524
14,976
76,117
485
10,847
221
3,023
3,010
199
2,063
438,079
Option 4
19,632
24,087
4,683
17,866
365
23,550
3,402
23,847
108,163
1,108
18,624
494
7,817
5,462
358
3,985
706,378
Table 10-14. Sediment removals—primary analysis case
State
AL
AK
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
Ml
MN
MO
MS
MT
NC
Tons per year
Option 1
40,557
-
26,576
3,639
8,251
4,113
969
472
1,164
50,716
60,766
—
13,234
610
30,178
18,131
34,024
17,375
66,288
2,106
10,026
2,482
10,651
5,936
29,160
48,897
1,231
31,250
Option 2
84,295
-
55,309
7,587
17,103
8,538
1,993
981
2,410
104,125
125,926
—
27,485
1,265
62,632
37,581
70,775
35,990
138,237
4,294
20,793
5,018
21,814
12,237
60,706
101,874
2,546
64,717
Option 3
105,094
-
68,973
9,465
21,312
10,642
2,479
1,222
3,002
129,523
156,912
—
34,261
1,576
78,065
46,831
88,251
44,842
172,452
5,334
25,913
6,224
27,123
15,233
75,708
127,067
3,171
80,632
Option 4
94,568
-
63,044
8,842
18,572
9,437
1,953
1,092
2,575
99,323
136,224
—
30,541
1,385
69,031
40,766
80,195
38,704
161,398
3,671
22,722
3,703
20,132
12,363
69,456
117,630
2,692
69,431
                        10-17

-------
                                     Section 10: Estimating Pollutant Load Reductions
State
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
Rl
SC
SD
TN
TX
UT
VA
VT
WA
Wl
WY
WV
NATIONAL
Tons per year
Option 1
2,886
4,898
1,092
4,839
897
749
5,823
21,201
34,023
3,069
18,067
325
26,811
5,179
29,865
154,888
886
21,311
407
5,301
5,844
386
3,947
871,492
Option 2
6,010
10,177
2,230
9,996
1,837
1,549
1 1 ,962
43,821
70,827
6,216
37,268
667
55,503
10,798
61,987
322,536
1,811
44,123
833
10,756
12,075
797
8,137
1,808,143
Option 3
7,496
12,688
2,770
12,448
2,283
1,929
14,881
54,578
88,329
7,713
46,398
829
69,147
13,471
77,262
402,260
2,251
54,971
1,036
13,350
15,038
992
10,129
2,253,559
Option 4
6,901
1 1 ,387
1,943
10,408
1,709
1,634
11,513
45,848
80,995
4,868
37,936
638
59,256
12,592
68,391
370,217
1,635
47,195
763
8,662
12,586
835
8,209
1,985,567
Table 10-15. Sediment removals—low sensitivity analysis case
State
AL
AK
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
Tons per year
Option 1
37,081
—
24,298
3,327
7,543
3,761
886
432
1,064
46,369
55,557
—
12,100
557
27,591
16,577
31,108
15,886
60,606
1,926
9,166
2,269
Option 2
65,975
—
43,305
5,944
13,376
6,680
1,555
767
1,884
81,216
98,478
—
21,507
989
49,001
29,392
55,406
28,141
108,295
3,342
16,264
3,896
Option 3
79,716
—
52,343
7,188
16,149
8,068
1,873
927
2,274
97,787
118,889
—
25,980
1,194
59,182
35,485
66,961
33,970
130,972
4,016
19,640
4,670
Option 4
69,190
—
46,414
6,568
13,433
6,890
1,347
796
1,848
68,149
98,202
—
22,260
1,004
50,148
29,420
58,904
27,832
119,921
2,361
16,450
2,379
                           10-18

-------
                                      Section 10: Estimating Pollutant Load Reductions
State
Ml
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
Rl
SC
SD
TN
TX
UT
VA
VT
WA
Wl
WY
WV
NATIONAL
Tons per year
Option 1
9,738
5,427
26,661
44,706
1,125
28,572
2,639
4,478
999
4,424
820
685
5,324
19,384
31,107
2,806
16,519
297
24,513
4,735
27,305
141,612
810
19,484
372
4,847
5,343
352
3,609
796,791
Option 2
17,004
9,556
47,535
79,787
1,990
50,601
4,706
7,965
1,736
7,810
1,431
1,210
9,331
34,245
55,459
4,830
29,107
520
43,393
8,459
48,497
252,573
1,410
34,497
650
8,362
9,435
622
6,354
1,414,486
Option 3
20,459
11,519
57,461
96,470
2,401
61,077
5,690
9,623
2,087
9,420
1,722
1,461
1 1 ,237
41,312
67,039
5,792
35,093
626
52,371
10,230
58,575
305,340
1,696
41,637
781
10,033
11,381
751
7,659
1,708,225
Option 4
13,609
8,667
51,209
87,032
1,930
49,887
5,095
8,322
1,261
7,402
1,187
1,167
7,902
32,581
59,705
3,257
26,634
435
42,479
9,352
49,703
273,323
1,118
33,860
511
5,900
8,936
598
5,741
1,442,318
Table 10-16. Sediment removals—high sensitivity analysis case
State
AL
AK
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
Tons per year
Option 1
44,034
—
28,854
3,951
8,958
4,466
1,052
513
1,264
55,063
65,974
—
14,369
662
32,765
19,685
36,941
Option 2
102,615
—
67,314
9,231
20,830
10,396
2,430
1,194
2,936
127,033
153,374
—
33,463
1,540
76,264
45,771
86,144
Option 3
130,473
—
85,603
1 1 ,742
26,475
13,216
3,086
1,518
3,731
161,258
194,936
—
42,543
1,958
96,949
58,176
109,542
Option 4
119,947
—
79,674
11,119
23,729
12,002
2,559
1,387
3,302
131,035
174,247
—
38,822
1,766
87,915
52,111
101,485
                            10-19

-------
                                                     Section 10: Estimating Pollutant Load Reductions
State
KY
LA
MA
MD
ME
Ml
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
Rl
SC
SD
TN
TX
UT
VA
VT
WA
Wl
WY
WV
NATIONAL
Tons per year
Option 1
18,864
71,970
2,287
10,885
2,695
1 1 ,563
6,445
31,659
53,088
1,336
33,929
3,133
5,317
1,186
5,254
974
813
6,322
23,018
36,939
3,332
19,616
352
29,109
5,622
32,425
168,164
961
23,137
442
5,756
6,344
419
4,286
946,192
Option 2
43,838
168,179
5,245
25,321
6,139
26,625
14,918
73,878
123,961
3,102
78,833
7,314
12,389
2,723
12,182
2,242
1,887
14,592
53,398
86,195
7,602
45,429
813
67,614
13,138
75,477
392,499
2,211
53,749
1,017
13,151
14,714
971
9,920
2,201,800
Option 3
55,714
213,931
6,652
32,186
7,777
33,788
18,948
93,954
157,664
3,941
100,187
9,302
15,752
3,454
15,476
2,845
2,398
18,524
67,845
109,619
9,633
57,704
1,032
85,924
16,711
95,950
499,180
2,805
68,306
1,291
16,667
18,694
1,233
12,599
2,798,892
Option 4
49,576
202,876
4,988
28,994
5,253
26,773
16,075
87,703
148,227
3,461
88,980
8,707
14,451
2,627
13,421
2,256
2,102
15,152
59,114
102,284
6,716
49,242
841
76,032
15,833
87,079
467,135
2,182
60,530
1,018
11,873
16,242
1,074
10,677
2,530,594
Table 10-17 provides the reductions of sediment discharged for the nation under each regulatory
option, the percent reduction for the primary analysis, and the sensitivity analysis, after full
implementation.

                Table 10-17. National sediment reductions for regulatory options
Load reduction
(tons)

Option 1
Option 2
Option 3
Option 4
Low-end estimate
796,791
1,414,486
1,708,225
1,442,318
Average estimate
871 ,492
1,808,143
2,253,559
1,985,567
High-end estimate
946,192
2,201,800
2,798,892
2,530,594
                                           10-20

-------
                                                  Section 10: Estimating Pollutant Load Reductions
Load reduction
(billion pounds)

Option 1
Option 2
Option 3
Option 4
Low-end estimate
1.594
2.829
3.416
2.885
Average estimate
1.743
3.616
4.507
3.971
High-end estimate
1.892
4.404
5.598
5.061
Percent load reduction

Option 1
Option 2
Option 3
Option 4
Low-end estimate
41%
73%
88%
74%
Average estimate
34%
70%
87%
77%
High-end estimate
29%
68%
86%
78%
The RF1-level estimates of baseline discharges and discharges under the regulatory options were
used as inputs to the SPARROW model to estimate changes in sediment flux in the nation's RF1
river network. EPA used, as inputs to the SPARROW model, only the set of RFls watersheds
that have 1 or more acres of annual development. Of the 42,288 state/RFl combinations in the
model, there were 40,591 individual RF1 watersheds. Of those, 33,083 had 1 or more acres of
development. The total loads modeled in SPARROW are as shown in Table 10-18. The total
number of acres represented in the SPARROW loads is 848,986, which represents 99.6 percent
of the total annual acres estimated to be developed (852,650).
              Table 10-18. Total discharge loads and loads modeled in SPARROW

Baseline
Option 1
Option 2
Option 3
Option 4
All RF1s
2,589,577
1,718,085
781,433
336,018
604,009
RF1s modeled in SPARROW
2,582,272
1,713,238
779,217
335,053
602,164
Results of the SPARROW modeling are in The Environmental Impact and Benefits Assessment
for Final Effluent Guidelines and Standards for the Construction and Development Category
(EPA 2009b), which discusses the results of the SPARROW modeling and the monetized
benefits of the regulatory options. The entire loading analysis is in the C&D Load Spreadsheet
Model (DCN 43121).

  10.6.   REFERENCES

Miller, D.A., and R.A. White. 1998. A conterminous United States multilayer soil characteristics
    data set for regional climate and hydrology modeling. Earth Interactions 2(2): 1-26.
                                         10-21

-------
                                                  Section 10: Estimating Pollutant Load Reductions
TxDOT (Texas Department of Transportation). 2009. NRCS Runoff Curve Number Methods.
   Chapter 5, section 7 in Hydraulic Design Manual. Texas Department of Transportation, Austin,
   TX. . Accessed September 27, 2009.

PRISM Group. 2006. United States Average Monthly or Annual Precipitation (1971-2000).
   Oregon State University, Corvallis, OR. . Accessed March
   2008.

USEPA (U.S. Environmental Protection Agency) 2009b. Environmental Impact and Benefits
   Assessment for Final Effluent Guidelines and Standards for the Construction and
   Development Category (EPA-821-R-09-012).

URS (URS Corporation). 2008. Analysis of Draft General Permit Risk Factors.
   . Accessed September 27, 2009.

USDA (U.S. Department of Agriculture). 1997. Predicting Soil Erosion by Water: A Guide to
   Conservation Planning With the Revised Universal Soil Loss Equation (RUSLE). Agriculture
   Handbook Number 703. U.S. Department of Agriculture, Washington, DC.

Wischmeier, W.H., and D.D. Smith. 1978. Predicting Rainfall Erosion Losses -A Guide to
   Conservation Planning. AH537. U.S. Department of Agriculture, Washington, DC.
                                         10-22

-------
                                             Section 11: Non-Water Quality Environmental Impacts
11.  NON-WATER QUALITY ENVIRONMENTAL IMPACTS

Sections 304(b) and 306(b) of the Clean Water Act require the U.S. Environmental Protection
Agency (EPA) to consider non-water quality environmental impacts (including energy
requirements) associated with effluent limitations guidelines and standards. In accordance with
those requirements, EPA has considered the potential impacts of the options on energy
consumption, solid waste generation, and air emissions. The estimates of the impacts for the
construction and development (C&D) industry are summarized in Sections 11.1, 11.2, and 11.3.
Additional information on the calculation of the estimates is in DCN 43111 in the Administrative
Record.

 11.1.   ENERGY REQUIREMENTS

EPA considered the additional energy requirements attributable to the regulatory options
(Section 11.1.1) and the production of treatment chemicals (Section 11.1.2) and compared the
option energy requirements with the energy requirements of the C&D industry (Section 11.1.3).

   11.1.1.  ENERGY REQUIREMENTS ATTRIBUTABLE TO THE REGULATORY
            OPTIONS

EPA estimates that additional energy requirements attributable to the regulatory options being
considered are the result of the additional sediment removed from basins, traps, and other areas
of accumulation. In addition, Options 2 and 3, which rely on the use of active treatment systems
(ATS), would have additional energy requirements for operating pumps and generators. EPA
assumes that diesel powered generators and pumps would be used to operate ATS. For a 500-
gallon-per-minute (gpm) system, fuel consumption is approximately 10 gallons per hour (Rain
for Rent 2008). Table 11-1 presents estimates of energy usage by regulatory option considered.
Under Option 4, a small amount of energy could be required if metering pumps are used for
introducing liquid polymer. EPA has not quantified energy usage for the pumps, but the amount
of energy usage is expected to be minimal. The passive treatment technologies of Option 4
generally rely on gravity, so can be configured so as to utilize gravity flow of water through
channels and basins, so the significant use of pumps and generators is not anticipated. However,
permittees may utilize pumping to move water around construction sites and for dewatering
trenches and excavations, but EPA has not quantified potential energy usage for pumping as the
need for pumping would be highly site-specific.
                                        11-1

-------
                                          Table 11-1. Estimated energy consumption by regulatory option
Option
1
2
3
4
Sediment
removal
(tons/year)
871,492
1,808,143
2,253,559
1,985,567
Sediment excavation
Equipment
run time for
excavation
(hours/year)
3,320
6,888
8,585
7,564
Excavator
fuel
consumption
(gallons/year)
28,552
59,238
73,831
65,051
On-site trucking
#of
truckloads
24,900
51,661
64,387
56,730
Truck fuel
consumption
(gallons/year)
4,980
10,332
12,877
1 1 ,346
Active treatment
Water volume
treated (billion
gallons/ year)
N/A
180
273
N/A
Equipment
run time
(hours/year
@500 gpm)
N/A
5,989,567
9,090,525
N/A
Fuel
consumption
(gallons @ 10
gallons/hr)
N/A
59,895,567
90,905,253
N/A
Total fuel
consumption
(gallons)

33,532
59,965,244
90,991,961
76,397
IV)

-------
                                               Section 11: Non-Water Quality Environmental Impacts
   11.1.2.  TREATMENT CHEMICAL PRODUCTION

EPA considered the availability and additional energy consumption from treatment chemicals.
Section 11.1.2.1 describes chitosan and Section 11.1.2.2 describes polyacrylamides (PAMs).
EPA expects that under Options 2 and 3, chitosan would primarily be used. Under Option 4,
EPA anticipates a mix of both chitosan and PAMs. Results for both 100 percent chitosan use and
100 percent PAM use are presented for Option 4.

      11.1.2.1.  Chitosan

Chitosan is derived from chitin, the major component of crustacean shells and is a cationic
polyelecrolyte. Chitosan (poly-D-glucosamine) is one of the most common polymers found in
nature (USEPA 2003). Chitin is the second-most abundant natural fiber after cellulose and is
similar to cellulose in many respects (Hennen 1996). Global Industry Analysts, Inc., estimates
that the global chitin market will exceed 51.4 thousand metric tons (113 million pounds) by 2012
(Global Industry Analysts, Inc. 2008). The United States could produce approximately 30
percent of the worldwide shellfish harvest each year (Hennen 1996). Therefore, EPA estimates
that the U.S. chitin market could approach 34 million pounds by 2012.

Minton (2006) reports an average chitosan acetate dose rate of 2 milligrams per liter (mg/L).
Minton (2006) notes that the Washington State Department of Ecology specifies a maximum
dosage of 1 mg/L but that variances are granted for turbidities greater than 600 nephelometric
turbidity units (NTUs). Minton (2006) reports chitosan acetate dosages as high as 3 mg/L. Table
11-2 presents the amount of chitosan acetate required from applying a 2 mg/L chitosan acetate
dosage to the stormwater volumes (from Section 11.1.1, Table 11-1) requiring treatment. Note
that under Option 4, it is not likely that chitosan acetate would be used to treat all stormwater
generated. Nonetheless, EPA has included estimates here. The option 4 volumes are the  same as
Option 3, because the acreage threshold (10 acres) is the same.
              Table 11-2. Maximum chitosan acetate required under EPA options
Option
2
3
4
Stormwater treated
(billions of gallons)
180
273
273
Chitosan acetate required
(pounds)
3,000,000
4,560,000
4,560,000
The amount of chitosan acetate in Table 11-2 represents a fraction of the total chitin market. In
addition, EPA expects the amount of chitosan would be less than the amount presented in Table
11-2 because many construction sites would use other treatment chemical alternatives, including
PAMs, described in Section II.1.2.2.

Because chitosan is manufactured from crustacean shells and not petroleum products, additional
energy consumption from chitosan production and use is expected to be minimal.
                                          11-3

-------
                                              Section 11: Non-Water Quality Environmental Impacts
      11.1.2.2. Polyacrylamides (PAMs)

PAMs are a broad class of compounds that include cationic (positively charged) and anionic
(negatively charged) PAM. PAMs are water soluble over a wide pH range and exhibit a high
affinity for suspended sediment. PAMs are derived from acrylamide, of which 94 percent is used
as PAMs (ICIS Chemical Business 2008). U.S. demand for PAMs is presented in Table 11-3.

                        Table 11-3. U.S. acrylamide/PAMs demand*
Year
2007
20 11 (projected)
Acrylamide demand
(million Ibs)
253
290
PAM demand
(million Ibs)
238
273
       Source: ICIS Chemical Business 2008
       * U.S. demand equals production plus imports less exports

Polymers such as PAMs are produced from petroleum, so additional PAMs consumption to treat
construction site stormwater runoff would result in increased petroleum consumption. However,
consumption on construction sites is not expected to significantly increase demand for
acrylamide. EPA estimates that total treatment volumes under Option 4 are 273 million gallons
per year. Assuming a PAMs dosage of 2 mg/L to all stormwater generated, incremental PAM use
under Option 4 would be 4,560,000 pounds per year.

   11.1.3. COMPARISON OF OPTION ENERGY REQUIREMENTS TO
           CONSTRUCTION INDUSTRY

Table 11-4 presents an estimate for construction industry fuel consumption based on the 2002
census.

                     Table 11-4. 2002 Energy use in NAICS Category 23
Census category
Gasoline and diesel fuel
NAICS category 23
$10,953,670,000a
2002 unit cost
$1.32/gallonb
NAICS category 23 energy use
(millions of gallons)
8,300
' U.S. Census Bureau 2002.
1 Energy Information Administration 2002.
Table 11-5 presents estimates of energy usage by regulatory option considered, compared to the
total annual diesel and gasoline consumption in NAICS Category 23 (Construction).
             Table 11-5. Estimated incremental energy usage by regulatory option
Option
Option 1
Option 2
Option 3
Option 4
Option diesel consumption
(gallons)
33,532
59,965,244
90,991,961
76,397
Fraction of NAICS category 23
energy (gallons)
0.000004
0.007
0.011
0.000009
                                         11-4

-------
                                               Section 11: Non-Water Quality Environmental Impacts
EPA does not expect any adverse effects to occur as a result of the small incremental energy
requirements for the regulation.

 11.2.   AIR EMISSIONS IMPACTS

The Agency believes that none of the regulatory options for this rule would generate significant
air emissions.

According to the Construction Industry Compliance Assistance Compliance Summary Tool
(http://www.cicacenter.org/cs.cfm) no federal Clean Air Act (CAA) requirements apply to the
C&D industry. CAA requirements are implemented primarily by states through their State
Implementation Plans (SIPs). Following are examples of construction-related emissions that
might require a state permit under an SIP:

       •  Nitrogen oxides (NOx) and fine particulates from construction equipment diesel
          engines

       •  Dust from vehicle traffic, from loading and unloading of construction materials at
          transfer points, and from conveyor systems transporting building materials

       •  Visible stack emissions from off-road equipment

       •  Volatile organic compounds (VOCs) from paint and cleaning solvents

To the extent that use of heavy construction equipments would be expected to increase from
removing accumulated sediment, or portable generators or diesel powered pumps are used to
power ATS, there would be an increase in fine paniculate matter, VOCs, and NOx, and other
pollutants, as well as increased CO2 emissions, as estimated below.

EPA estimates air emissions on the basis of emission factors from diesel generators, the primary
source of construction site air emissions, and excavators and trucks to remove accumulated
sediment. A 135-kilowatt generator  (210 horsepower [hp]) generator would consume
approximately 10 gallons of diesel per hour (Diesel Supply and Service, No date). EPA
multiplied the total system run times presented in Section 11.1 by the emission factors from the
California South Coast Air Quality Management District (SC AQMD 2008)). Table 11-6
presents the estimated incremental air emissions by regulatory option.
       Table 11-6. Estimated incremental air emissions by regulatory option (pounds/year)
Option
Option 1
Option 2
Option 3
Option 4
Reactive
organic
gases
(ROG)
2,066
1,116,150
1,692,847
4,707
Carbon
monoxide
(CO)
6,731
3,997,266
6,062,967
15,335
Nitrogen
oxides
(NOx)
19,299
11,595,274
17,587,588
43,970
Su If uric
oxides
(SOx)
20
1 1 ,860
17,989
45
Particulate
matter (PM)
794
442,759
671,540
1,809
Carbon
dioxide (CO,)
1,829,303
1,052,768,276
1,596,784,327
4,167,800
Methane
(CH4)
186
100,692
152,718
424
                                          11-5

-------
                                              Section 11: Non-Water Quality Environmental Impacts
Because construction air emissions are primarily from fuel combustion, EPA estimates that the
increase in air emissions relative to the construction industry air emissions would be similar to
the estimates for the fraction of construction industry fuel consumption presented in Table 11-5.

 11.3.   SOLID WASTE GENERATION

Solid waste generated at C&D sites include treatment residuals generated as part of coagulation
and flocculation from ATS, and sediment that accumulates in channels, basins, and traps that are
used as part of passive treatment systems. If ATS are used, solid waste can include spent
cartridge or bag filters, or filter media (usually sand). EPA did not quantify solid waste generated
from spent cartridge or bag filters because it is not clear whether permittees would require
cartridge or bag filters as a final finishing step after ATS. Sediment removed from sediment
basins and ATS, including sediment-containing polymers, can generally be used as fill material
on the construction site. Therefore, EPA expects that solid waste generation would be minimal
under any of the options.

 11.4.   REFERENCES

Diesel Service and Supply, Inc. No date. Approximate Fuel Consumption Chart.
   . Accessed November 14, 2008. (DCN 43090)

Energy Information Administration. 2002. Average 2002 dieselfuel cost. U.S. On-Highway
   Diesel Fuel Prices. Energy Information Administration, Washington, DC.
   . Release Date April 21, 2008. (DCN
   43088)

Global Industry Analysts, Inc. 2008. PRWeb Press Release Newswire. Global Chitin Market to
   Exceed 51.4 Thousand Metric Tons by 2012. (DCN 43084)

Hennen, WJ.  1996. Chitosan. Woodland Publishing, Pleasant Grove, UT. (DCN 43085)

ICIS Chemical Business. 2008. Chemical profile: acrylamide.
   . Accessed November 6,
   2008. (DCN 43086)

Minton, G.R. 2006. Technical Engineering Evaluation Report (TEER)for the Chitosan-
   Enhanced Sand Filtration Technology for Flow-through Operations. Technology developed
   by Natural Site Solutions, LLC, Redmond, WA, study prepared by Resource Planning
   Associates, Seattle, WA. (DCN 43002)

SC AQMD (South Coast Air Quality Management District). 2008. Off-road Mobile Source
   Emission Factors. California South Coast Air Quality Management District, Diamond Bar,
   CA.  Accessed May 22, 2008.
   (DCN 43089)

Rain for Rent. 2008. Chitosan-Enhanced Sand Filtration Example Quote #1 and #2 (March
   2008). Rain for Rent, Bakersfield, CA. (DCN 43007)
                                         11-6

-------
                                               Section 11: Non-Water Quality Environmental Impacts
U.S. Census Bureau. 2002. Economic Census Construction Industry General Summary 2002.
   Issued October 2005. U.S. Census Bureau, Washington, DC. (DCN 43079)

USEPA (U.S. Environmental Protection Agency). 2003. Chitosan: Poly-D-glucosamine
   (128930) Fact Sheet.
   .
   Accessed November 14, 2008. (DCN 43087)
                                          11-7

-------