SEPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
EPA-600/6-82-004a
September 1982
Research and Development
Water Quality
Assessment:
A Screening
Procedure for Toxic
and Conventional
Pollutants—Part 1
-------
EPA-60Q/6-82-004a
September 1982
WATER QUALITY ASSESSMENT:
A Screening Procedure for Toxic
and Conventional Pollutants
Part 1
by
W.B. Mills, J.D. Dean, D.B. Porcella, S.A. Gherini, R.J.M. Hudson,
W.E. Frick, G.L. Rupp, and G.L. Bowie
Tetra Tech, Incorporated
Lafayette, California 94549
Contract No. 68-03-2673
Prepared in Cooperation with U.S. EPA's
Center for Water Quality Modeling
Environmental Research Laboratory
Athens, Georgia
Monitoring and Data Support Division
Office of Water Regulations and Standards
Office of Water
Washington, D.C.
Technology Transfer
Center for Environmental Research Information
Cincinnati, Ohio
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ATHENS, GEORGIA 30613
-------
DISCLAIMER
Mention of trade names or commercial products does not constitute
endorsement or recommendation for use by the U.S. Environmental Protection
Agency.
-------
FOREWORD
As environmental controls become more costly to Implement and the
penalties of judgment errors become more severe, environmental quality
management requires more efficient analytical tools based on greater know-
ledge of the environmental phenomena to be managed. As part of this Lab-
oratory's research on the occurrence, movement, transformation, impact,
and control of environmental contaminants, the Technology Development and
Application Branch develops management or engineering tools to help pollu-
tion control officials achieve water quality goals through watershed manage-
ment.
Basin planning requires a set of analysis procedures that can provide
an assessment of the current state of the environment as a means of predict-
ing the effectiveness of alternate pollution control strategies. This manual
contains a description of a set of consistent analysis procedures that can
help to accomplish these tasks. It is directed toward local and state gov-
ernment planners who must interpret technical information from many sources
and recommend the most prudent course of action that will maximize the en-
vironmental benefits to the community and minimize the cost of implementation.
The manual was prepared in cooperation with the Office of Water and the
Center for Environmental Research Information. The Office of Water is re-
viewing the manual for its potential use as a screening guide for the Waste-
load Allocation-Total Maximum Daily Load program. User comments on the
methodologies in the manual are encouraged and should be directed to the
Center for Water Quality Modeling at the Athens Environmental Research Lab-
oratory.
David W. Duttweiler
Director
Environmental Research Laboratory
Athens, Georgia
-------
ABSTRACT
New technical developments in the field of water quality assessment
and a reordering of water quality priorities prompted a revision of Water
QuaIJty Assessment: A Screening Methodology for Nondesignated 208 Areas
(EPA-600/9-77-023)! The utility of the revised manual is enhanced by the
inclusion of information on the accumulation, transport, and fate of toxic
chemicals in the environment. The new subtitle—A Screening Procedure for
Toxic and Conventional Pollutants — reflects the added information.
Applying the manual's simple techniques, the user is now capable of
assessing the loading and fate of conventional pollutants (temperature,
biochemical oxygen demand-dissolved oxygen, nutrients, and sediments) and
toxic pollutants (from the U.S. EPA list of priority pollutants) in streams,
impoundments, and estuaries. The techniques are readily programmed on hand-
held calculators. Most of the data required for using these procedures are
contained in the manual.
Because of its size, the manual has been divided into three parts. Part
1 contains the introduction and chapters on the aquatic fate of toxic organic
substances, waste load calculations, and the assessment of water quality para-
meters in rivers and streams. Part 2 continues with chapters on the assessment
of impoundments and estuaries and appendices A, B, C, E, F, G and H. Appendix
D is provided in the third part (on microfiche in the EPA-printed manual).
This report is submitted in fulfillment of Contract No. 68-03-2673 by
Tetra Tech, Inc., under the sponsorship of the U.S. Environmental Protection
Agency. Work was completed as of February 1982.
-------
TABLE OF CONTENTS
PART 1
FOREWORD i i I
ABSTRACT -j v
LIST OF FIGURES (PART 1) vi ^
LIST OF TABLES (PART 1) xi1
ACKNOWLEDGMENTS xix
CHAPTER
1 INTRODUCTION I
1.1 Background 1
1.2 Purpose and Scope 2
1.3 Methodology Application 4
1.4 Limitations 6
References for Chapter 1 7
2 AQUATIC FATE OF TOXIC ORGANIC SUBSTANCES 8
2.1 Introduction 8
2.2 Screening Methods for Toxic Organic Compounds 27
2.3 Speciation Processes 54
2.4 Transport Processes 76
2.5 Transformation Processes 101
References for Chapter 2
3 WASTE LOAD CALCULATIONS 168
3.1 Introduction 168
3.2 Nonurban Nonpoint Source Loads 170
3.3 Urban Nonpoint Source Loads 253
3.4 Point Source Waste Loads 297
References for Chapter 3 315
-------
Page
4 RIVERS AND STREAMS 321
4.1 Introduction 321
4.2 Carbonaceous and Nitrogenous Oxygen Demand 362
4.3 Dissolved Oxygen 380
4.4 Temperature 423
4.5 Nutrients and Eutrophication Potential 469
4.6 Total Coliform Bacteria 480
4.7 Conservative Constituents 486
4.8 Sedimentation 488
4.9 Toxic Substances 509
References for Chapter 4
PART 2
5 IMPOUNDMENTS 1
5.1 Introduction 1
5.2 Impoundment Stratification 3
5.3 Sediment Accumulation 24
5.4 Eutrophication and Control 65
5.5 Impoundment Dissolved Oxygen 92
5.6 Toxic Chemical Substances 128
5.7 Application of Methods and Example Problem 140
References for Chapter 5 185
Glossary of Terms 187
6 ESTUARIES 191
6.1 Introduction 191
6.2 Estuarine Classification 207
6.3 Flushing Time Calculations 222
6.4 Far Field Approach to Pollutant Distribution in Estuaries 251
6.5 Pollutant Distribution Following Discharge from a Marine
Outfall 314
6.6 Thermal Pollution 367
6.7 Turbidity 390
6.8 Sedimentation 408
References for Chapter 6
APPENDICES
A Monthly Distribution of Rainfall Erosivity Factor R A-l
B Methods for Predicting Soil Erodibility Index K B-l
C Stream and River Data C-l
D Impoundment Thermal Profiles D-l
E Modeling Thermal Stratification in Impoundments E-l
F Reservoir Sediment Deposition Surveys F-l
G Initial Dilution Tables G-l
H Equivalents of Commonly Used Units of Measurement H-l
vi
-------
LIST OF FIGURES
PART 1
Figure Page
II-l Environmental Fate of a Toxic Pollutant 9
II-2 Speciation, Transport and Transformation Processes in the 28
Aquatic Environment
11-3 Flow System Representations 32
II-4 Isotherms for Adsorption of a Hydrophobic Pollutant on Sediments 64
II-5 Relationships Between Koc and Octanol-Water Partition Coeffi- 69
cient (Kow) of Energy-Related Organic Pollutants
II-6 Correlation of Aqueous Solubility with Octanol-Water Partition 70
Coefficient
11-7 Relationship Between KQC and Kowfor Coarse Silt 71
II-8 Schematic Representation of Volatilization form Solution Phase 84
to Liquid Phase
II-9 Microbial Transformations of Phenoxy Herbicides 104
11-10 Ultraviolet Absorption Spectrum of Naphthacene 128
11-11 Spectral Distribution of Solar Energy 129
11-12 Solar Radiation in the United States 132
11-13 Photochemical Pathways of an Excited Molecule 138
11-14 Direct Photochemical Reactions of a 2,4-D Ester 140
11-15 Comparison of Solar Irradiance with the Absorption Spectra 143
11-16 pH Dependence of Hydrolysis Rate Constants 157
III-l Flow Diagram for Calculating Sediment Loading from Surface 172
Erosion
III-2 Average Annual Values of the Rainfall-Erosivity Factor, R 174
vii
-------
n_cju_re Page
III-3 Mean Annual Values of Erosion Index for Hawaii 175
III-4 Soil Moisture-Soil Temperature Regimes of the Western U.S. 177
III-5 Relationships between Annual Average Rainfall Erosivity Index 179
and the 2--year, 6-hour Rainfall Depth for 3 Rainfall Types
in Western U.S.
III-6 Storm Distribution Regions in Western U.S. 180
III-7 Slope Effect Chart Applicable to Areas A-l in Washington, 184
Oregon, and Idaho and All of A-3
III-8 Slope Effect Chart for Areas Where Figure III-7 is Not Appli- 185
cable
111-9 Sediment Delivery Ratio for Relatively Homogeneous Basins 196
111-10 Percentage Nitrogen in Surface Foot of Soil 213
III-ll Soil Nitrogen vs. Humidity Factor and Temperature 215
111-12 Nomograph for Humidity Factor, H 216
111-13 Phosphorus Content in the Top 1 foot of Soil 218
111-14 Nitrogen (NH4-N and N03-N) in Precipitation 231
111-15 Climate Zone for the Cities from which Data Are Available 270
and Used in the URS Study
111-16 Correlation between Population Density and Curb Length Density 273
111-17 Street Surface Contaminant Removal as a Function of Runoff 276
111-18 Correlation of Influent Total Metals Concentration to Percent 310
Industrial Flow
IV-1 Illustration of River Segmentation Procedure on the James River 340
IV-2 Hypothetical River Having a Variety of Pollutant Sources and 342
Sinks
IV-3 River Segmentation for BOD Distribution 343
IV-4 Pollutant Discharge Where Initial Mixing Occurs a Fractional 346
Distance Across the River
IV-5 Illustration of Water Balance 353
vm
-------
IV-6 Sketch of Snake River from Heise to Neeley, Idaho 355
IV-7 Example of Flow Rate Information Tabulated in U.S. Geological 357
Survey's Water Data Report
IV-8 Example Set of User's Instructions for Hand-Held Calculator 361
Programs
IV-9 The BOD Curve, (a) Curve for Oxidation of Carbonaceous Matter. 364
(b) Curve Showing Influence of Nitrification
IV-10 Mechanisms of BOD Removal from Rivers 366
IV-11 Deoxygenation Coefficient as a Function of Depth 367
IV-12 Example of Computation of KL from Stream Data 369
IV-13 Hypothetical BOD Waste Loadings in a River 375
IV-14 Variability of Dissolved Oxygen by Season for 22 Major Water- 381
ways, 1968-72
IV-15 Reaeration Coefficient as a Function of Depth 385
IV-16 Reaeration Coefficient for Shallow Streams 386
IV-17 Reaeration Rate vs. Depth and Velocity 387
IV-18 Characteristic Dissolved Oxygen Profile Downstream from a Point 398
Source of Pollution
IV-19 Flow Process of Solution to Dissolved Oxygen Problem in Rivers 399
IV-20 Daily Dissolved Oxygen Variation in Two Rivers 408
IV-21 Flow Process in Reach by Reach Solution to Critical Dissolved 415
Oxygen Values
IV-22 Hypothetical River Used in Example IV-9 420
IV-23 Mechanisms of Heat Transfer Across a Water Surface 426
IV-24 Schematic of Site No. 3 Cooling Lake 427
IV-25 Observed Temperatures, Site No. 3, July 18-July 24, 1965 428
IV-26 Comparison of Computed Equilibrium and Ambient Temperatures 429
with Observed Mean Diurnal Temperature Variations for Site
No. 3, July 18-July 24, 1966
IX
-------
Figure Page
IV-27 Mean Daily Solar Radiation throughout the U.S. for July and 431
August
IV-28 Mean Dewpoint Temperature (Deg. F) throughout the U.S. for 441
July and August
IV-29 Mean Daily Wind Speeds (mph) throughtout the U.S. for July 442
and August
IV-30 Idealization of a Run-of-the-River Power Plant 447
IV-31 Downstream Temperature Profile for Completely Mixed Stream, 460
(T-E)/(Tm-E) vs. r
IV-32 Measured Air and Water Temperatures for the Santa Ana River 463
near Mentone, California, in June 1979
IV-33 Measured Dissolved Oxygen Concentration and Predicted Satura- 464
tion Concentration for the Santa Ana River near Mentone,
California, in June 1979
IV-34 Flow Duration Curve, Hatchie River at Bolivar, Tennessee 466
IV-35 Frequency of Lowest Mean Discharges of Indicated Duration, 467
Hatchie River at Bolivar, Tennessee
IV-36 Three River Temperature Profiles 468
IV-37 Total Coliform Profiles for the Willamette River 481
IV-38 Salinity Distribution in a Hypothetical River 488
IV-39 Division between Wash Load and Bed Material Load 491
IV-40 y and TC for DuBoys Relationship as Functions of Median Size 494
of Bed Sediment
IV-41 Hydraulic Radii for Different Channel Shapes 497
IV-42 User Instructions for Yang's Sediment Transport Equation 502
IV-43 Program Listing and Sample Input/Output for Yang's Sediment 503
Transport Equation
IV-44 Sediment Discharge as a Function of Water Discharge for the 507
Colorado River at Taylor's Ferry
IV-45 Sediment Discharge as a Function of Water Discharge for the 508
Niobrara River at Cody, Nebraska
-------
Figure
IV-46 Toxicant Concentrations Following Initiation and Cessation 513
of Point Source
IV-47 Vertical Equilibrium Distribution of Suspended Solids in a 516
River
IV-48 Vertical Distribution of Relative Solute Concentration, 517
KPSA = 10
IV-49 Vertical Distribution of Relative Solute Concentration, 518
KPSA = 100
IV-50 Instreatn Transformation Processes Analyzed for Toxicants 527
IV-51 Location Map of Hudson River, New York 537
IV-52 Hypothetical Concentration Distributions of Finitely Soluble 542
and Infinitely Soluble Toxicants
IV-53 Hypothetical Distribution of Toxicant at Various Locations 546
Following a Spill
IV-54 Illustration of Hypothetical Spill Incident 552
IV-55a Chloroform Concentration in Water Column for First 60 Hours 560
Following a Spill 16.3 Miles Upstream
IV-55b Chloroform Concentration in the Mississippi River at a Loca- 561
tion 16.3 Miles Below the August 29, 1973, Spill
-------
LIST OF TABLES
PART 1
Table Page
II-l Brief Comparison of Conventional and Toxic Pollutants 10
II-2 Proposed Criteria for Toxic Substances Designed to Protect 13
Aquatic Life
II-3 EPA List of 129 Priority Pollutants and the Relative Frequency 16
of These Materials in Industrial Wastewaters
II-4 Most Commonly Discharged Priority Pollutants 17
II-5 Selected Characteristics of Various Aliphatic Hydrocarbons 20
II-6 Various Characteristics of Selected Pesticides 21
II-7 Selected Characteristics of Polychlorinated Biphenyls and Rela- 22
ted Compounds
II-8 Selected Characteristics of Monocyclic Aromatic Hydrocarbons 23
II-9 Selected Characteristics of Various Polycyclic Aromatic Hydro- 24
carbons
11-10 Expressions for Toxic Pollutant Levels in Various Water Bodies 35
11-11 Relative Importance of Processes Influencing Aquatic Fate of 39
Organic Priority Pollutants
11-12 Occurrence of Acids and Bases in Neutral and Charged Forms as 57
as a Function of p^1, pKa > anc! PKb
11-13 pKa and pK^ Values for Selected Toxic Organic Acids and Bases 58
and Values of p ^ for Water
11-14 Relationship of Dissolved and Sorbed Phase Pollutant Cone 73
trations to Partition Coefficient and Sediment Concentration
11-15 Henry's Law Constant for Selected Hydrocarbons 81
11-16 Henry's Law Constants for Selected Compounds 89
xii
-------
Table Pac]§
11-17 Typical Values of Pollutant Volatilization Rates in Surface 89
Waters
11-18 Comparison of Tabulated and Predicted Values of Diffusion Coeffi- 91
cients for Selected Pollutants
11-19 Results of a Study to Directly Determine Volatilization Rates 94
of Several Priority Pollutants in Rivers
11-20 Relative Volatilization Mass Fluxes of Several Chemicals in 99
Saturated Solutions
11-21 Size of Typical Bacterial Populations in Natural Waters 109
11-22 Summary of the Characteristics of the Two General Types of Bio- 110
degradation: Metabolism and Cometabolism
11-23 Potential Biodegradability of Organic Pollutants in an Aerobic 113
Environment
11-24 Biodegradation Rate Constants under Aerobic Conditions 116
11-25 Calculated Solar Radiant Energy Flux to a Horizontal Surface 130
under a Clear Sky
11-26 Calculated Solar Irradiance in a Water Body Just Beneath the 133
Surface, Annual Mean at 40 N
11-27 Contributions to Light Attenuation Coefficient 136
11-28 Disappearance Quantum Yields, ^d, for Direct Photolysis 142
11-29 Near-Surface Direct Photolysis Rate Constants , 148
11-30 Generalized Hydrolytic Reactions of Organic Compounds 156
11-31 Hydrolysis Rate Parameters and Estimated Environmental Hydro- 160
lysis Rates
III-l Applicability of Rr and RS Factors in the Areas West of the 178
the Rocky Mountains
III-2 Generalized Values of the Cover and Management Factor, C, in 187
the 37 States East of the Rocky Mountains
III-3 "C" Values for Permanent Pasture, Rangeland, and Idle Land 189
III-4 "C" Values for Woodland 191
III-5 "C" Values for Construction Sites 192
xi i i
-------
Table Page
III-6 Practice Factors (P) Used in Sediment Loading Equation 194
III-7 Typical Values of Drainage Density 198
III-8 Erosion Equation Factor Precision Error 201
III-9 Runoff Curve Numbers for Hydrologic Soil-Cover Complexes 206
(For Antecedent Rainfall Condition II)
111-10 Antecedent Rainfall Conditions and Curve Numbers (for Ia = 0.25)208
111-12 Enrichment Ratios for Nitrogen 212
111-13 Enrichment Ratios for Phosphorus 221
111-14 Enrichment Ratios for Organic Matter in Surface Runoff 224
111-15 Calculated Sediment, Nitrogen, Phosphorus and Organic Matter 226
Loads for Parke County, Indiana, Watershed
II1-16 Variation in Constituent Accounted for by Regression on Suspend- 229
ed Solids (Linear Models Only)
111-17 Atmospheric Contributions of Nitrogen and Phosphorus in Rain- 232
fall
111-18 Nutrient Budgets for Various Terrestrial Ecosystems of the 233
World
111-19 Salt Yields from Irrigation in Green River Subbasin 234
111-20 Salt Yields from Irrigation in Upper Colorado Main Subbasin 235
111-21 Salt Yields from Irrigation in San Juan River Subbasin 236
111-22 Salt Yields from Irrigation in Lower Colorado River Basin 237
111-23 Salt Yields from Irrigation for Selected Areas in California 237
111-24 Values of k for Dissipation of Pesticides from Soil Surfaces 243
111-25 Degradation Rate Coefficients for Selected Pesticides 247
111-26 Octanol-Water Partition Coefficients for Selected Pesticides 249
111-27 General Land Consumption Rates fir Various Land Uses 256
111-28 Pollutant Loading Factors 256
111-29 Comparison of Quality of Storm Sewer Discharges for Cities 258
xiv
-------
Tab!e Pacj^
111-30 Comparison of Quality of Combined Sewers for Various Cities 259
111-31 Summary of Stormwater Pollutant Concentrations 260
111-32 Summary of Street Cleaning Methods 263
111-33 Removal Rates for Selected Contaminants by Size 264
111-34 Solid Loading Rates and Composition—Nationwide Means and 266
Substitutions of the Nationwide Means at 80% Confidence Level
111-35 Values of Runoff Coefficient, k 274
111-36 Relationships between Total Suspended Solids and Other Pollutants 277
111-37 Field-Measured Dry Deposition Velocities 282
111-38 Washout Ratios for Selected Trace Organics 288
111-39 PCBs, DDTs, and Phthalate Esters in the Gulf of Mexico Atmos- 289
phere
111-40 1975 Monthly Average Concentrations of Three Organic Compounds 290
at Three New York City Locations
111-41 Seasonal Fluctuations in the Geometric Mean PAH Concentrations 291
in Air Samples Collected at 13 Stations in the Los Angeles,
California, Area
111-42 Average Monthly Atmospheric Levels of Four Pesticides at Stone- 292
ville, Mississippi
111-43 Flow Weighted Mean Concentrations of Trace Metals and Chlorina- 294
ted Hydrocarbons in the Los Angeles River
111-44 Concentrations of PAH in Municipal Wastewater Effluents in the 295
GFR
111-45 Water Withdrawals for Public Supplies by States and by Selected 300
Municipal Systems, 1970
111-46 Typical Municipal Waste Concentrations 301
111-47 Municipal Wastewater Treatment System Performance 302
111-48 Point Source Loadings of Six Major Wastewater Treatment Facili- 303
ties in One North Carolina 208 Area
111-49 Typical Industrial Discharge Pollutant Concentrations 305
xv
-------
Table Page
II1-50 Summary of Current and Projected Waste Loads in One Region 306
208 Area (by SIC Code)
111-51 Predicted Priority Pollutants in Household Wastewater 308
111-52 Occurrence of Priority Pollutants in POTW Influent Samples 309
111-53 Industrial Categories and Frequently Detected Priority Pollu- 311
tants by Category
111-54 Reduction of Conventional and Priority Pollutants by POTW 313
Treatment Processes
111-55 Concentrations (Mean + Standard Error) of EPA Priority Pollu- 314
tants in the Los Angeles County JWPCP Effluents
IV-1 Reference Level Values of Selected Water Quality Indicators for 323
U.S. Waterways
IV-2 Condition of Eight Major Waterways 324
IV-3 Water Quality Problem Areas Reported by States 327
IV-4 Example River Water Quality Standards 328
IV-5 Water Quality Parameters Commonly Monitored by States 329
IV-6 Annual Phosphorus and Nitrogen Load for Selected Iowa River 332
Basins
IV-7 Major Waterways: Seasonal Flow Analysis, 1968-72 336
IV-8 Water Quality Analyses for River Screening Methodology 337
IV-9 Experimental Measurements of Transverse Mixing in Open Channels 348
with Curves and Irregular Sides
IV-10 Suggested Configuration for Water and Nutrient Balance Table 351
IV-11 Solution to Snake River Water and Phosphorus Balance Problem 358
IV-12 Municipal Waste Characteristics before Treatment 363
IV-13 Comparison of Predicted and Observed Reaeration Rates on Small 389
Streams in Wisconsin
IV-14 Typical Hydraulic Properties in Patuxent River (September 1969) 390
IV-15 Solubility of Oxygen in Water 394
xvi
-------
Table p_a_g_e
IV-16 Dissolved Oxygen Saturation vs. Temperature and Altitude 396
IV-17 DC/L0 Values vs. D0/L0 and ka/KL 402
IV-18 kat vs. D0/L0 and ka/kL 403
IV-19 Some Average Values of Gross Photosynthetic Production of Dissol- 407
ved Oxygen
IV-20 Average Values of Oxygen Uptake Rates of River Bottoms 412
IV-21 Compilation of Information in Example IV-8 417
IV-22 Critical Time Results 419
IV-23 Net Long Wa e Atmospheric Radiation, Han 433
IV-24 Water Vapor Pressure vs. Air Temperature, Ta, and Relative 434
Humidity
IV-25 B and C(B) as Functions of Temperature 435
IV-26 Summary of Solar-Radiation Data for Mineola, Brookhaven, and the 437
Connetquot River Sites
IV-27 Data Needed for Thermal Discharge Screening 449
IV-28 Eutrophication Potential as a Function of Nutrient Concentrations 472
IV-29 Regional Stream Nutrient Concentration Predictive Models 476
IV-30 Total Nitrogen Distribution in a River in Response to Point and 479
and Non-Point Source Loading
IV-31 Total Coliform Analysis 482
IV-32 Salinity Distribution in a Hypothetical River 489
IV-33 Relationship of Total Suspended Sediment Concentration to Problem 492
Potential
IV-34 Sediment Grade Scale 495
IV-35 Computing D/T for Determining the Hydraulic Radius of a Parabolic 496
Section
IV-36 Relationship between Width to Depth Ratio of a Graded Stream and 498
the Suspended and Bed Load Discharge
xvi
-------
Table Pag_e_
IV-37 Characteristics of the Colorado and Niobrara Rivers 505
IV-38 Methods of Introduction of Toxic Organic Compounds into Rivers 510
and Fate in Terms of Volatilization and Sorption
IV-39 Mass of Contaminated Sediments and Equivalent Water Depth as a 536
Function of Depth of Contamination
IV-40 Water-Soluble, High Density, Immiscible Chemicals 549
xvm
-------
ACKNOWLEDGEMENTS
This publication is the result of the labors of a number of individuals
who contributed to both this document and the previous edition. Two of the
authors of the previous document, Stanley Zison and Kendall Haven, were
instrumental in producing this work because many of their original ideas
have been retained. In addition, all of the individuals in the U.S. EPA,
especially Dr. James Falco, Mr. Orville Macomber, Mr. Robert Ambrose, and
Mr. Tom Barnwell, who supported this work must be thanked for their input,
consideration, and patience.
Because of the size of this document a phenomenal amount of typing and
graphic art work was done. The authors would like to acknowledge
Susan I. Madson, Pencie Shrewsbury, Barbara Koch and Trudy Rokas for their
tireless efforts in these areas. Finally, the authors would like to thank
Ms. Carrie Campbell who tabulated much of the data on toxic substances found
in Chapter 2.
xix
-------
-------
CHAPTER 1
INTRODUCTION
1.1 BACKGROUND
In 1977, the United States Environmental Protection Agency published
Hater Quality Assessment: A Screening Method for Nondeslgnated 208 Areas
(Zison et a]_., 1977). This document was intended as a simplified
methodology that water quality planners in nondesignated 208 areas could
use to perform preliminary assessments of surface water quality. The
method covered primarily the identification of problem areas for sediment,
nutrients, dissolved oxygen, and some urban pollutants in streams, lakes
and estuaries.
The original methodology was used by its developers, Tetra Tech, Inc.,
as instructive materials in EPA workshops on water quality assessment. It was
also used in an EPA project designed to test the methodology. In that project,
elements of the procedures were applied to the Sandusky River in northern
Ohio and to the Ware, Patuxent, Occoquan, and Chester Rivers in Virginia
and Maryland. Testing results were favorable for phosphorus and reasonable
for nitrogen (nitrate loading was a problem) and were reported in two
publications (Dean et al_., in press; Dean ejt al_., 1981).
As feedback was acquired from individual users and workship partici-
pants, there arose a need to reassess the methods available in light of
new technical developments and new priorities in water pollution assess-
ment and control. To this end, the original screening method (Zison e_t al_.,
1977) has been revised and updated to include the assessment of toxic
chemicals in the environment. The title has been changed to reflect the
new content of the screening procedure.
-------
1.2 PURPOSE AND SCOPE
This report contains a simplified methodology which can be used by
planners or engineers to perform preliminary assessment of toxic and
conventional pollutants in surface waters. Conventional pollutants include
suspended sediments, nitrogen, phosphorus, coliform bacteria, BOD and
dissolved oxygen deficits. The 129 EPA priority pollutants are included in
the sections on toxic chemicals. The analyses require little external
input, since much data are supplied by figures and tables in the text or
appendices. Additional data can be found in Zison et_ al_., 1978. All the
algorithms are intended to be used on a desk-top calculator.
Where instructive, introductory material has preceded the actual
presentation of water quality assessment methodologies. This was done to
orient the planner toward pertinent background material, as well as to
clearly state limitations of the methodologies due to assumptions and
simplifications. Further, example calculations of the major emphases within
each chapter are included to illustrate the ideas being presented. These
examples are designed to unify the theory that has preceded it, as well as
in some cases to introduce new but related ideas.
The units most commonly used in this report are those that historically
appear in the literature. Often, the units are not metric. Consequently an
english-metric-conversion appendix is included at the end of this report.
Many equations are presented with both English and metric constants.
The report is divided into five major chapters (two through six). A
brief description of the content of each chapter is presented in the
following paragraphs.
• Chapter 2 deals with the environmental chemistry of toxic
chemicals. Processes considered include photolysis,
hydrolysis, volatilization, biodegradation and adsorption.
The purpose of the chapter is to provide an understanding of
the processes and to provide procedures for estimating
associated rate and equilibrium constants.
-------
• Chapter 3 deals with the estimation of pollutant loads from
nonpoint and point sources for both toxic and conventional
pollutants. Procedures include load estimation for single
event and annual loads from agricultural, forested, and urban
areas.
0 In Chapter 4, impacts of point and nonpoint sources of
conventional and toxic pollutants in rivers are addressed.
Conventional pollutant interactions presented include BOD-DO,
temperature, coliform bacteria, nutrients, and sediment
transport. Fate of toxic chemicals is assessed using
volatilization, sorption and first order degradation. Methods
are also presented to handle large spills of toxic chemicals
having density the same as or different from the receiving
waters.
• Chapter 5 contains methods for assessing water quality in
impoundments. The topics covered are sediment accumulation,
thermal stratification, BOD-DO interactions, eutrophication,
and fate of toxic materials. The physical/chemical processes
governing the fate of toxicants as well as biological uptake
and bioconcentration are considered.
• In Chapter 6, methods are presented for estuary
classification, flushing time prediction, and transport of
conservative and non-conservative pollutants and dissolved
oxygen in well-mixed extuaries. For stratified estuaries,
Pritchard's box model is used to determine the distribution of
conservative materials. Additionally, methods are presented
to calculate initial dilution from a waste water discharge and
pollution distribution at the completion of and subsequent to
initial dilution.
-------
1.3 METHODOLOGY APPLICATION
For each category in the methodology, the six conceptual steps shown
below should be followed to screen a river basin:
t obtain necessary tools and data to make calculations and
utilize nomographs, etc.;
• identify problems that are obvious from inspection of the data
base;
• determine the state variables which will be screened;
t apply procedures and compare where possible to observed data;
t consider consequences of errors; and
• reevaluate and make recommendations.
The techniques in the screening procedure are designed to interact which
makes them ideal for use as an analytical tool for river basin surface
waters which may include rivers, lakes, and/or estuaries. Although the
procedures may interact, they can be applied individually and with
identified data sets for specific case studies.
1.3.1 Base Maps
The first step in the screening process can be to obtain large scale
topographic maps of the study area. These can be used to determine which
water bodies are to be examined and to establish an order of study. Once
this has been done, selected small scale (7 1/2-minute or 15-minute series)
topographic sheets can be obtained. On these, the planner can locate and
mark point source discharges, regions of specific kinds of land use,
population centers, and industrial complexes. Use of overlays or push pins
may be helpful in preparing these displays.
-------
-------
The maps are also very important in showing the relationships among
water bodies and the flow patterns for stormwater runoff. Mnally, control
strategies may be displayed for examination on the maps.
1.3.2 Data Collection
Once the base maps are prepared, the kinds of data needed should be
fairly clear in most cases. In general terms, the only data that will not
be provided in this methodology are climatic and hydrologic data.
Hydrologic data includes such items as:
0 Runoff quantity
• Stream flows (low flows, statistical flows such as 7Q10,
critical flows to be protected as decreed by law)
« Inflows and outflows, stagnant regions, stratification,
internal flow patterns
e Estuarine tidal prism
Much of the necessary hydrologic data will be available from the USGS,
state geological surveys, state environmental protection agencies, and other
governmental organizations. In addition, data may be available from the
private sector, from universities, local citizens groups, and private firms.
Hydrologic data must usually be analyzed to serve as a basis for
subsequent water quality analyses. Statistical methods may be applied to
determine the annual runoff, monthly runoff, and critical flow for a
stipulated return frequency, on a selected time basis.
To select critical flow, for example, one must have some base knowledge
of the seasonal distribution of stream flow and quantity-quality
relationships. In general, the summer low flow is considered as the
critical condition for stream and estuarine analyses. Average annual runoff
-------
is to be used for lake analyses, even though wet years are generally more
critical from the standpoint of lake water quality.
Climatic data may also be needed. Generally these are available from
the National Climatic Center in Ashville, North Carolina. This agency can
provide data summaries of various kinds for a large number of weather
stations. Data include precipitation, cloud cover, humidity, and other
important parameters. Computer tapes can often be provided.
In collecting data for the area to be screened, the Reach File data
base (EPA, In Press) may also be useful. This data base contains
information for over 68,000 river reaches in the 48 contiguous states. The
Reach File provides a unique index for each of these river reaches and a
systematic way of retrieving the hydrologic or water quality information
which is available.
1.4 LIMITATIONS
The processes which govern the fate of pollutants in the environment
are complex. A methodology such as this, designed for hand calculation,
cannot be inclusive of all of these processes nor in all cases are the
methods state-of-the-art. An attempt has been made in each chapter to cover
the assumptions under which the algorithms are developed. Users should be
aware of the assumptions, potential errors, and limitations of the tools
presented. When deficiencies are noted or the methods deemed inappropriate,
the user should be prepared to use a higher level analytical tool.
-------
REFERENCES
Dean, J.D., W.B. Mills and D.B. Porcella. 1981. A Screening Methodology
for Basin Wide Water Quality Management. Symposium on Unified River-
Basin Management. R.M. North, L.B. Dworsky and D.J. Allee (editors)
May 4-7, 1980, Gatlinburg, Tennessee.
Dean, J.D., B. Hudson and W.B. Mills. (In Press) River Basin Validation of
the MRI Nonpoint Calculator and Tetra Tech's Nondesignated 208
Screening Methodologies, Volume II. Chesapeake-Sandusky Nondesiqnated
208 Screening Methodology Demonstration. U.S. Environmental Protection
Agency, Athens, Georgia.
U.S. Environmental Protection Agency (In Press). An Introduction to the
Reach File and Reach File Directory. MonHoring Branch (WH-553).
U.S. Environmental Protection Agency, Washington, D.C. 20460.
Zison, S.W., K. Haven, and W.B. Mills, 1977. Water Quality Assessment: A
Screening Methodology for Nondesignated 208 Areas. F.PA-600/9-77-023,
U.S. Environmental Protection Agency, Athens, Georgia.
Zison, S.W., W.B. Mills, D. Deimer, and C. Chen, 1978. Rates, Constants,
and Kinetics Formulations in Surface Water Quality Modeling.
EPA-600/3-78-105, U.S. Environmental Protection Agency, Athens,
Georgia.
-------
CHAPTER 2
AQUATIC FATE OF TOXIC ORGANIC SUBSTANCES
2.1 INTRODUCTION
2.1.1 Background
Today's technological society generates enormous quantities of
chemicals both as products for consumption and as waste. As the volume and
number of chemicals has increased, numerous unintended adverse effects of
these chemicals have been observed in the environment. Because of the
potential hazard that exposure to these compounds poses to biota, the levels
of toxic and carcinogenic substances in the environment have become
important criteria for evaluating environmental quality.
The level, or concentration, of a toxic compound in the environment
depends on the quantity added to the environment and the processes which
influence its fate. "Transport" processes tend to distribute chemicals
between the atmospheric, aquatic, and soil environments depending on the
affinity of the compound for each phase. "Transformation" processes within
each phase chemically alter pollutants to forms of lesser, equivalent, or
sometimes greater toxicity. These processes occur at rates which are
specific to each chemical and to each environmental compartment. The sum of
these processes and their interactions, as Figure II-l illustrates,
determines the environmental fate and consequent exposure of biota to a
toxic pollutant. The fate of toxic substances in the aquatic environment is
the concern of this chapter.
2.1.2 Comparison of Conventional arLd_Toxic__Pql_]_L[tants
Toxic substances frequently exhibit properties which are quite
different from the properties of conventional aquatic pollutants. It is
worthwhile to compare these differences in order to better appreciate the
-------
SOURCE
TRANSPORT AND
TRANSf ORMATION
CHEMICAL
EXPOSURE
FIGURE II-l ENVIRONMENTAL FATE OF A Toxic
POLLUTANT (AFTER HAGUE, 1980)
problems of analyzing impacts of toxicants in surface water systems. Table
II-l shows some of the more important differences.
Typically, one to two dozen pollutants and water quality parameters are
classified as "conventional". Until the past several years, these
parameters (e.g. BOD, nutrients) have received most of the attention of
water quality planners. In contrast to the small number of conventional
pollutants there are thousands of toxicants and many more synthetic
chemicals are continually being developed. Potentially, any of these
toxicants could enter the environment.
-------
TABLE II-l
BRIEF COMPARISON OF CONVENTIONAL AND TOXIC POLLUTANTS
Conventional
Toxic
One to two dozen pollutants fall into
this category
Often large quantities required to
produce impact (e.g. thousands
Ibs/day)
Concentrations often expressed as
ppm (mg/1)
Often travel in dissolved form
Mean residence time within water
bodies often equal to or less
than the mean residence time of
moving waters
Many biodegrade into harmless
substances
Thousands fall into this category;
many more being synthesized
Small quantities can produce
impact (e.g. few Ibs/days)
Concentrations often expressed as
ppb (yg/1), or in smaller units
May be highly sorbed to suspended
and bedded sediments
Can reside in bedded sediments
for years
Many are transformed to chemicals
which are also toxic; others are
resistant to degradation and
bioconcentrate
10
-------
Even though there are relatively few types of conventional pollutants,
numerous sources combine to routinely discharge large quantities. However,
because many surface water bodies have a capacity to assimilate conventional
pollutants (e.g. BOD) without apparent adverse effects, this practice is,
within limits, both acceptable and pragmatic. Toxic substances, on the
other hand, can cause adverse effects even at low discharge rates.
Concentrations of conventional pollutants are most often expressed in
units of ppm (or mg/1). Because of the small quantities of toxicants which
are typically released, concentrations are often expressed in the ppb (or
yg/1) range, or in even smaller units. This represents at least a thousand
fold difference relative to concentrations of conventional pollutants.
However, because toxic substances present in small amounts can adversely
impact the environment, these small concentrations can not always be
ignored.
Many conventional pollutants are transported in dissolved form. The
mean residence times of dissolved, conservative pollutants in a system is
equivalent to the mean residence time of water in the system, which is:
• the hydraulic detention time for freshwater lakes
• the travel time for freshwater rivers; and
• the flushing time for estuaries.
Many toxic chemicals strongly sorb to suspended and bedded sediments and
consequently can become a part of the immobile sediments in the bed. The
residence time of such chemicals can be on the order of years. Therefore,
depending on the properties of the toxicant the period of impact can greatly
exceed the period of discharge (e.g. a PCB spill may occur in a few minutes,
but quantities of the PCB may remain in immobile, bedded, sediments for
years). Consequently the recovery period of a system can be years.
11
-------
2.1.3 Water Quality Criteria
As previously indicated, toxicants are present in the environment in
quantities which are often measured in the ppb range. Such small
concentrations are often foreign to many workers in the field. When data or
model predictions contain concentrations in the ppb range, the significance
of the toxicant level is not always obvious (i.e. there is no "feel" as to
whether the concentration is large or small). Proposed criteria for toxic
substances can serve as a basis to gauge the significance of observed or
predicted levels. Table II-2 shows proposed criteria for numerous
toxicants. Since proposed criteria evolve over time the criteria shown in
the table are not necessarily the most current. Nevertheless, their
function remains: to provide a comparison with levels observed or predicted
in real systems. The data in these tables come from the "Red Book"
(U.S. EPA, 1976) and the Federal Register, March 15, 1979; July 25, 1979;
October 1, 1979; and November 28, 1980. Criteria, designed to protect
human health, for levels of toxicants in domestic water supplies, are
available from these same sources as well.
2.1.4 F reguencyJ3f.J31scharge of Toxic_S_ubstances__frqrn_I_ndustr|es
Numerous organizations, including the U.S. Department, of
Transportation and the U.S. Environmental Protection Agency, continually
collect and analyze data on the discharge of toxic substances. Table II-3
summarizes the results of a study reported by Keith and Telliard (1979)
which shows the frequency of detection of the 129 priority pollutants in
industrial wastewaters. A total of 32 industrial categories were analyzed
for organics and 28 for metals. The number of samples ranged from 2532 to
2988. Table 11-4 summarizes the most commonly discharged priority
pollutants. Table 111-53, shown in the next chapter, provides a breakdown
by industry of the occurrence of priority pollutants in industrial effluent.
It is common in this country for numerous industrial plants to release
their effluent into a single water body. Because of this situation a
question that naturally arises is: Based on the number and type of
industries located on the water body, what kinds of toxic chemicals are
likely to be discharged there ? If the industrial categories of each plant
12
-------
TABLE 11-2
PROPOSED CRITERIA FOR TOXIC SUBSTANCES DESIGNATE)
TO PROTECT AQUATIC LIFE
Sa i i
COMPOUND
Acenaphthene
Acrolein
Acrylonl trlle
Aldrin/Dieldrin
Antimony
Arsenic
Asbestos
Benzene
Benzidine
Beryllium
Cadmium
Carbon Tetrachlorlde
Chlordane <•
Chlorinated benzenes
Chlorobenzene
1,2,4 - Trichlorobenzene
1,2,3,5 - Tetrachlorobenzene
1,2,4,5 - Tetrachlorobenzene
Pentachlorobenzene
Chlorinated Ethanes
1,2 - Dichloroethane
1,1,1 - Trichloroethane
1,1,2 - Trichloroethane
1.1,1.2 - Tetrachloroethane
1,1,2,2 - Tetrachloroethane
Pen tac hi oroe thane
Hexachloroethane
Chlorinated Naphthalenes
Chlorinated Phenols
4 - Chlorophenol
2,4,6 - Trichlorophenol
CMoroalkyl Ethers
Chloroform
2 - Chlorophenol
Cftroriun (Hcxavalent)
Ccpper
C,;r,,de
DDT
24 Hour
Average
M9/1
LDa
21C
2600°
0.0019
1600
40C
LD
LD
LD
5.3°
d
620
0.0043
150Qh
210h
170h
97h
16h
3900h
h
5300h
310h
420h
170h
440h
62h
29
45
52
LD
500
60
0.29
5.6
3.5
0.00023
Maximum
ug/l
1700b
68b
7550b
2.5
9000
440b
LD
5300b
2500
130b
e
1400
2.4
3500h
470h
390h
220h
36h
8000h
i.
12000"
710h
960h
380h
ioooh
140h
67
180
150
LD
1200
180
21
1
52
0.41
.. .. 24 Hour
"Red Book" Average
710C
LO
LD
0.003 0.0019
LD
LD
LD
700°
LD
11. -1100 LD
0.4-l.Zf 4.5
4. 0-12. O9
2000
0.01 0.0040
120h
3.4h
2.6h
9.6
1.3h
880h
t.
240h
LD
LD
70h
38
7.0h
2.B
LD
LD
LD
620h
LD
100 18
j 4.0
5.0 LD
.001 O.C067h
Maximum "Red Book"
pg/1 yg/l
970b
55b
ID
0.71 0.003
LD
508b
LD
5100b
LD
LD
59 5.0
4000
D.09 0.004
280h
7.8h
5.9h
26
2.9h
2000h
h
54 Oh
LD
LD
160h
87
16h
6.4
LD
LO
LO
HQOh
LD
1?60
23 J
LD 5.0
0.021h .001
L i c f "!(jrc,L(_r,2eriei
1,2 - DicMor&t t r,7tne
1 • ^ - C-')cMt,rcI;r.7ene
44
310
59
13
-------
TABLE 11-2 (Continued)
COMPOUND
3,3' - Dichlorobenzidine
Dichloroethylenes
1,1 - Dichloroethylene
1,2 - Dichloroethylene
2,4 - Dichlorophenol
Olchloropropanes and
Dlchloropropenes
1,1 - Dichloropropane
1,2 - Dichloropropane
1,3 - Dichloropropane
1,3 - Dichloropropene
2,4 - Dimethyl phenol
Dim trotoluenes
2,3 - Dim trotoluene
2,4 - Dim trotoluene
1,2 - Diphenylhydrdzine
Cndosul fan
Endrui
Enthylbenzene
Fluordnthene
Haloetherb
4 - bi oniopheny 1 pht^nyl ether
Ha lunie thdnes
Chloromethane
Broinomethdrie
DiUiloromethdrie
Tr i bi onioiiietharie
Heptdchlor
Hexdchlorobutadiene
Hexd^h) orOLyclohexdne
L i nddtie
Other isomers
Huxav.nl or ucyclopen tad leiie
1 ^optior une
tedd
Mercury (tutdl)
Naphthalene
nickel
M trobenzene
24 Hour
Average
ug/1
LD
0.4
410
920
4800
18
38
12
620
17
0.042
0.0023
LD
250h
6.2
7000
140
4000h
840h
0.0038
LD
0.080
LD
0.39
2100
k
0.2
LD
n
480
Freshwater
Maximum "Red Book"
M9/1 pg/1
LD
11600
11600
110
930
2100
11000
250
86
27
1400
38
0.49 0.003
0 18 0.004
LD
560h
14
16000
320
9000h
1900h
0.52 0.001
LD
2.0
LD
7.0
4700
1 m
4 1 0.05
LD
0 p
1100
24 Hour
Average
M9/I
LD
224000
224000
LD
LD
400h
79
5.5h
LD
4.4h
LD
ID
LD
0.0023
LD
0.30
LO
3700h
170h
1900h
180
0.0036
LD
LD
LD
LD
97
25b
0.10
LD
7 1
53
Saline Watei
Maximum "Red Book"
M9/1 ug/I
LD
LD
LD
910h
180
14h
ID
10"
ID
LD
LD 0 OU1
0 037 0.004
LD
0.69
Lll
8400h
380h
440Uh
420
0 053 0.001
LD
0.16
LD
LO
220
668b
3.7 U.10
LD
140 p
120
14
-------
TABLE 11-2 (Continued)
COMPOUND
Nitrophenols
2 - Nitrophenol
4 - Nitrophenol
2,4 - Oinitrophenol
2,4 - Dinitro-6-methylphenol
2,4,6 - Trinitrophenol
N-Nitrosodiphenylamine
Pentachlorophenol
Phenol
Phthalate esters
Polychlorinated biphenyls
Polynuclear aromatic hydrocarbons
Selenium
Silver
2,3,7,8 - Tetrachlorodibenzo-
p-dioxin
Tetrachloroethene
Tha 1 1 i urn
Toluene
Toxaphene
Trichloroethene
Vinyl chloride
Zinc
dLO denotes IJCK of data
SA^ute toxicit, ievel
he vjlue in .g/1 should not excee exp [1
"The value in ug/1 should not exceed exp [1
'the value in ug/1 should not exceed exp [0
For freshwater and marine aquatic life, 0.1
The value in ug/1 should not exceed exp [2
The value in ug/l should not exceed exp [1
24 Hour
Average
ug/1
2700h
240h
79h
57h
1500h
LD
6.2
600
LD
0.014
LD
35
0.0090
LD
310
LD
2300h
0.013
1500
LD
47
03 n (hardness,
05 In (hardness)
94 In (hardness
Freshwater
Maximum "Red Book"
P9/1 ug/1
6200h
550h
180h
140h
3400h
LD
14
3400
LD 3.0
2.0b 0.001
LD
260 p
1.9 p
LD
700
LD
5200h
1.6 0.005
3400
LD
q P
-u DJJ w ere hardness is expressed in .nq;
-3 73] where hardness is expressed in nq,
-1 23)] .where hardness is expressed in rtg
times a 96 hr LCcn ds determined through nonaerated tnoss
15 In , hardness)
22 In (nardness)
-9 48] where hardness is expressed in mq/
-0 47] where hardness is expressed in rug.
0.01 times the 96 hour LC,~ value, using the receiving ar comparable water as the diluent and soluble
nlhe value in ug/1 should not exceed exp [0
°7he value in ^g/1 should not exceed exp [0
°For marine and/or fresh water aquatic life.
resident species
The value in t.g/1 should not exceed fxp [0
76 In ' hardness)
76 In (nardness)
0 01 of the %
;;j In (hardness)
*•! 06] where hardness is expressed in mg/1
+4 02] where haraness is expressed it\ mg/1
hour LCcn as determined through bioassay j
* 1 95] where hardness is expressed in "ig
Saline Hater
24 Hour
Average Maximum
ug/1 ug/1
LD LD
S3 120
37h B4h
LD LD
150h 340h
LD LD
3.7 8.5
LD LD
LD LD
0.030 10b
LD LD
54 410
0.26 2.3
79 180
LD LD
100 230
LD 0.07
LD LD
LD LD
58 170
1 is CaC03
1 as CaCOj
/I as CaC03
ay using a sensitive
i is CaCOj
: as CaC03
lead measurements
as CaC03
as CaC03
sing a sensitive
, 1 as CaCOj.
"Red Book"
ug/1
0.001
P
P
0.005
"Red Book" (U 5.EPA 1976)
Federal Register nn these dates.
March 15, 1979 - July 25, 1979 - October 1, 1979 - November 29, 198C
15
-------
TABLE 11-3
EPA LIST OF 129 PRIORITY POLLUTANTS AND THE RELATIVE FREQUENCY OF
THESE MATERIALS IN INDUSTRIAL WASTEWATERS
(After Keith anH Telliard, 1979)
Percent
Of »
Sables'
Number of Percent . Number of
Industrial. of Industrial
Categories Samples Categories
b
Purgeable Organic*
1.2
2.7
29.)
29.3
16.7
7.7
5.0
6.5
10.2
1.4
7.7
1.9
4.2
0.4
1.5
40.2
5
10
25
28
24
14
10
16
25
8
17
12
13
2
1
28
Acroleln
Acrylonitrlle
Benzene
Toluene
Ethylbenzene
Carbon tetrachloride
Chloroaer.zer.e
1,2-Dichlorcethane
1,1 . 1-Trichloroe thane
1.1-Dichloroethane
1,1-Dichloroethylene
1,1.2-Tnchloroetnane
1,1 ,2,2-Tetrachloroetnjne
Chloroethane
2-Chloroethyl vinyl ether
Chloroform
2.1
1.0
34.2
1.9
0.1
1.9
4.3
6.8 x
0.3
2.5
10.2
10.5
0.2
7.7
0.1
5
5
25
6
1
12
17
11
4
15
19
21
2
18
2
1,2-Dichloropropane
1,3-Dtchloropropene
Hethylene chloride
Methyl chloride
Methyl bromide
Brornofonn
01 chlorobroT^oe thane
Trichlorofluorcme thane
Dtchlorodl fluorome thane
Chlorodi bronoTie thane
Tetrachloroethylene
Trlchloroethylene
Vinyl chloride
1,2- trans -DicMoroe thy lene
b1s(Chloronethyl)ether
B as e/Heutral Extract able Organic Compounds
6.0
0.5
0.2
1.1
1.0
0.4
10.6
0.9
1.5
1.8
1.1
1.5
0.04
41.9
6.4
5.6
7.6
18.9
4.5
4.2
8.5
26.1
2.3
2.2
1.6
t.l
fc.&
0.3
0.4
0.2
0.6
0.8
0.2
0.5
0.5
0.1
0.04
0.1
0.2
0.2
0.2
18.2
19.9
14.1
30.7
53.7
55.5
43.8
33.4
9
5
1
7
8
3
18
9
13
9
3
9
1
29
12
15
20
23
12
14
13
25
11
9
6
6
18
3
4
2
4
6
4
3
5
3
1
2
2
3
2
20
19
18
25
28
28
27
19
!l,2-D1chlorobenzene
1,3-Dtchlorobenzene
1,4-Dichlorobenzene
Hexachloroethane
Hexachlorob'jtadiene
Hexach) orotenzene
1,2,4-Trichlorobenzene
bi s ( 2-Chlo roe tnoxy)rr,e thane
Naphthalene
2-Chloronaphthalene
Isophorone
Nitrobenzene
2.4-Dimtrotoluene
2,6-Dim trotoluene
4-Bromopnenyl phenyl ether
bts(2-Ethylhexyl)phthalate
Bl-n-octyl phtnalate
Dimethyl phthalate
Diethyl phthalate
Di-n-butyl phthalate
Acenaphthylene
Acenaphthene
Butyl benzyl phthalate
Acid Extractable Organic
Phenol
2-NHrophenol
4-N1 trophenol
2,4-Dinitrophenol
4,6-Dfni tro-o-cresol
Pentachloropnenol
Pesticides/PCB1
a-Endosulfan
B-Endosulfan
Endosulfan sulfate
o-BHC
B-BHC
4-BHC
f-BHC
Aldrin
Dieldrln
4,4'-DOE
4.4I-DDO
4.4--DOT
Endrin
Endrin aldehyde
Ketals
Ant ir.ony
Arsenic
Beryllium
Cadniun
Chromium
Copper
Lead
Miscellaneous
Total cyanides
5.7
7.2
5.1
7.8
10.6
2.3
1.6
1.8
3.2
0.8
0.2
0.6
0.1
0
0.2
1.1
0.8
0.1
1.2
0.1
0.1
1.4
Compounds
1.9
2.3
3.3
4.6
5.2
s
0.3
0.1
0.2
0.2
0.6
0.5
0.9
0.8
0.6
0.6
0.5
-
16.5
34.7
18.9
22.9
19.2
54.6
Not available
Not available
11
12
9
14
• r
lo
6
6
6
a
4
4
7
2
0
4
4
7
1
5
1
2
6
8
10
12
12
15
3
1
4
2
2
1
2
3
2
3
1
-
20
27
21
25
19
28
Fluorene
Fluoranthene
Chrysene
Pyrene
{Phenanthrene
Anthracene
Benzo(a)anthracene
Benzojbjfluoranthene
Benzofkjfluoranthene
Benzo(a)pyrene
Indenot 1 ,2 ,3-c ,d)pyrene
D1 benzol a, h) anthracene
Benzo(g,h,1 jperylene
4-CMorophenyl phenyl ether
3,3'-Dichlorobenz1dine
Benzldine
bis(2-Chloroethyl)ether
1,2-Diphenylhyarazine
Hexachlorocylcl open tad iene
N-Hi trosodl phenyl am) ne
N-Ni trosodi methyl ami ne
K-Nitrosodi-n-propylamlne
b1s(2-Chlorouopropyl Jether
p-Chloro-m-cresol
2-Chlorophenol
2,4-Dichlorophenol
2,4,6-Tnchloropnenol
2,4-DiMthylphenol
Heptachlor
Heptachlor epoxide
Chlordane
Toxaphene
Aroclor 1016
Aroclor 1221
Aroclor 1232
Aroclor 1242
Aroclor 1248
Aroclor 1254
Aroclor 1260
2,3,7,8-Tetrachlorodibenzo-p-d!oxin (TCD
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
Asbestor (fiborous)
Total phenols
"The percent of sarples represents the Purler of tir^s this co-sound fcjs found In all Si-pies In wllch it was analyzed for divided
the total as of 31 August 1978. Nu-^ers of Si-jles ran5ed fror, 2532 to 2958 with the average being 2617.
^A tctal of 32 IrdjStrul categaries ft 5-tcategcries were analyzed for orgjnics and 28 for ratals as of 31 August 1978.
16
-------
TABLE 11-4
MOST COMMONLY DISCHARGED PRIORITY POLLUTANTS
Non-Metals
Pollutant
Bis (2-Ethylhexyl) Phthalate
Chloroform
Methyl ene Chloride
Total Cyanides
Toluene
Benzene
Phenol
Di-.n-Butyl Phthalate
Ethyl benzene
Naphthalene
Phenanthrene and Anthracene
Metal
Pollutant
Copper
Zinc
Chromium
Lead
Nickel
Percent of
Samples
41.9
40.2
34.2
33.4
29.3
29.1
29.1
18.9
16.7
10.6
10.6
s
Percent of
Samples
55.5
54.6
53.7
43.8
34.7
Percent of
Industries
91
88
78
59
88
78
78
72
75
56
50
Percent of
Industries
100
100
100
96
96
17
-------
are known, the probability that a particular pollutant is discharged from at
least one of the plants is:
n / f. . \
pj ~~l ~ n I1" roo ) j = lj M •TI~P
where
f _ = relative frequency of discharge of pollutant type j from
plant type i, expressed as a percent
P. = probability that pollutant type j is discharged from at
least one of the n plants located on the water body.
M = number of toxic substances being analyzed
If the industrial categories of the plants are not known, then the
probability that a particular pollutant is discharged can be estimated using
Table 11-3 together with the following formula:
j = 1,M (H-2)
J 1 J-UW I
where
g = percent of samples containing pollutant j
j
P - probability that pollutant j is detected in at least one of
the n discharges.
Equation II-l is obviously the more accurate of the two formulae, because it
is based on a knowledge of the types of industries which discharge.
Although the above equations provide information on the likelihood that
different chemicals are discharged into the environment, and thus can be
used to prioritize investigative efforts, they do not predict quantities of
pollutants which are discharged. Chapter III can be used to generate that
type of information.
18
-------
2.1.5 Physical _and Chemi_caT_ Characteristics of Tqxi_c Organic Compound-:;
The most intensively investigated toxic pollutants, as a qroup, are the
priority pollutants. Because of the greater availability of data on
priority pollutants from such sources as Cal lahan et _a_1_. (1979), Billing
_et al_. (1975), and Mackay and Leinonen (1975), data are presented for
organic priority pollutants in the following categories:
• Halogenated Aliphatic Hydrocarbons (Table II-5)
t Pesticides (Table II-6)
» Polychlorinated Biphenyls (Table II-7)
« Monocyclic Aromatic Hydrocarbons (Table 11-8)
• Polycyclic Aromatic Hydrocarbons (Table II-9)
The properties of the pollutants tabulated in Tables II-5 through II-9 are:
• Vapor pressure, Torr (1 Torr - 1 mrn-Hg)
9 Solubility
t Octanol-water partition coefficient (K )
ow
» Volatilization half-life
• Qualitative statement of the importance of sorption.
Specific information is included in the tables for volatilization and
sorption because of the demonstrated importance of these processes in
governing the fate of many pollutants. In particular, for the approxi-
mately 103 organic priority pollutants:
• Sorption processes are important for 60
• Sorption is not important for 28
• It is not certain if sorption is important for the remaining 15
0 Volatilization is important for 52
• Volatilization is not important for 44
• It is uncertain if volatilization is important for the remaining 7.
The volatilization half-lives presented in the tables were typically
measured under a specific set of laboratory conditions, and consequently are
shorter than in most natural systems. Other useful properties such as
19
-------
SELECTED CHARACTERISTICS
TABLE 11-5
OF VARIOUS ALIPHATIC HYDROCARBONS
ro
o
Halogcnated Al iphatic
Hydrocarbons
Chloromethane
Oichloromethane
Trichloromethane (chloroform)
Tetrachlorome thane
(carbon tetrachtorlde)
Chloroethane
1 ,1-Oichloroe thane
1 ,2-Dicnloroethane
1, 1, 1-Tric hi oroe thane
1,1 ,2-Trichloroethane
1 , 1,2,2 -Tetrac hi oroe thane
Hexachloroethane
Chloroetnene
(vinyl chloride)
1 ,1-Dichloroethene
1 ,2-trans-Dichloroethene
Trichloroethene
Tetrachloroethene
1,2-Dichloropropane
1,3-Dichloropropene
Hexachlorobutadiene
Hexdchlorocyclopentadiene
Rromo methane
Onmudi Chloromethane
Di bromochl orpmrthanp
Trihromome thane
Dichlorodi fluorome thane
Trichlorof luoromo thane
Vapor Pressure (Torr)
at 20"C
3700
362
150
90
1000
180
61
96
19
5
0.4
2660
591
200
57.9
14
42
25
0.15
0.081 at 25"C
1420
50
15
10
4306
667
Soluhil ity
6450-7250 mg/1
at 20"C
13000-20000 mg/1
at 25°C
8200 mg/1 at 20'C
785 mg/1 at 20°C
5740 mg/1 at 20"C
5500 mg/1 at 20°C
8690 mg/1 at 20'C
440-4400 mg/1 at 20"C
4500 mg/1 at 20°C
2900 mg/1 at 20"C
50 mg/1 at 22"C
60 rog/1 at 10°C
400 mg/1 at 20°C
600 mg/1 at 20" C
1100 mg/1 at 20°C
150-200 mg/1
2700 mg/1
2700 mg/1
2
0.8 mg/1
900 mg/1
-
-
3000 mg/1
280 mg/1
1100 mg/1
^ow
8
20
93
400
35
60
30
150
150
360
2200
4
30
30
200
760
190
95
5500
10"
10
75
120
200
145
3400
Volati 1 iza tion
Half-Life
27 minutes3
21 m1nutesa
21 minutes3
29 minutes3
21 minutes3
22 minutes3
29 minutes3
20 minutes3
21 minutes
56 minutes3
45 minutes3
26 minutes3
22 minutes3
22 minutes3
21 minutes3
26 minutes3
<50 minutes3
31 minutes3
-
-
%30 minutes
-
-
-
few minutes
few minutes
'orption
Important?
HO
Probably Not
Probably Not
Uncertain
Probably Not
Probably Not
Probably Not
Probably Not
Uncertain
Uncertain
Uncertain
Probably Not
Probably Not
Probably Not
Probably Not
Probably Not
Probably
Uncertain
Probably
Probably
Probably Not
Uncertain
Uncertain
Uncertain
Probably
Uncertain
Stirring in an open container of depth 65 mm at 200 'PM filling e^aj.
-------
ro
TABLE 11-6
VARIOUS CHARACTERISTICS OF SELECTED PESTICIDES
Pesticide
Acroleln
Aldri'n
Chlordane
ODD
DDE
DDT
Dieldrin
Endosulfan
Endrln
Heptachlor
Heptachlor Epoxide
He xachlo recycle he xane
Lindane
Isophorone
TCDD
Toxaphene
Vapor Pressure (Torn)
220 at 20°C
330 at 30°C
2.3xlO's at 20°C
6xlO'6 at 25°C
IxltT5 at 25°C
10.2-18.9xlO'7 at 30°C
6.2-6.5x10"' at 20°C
l.SxlO'7 at 20°C
1.9X10'7 at 25°C
1.8xlO"7 to
2.9xlO"7 at 20°C
IxlO'5 at 25°C
ZxlO'7
3xlO'"
-
lO'Mo-'
io-"-io~6
0.38
-
0.2-0.4
Solubility
20.8% at 20°C
17-180 ppb at 25°C
0.056-1.85 ppm
20-100 ppb at 25°C
1.2-140 ppb at 20°C
2-85 ppb
186-200 ppb at 25°C
100-260 ppb at 20°C
220 ppb at 25°C
56-180 ppb at 25°C
200-350 ppb at 25°C
0.70-21.3 ppm at 25°C
5-12 ppm at 25°C
12000 ppm
0.2 ppb
0.7-3. ppm
Kow
0.8
"410
600
106
5x10*
10--10*
-
4x10'
4xl05
-
-
10"
5xl03
50
-
2000
Volatil ization
Hal f-Life
Uncertain
Few hours to
few days
Several weeks
1 day to 1 month
1 to 10 hours
4 hours-1 week
Few hours to
few days
11 days-1 year
-
-
-
-
100-200 days
Probably great
-
-
Sorption
Important?
No
Yes
Probably
Yes
Yes
Yes
Probably
Yes
Uncertain
Probably
Probably
Probably
Probably
No
Yes
Yes
Conditions described in ".allahan et al . (1979)
-------
TABLE II-7
SELECTED CHARACTERISTICS OF POLYCHLORINATED BIPHENYLS AND RELATED COMPOUNDS
ro
ro
PCBs and Related
Compounds
Aroclor 1016
Aroclor 1221
Aroclor 1232
Aroclor 1242
Aroclor 1248
Aroclor 1254
Aroclor 1260
2-chloronaphthalene
Percent
Chlorine
41
21
32
42
48
54
60
-
Density
(gm/cmf)
1.33
1.15
1.24
1.35
1.41
1.50
1.58
-
Vapor Pressure
at 25 C°(Torr)
4x10'"
6.7x10"'
4xlO's
4x10"*
4.9x10'*
7.7xlO's
4xlO's
0.017
Solubility
mg/1
0.42
15.
1.45
0.1-0.3
0.054
0.01-0.06
0.0027
6.47
KOW
2xlO*-3xlOs
600-10*
1.5xlO'-3xlO*
10*-4xl05
MO1
MO*
>10'
10*
Volatil ization
Half-Lives
In
laboratory
(hrs)»
9.9
-
-
12.1
9.5
10.3
10.2
-
Loss in b
Natural Systems
3.6X after 24 hours
4.2% after 24 hours
-
-
-
-
34X-67X after 12 weeks
aAt 25°C 1n 1 m' of water, 1 m deep (MacKay and Lelnonen, 1975).
bConditions described in Callahan et aj_. (1979)
-------
TABLE 11-8
SELECTED CHARACTERISTICS OF MONOCYCLIC AROMATIC HYDROCARBONS
CO
Monocycl ic Aromatics
Benzene
Chlorobenzene
1 ,2-Dichlorobenzene
Hexachlorobenzene
Ethyl benzene
Toluene
2,4-Oinitrotoluene
2,6-Dinltrotoluene
Pentachlorophenol
2-Nitrophenol
4-Ni trophenol
2,4-Dinitrophenol
4,6-Dinitro-o-cresol
Vapor Pressure (Torr)
95. at 25°C
MO at 20°C
1.5 at 25°C
ID"5 at 20°C
7
29 at 25°C
0.001 at 59°C
low
0.0001
1.0 at 49°C
2.2 at 146°C
-
-
Solubil i ty
1800 mg/1 at 25°C
^500 mg/1
145 mg/1
^20 ug/1
152 mg/1
535 mg/1
270 mg/1 at 22°C
^300 mg/1
14 mg/1
2100 mg/1 at 20°C
16000 mg/1 at 25°C
5600 mg/1
-
KOW
100
700
2400
MO6
1400
500
100
100
10s
60
80
34
700
Volatil ization
Half-Life
4.8 hrs at 25°Ca
0.5-9 hrs
8-9 hours3
8 hours3
5-6 hours3
5 hoursa
MOO days
MOO days
>100 days
-
-
-
-
Sorption
Important?
Uncertain
Probably
Probably
Yes
Probably
Probably
Yes
Yes
Yes
Yes
Yes
Yes
Yes
°Mackay and Leinonen (1975). Calculated b^sed on water depth of 1 m, and using ii'ass transfer
coefficients of 20 cm/hr and 3000 cm/hr for the liquid and gas transfer phases, respectively.
-------
TABLE 11-9
SELECTED CHARACTERISTICS OF VARIOUS POLYCYCLIC AROMATIC HYDROCARBONS
Polycyclic Aromatics
Acenaphthene
Acenaphthylene
Flourene
Naphthalene
Anthracene
Fluoranthrene
Phenanthrene
Ben2o[a]anthracene
Benzo[b]fluoranthrene
Benzo[k]fluoranthrene
Chrysene
Pyrene
Benzo[ghi]perylene
Benzo[a]pyrene
Dibenzo [a] anthracene
lndeno[l,2,3-cd]pyrene
Vapor Pressure (Torr)
lO'Mo-2 at 20°C
lO-'-lO"2 at 20°C
IQ-'-lO'2 at 20°C
.0492
2x10-* at 20°C
10- 6 to 10-* at 20°C
6.8x10-* at 20°C
5xlO'9 at 20°C
II"11 to 10"6 at 20°C
9.6x10-" at 20°C
10'11 to 10'6 at 20°C
6.9x O'7 at 20°C
MO-"
5xlO-9
MO-10
MO'10
Solubility
3.4 mg/1 at 25°C
3.93 mg/1
1.9 mg/1
32. mg/1
0.05-0.07 mg/1 at 25°C
0.26 mg/1 at 25°C
1.0-1.3 mg/1 at 25°C
0.01 mg/1 at 25°C
-
-
0.002 mg/1 at 25°C
0.14 mg/1 at 25°C
0.00026 mg/1 at 25°C
0.0038 mg/1 at 25°C
0.0005 mg/1 at 25°C
-
KQW
21,000
12,000
15,000
2.300
28,000
340,000
29,000
4xl05
4x10'
7xl06
4xl05
2xl05
10'
106
106
5x10'
Volatilization
Important?
Less than sorption
Less than sorption
Less than sorption
Less than sorption
Probably
Probably Not
Probably Not
No
Probably Not
Probably Not
Probably Not
Probably Not
Probably Not
Probably Not
Probably Not
Probably Not
Sorption
Important?
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
-------
molecular weight and specific gravity are available in standard references
such as Perry and Chilton (1973).
2.1.6 Scope and Organization of Chapter
The complexity of the transport and transformation processes wlvch
influence the fate of toxicants require additional analytical tools beyond
those required for conventional pollutants. This chapter develops these
analytical tools in a general way that is applicable to rivers, lakes, and
estuaries. Individual chapters on the various surface water types refine
these tools further and provide a framework within which to use them. When
used together, the various chapters in this document should help the user
both understand and quantitatively represent the processes influencing the
aquatic fate of a pollutant.
This chapter presents both a general overview of the screening approach
for toxicants and a detailed description of the processes included in the
screening methodology. The various topics are organized as follows:
• Screening Methods for Toxic Organic Substances
• Speciation Processes
1) Acid-base Effects
2) Sorption
t Transport Processes
1) Solubility Limits
2) Volatilization
• Transformation Processes
1) Biodegradation
2) Photolysis
3) Hydrolysis
These methods apply primarily to the fate of toxic organic substances.
Some processes act on metals as well, but considerable expansion of the
material would be necessary to incorporate them in these screening methods.
Generally, the complexity of the environmental chemistry of metals makes
them more difficult to handle with simple methods. The utility of these
25
-------
methods remain high even without metals since the overwhelming majority of
toxic substances and over 100 of the 129 EPA Priority Pollutants are organic
compounds. In lieu of procedures designed specifically for metals, the user
may apply the screening methods for conservative substances. The advantages
and limitations of this approach are discussed in Section 2.2.2.2.
26
-------
2.2 SCREENING METHODS FOR TOXIC ORGANIC COMPOUNDS
2.2.1 Modeling the Fate of Toxic Organics
The goal of this screening methodology for toxic pollutants is to help
the user identify surface water bodies where toxicants could reach hazardous
levels. Multiple approaches for identifying pollution problems are
possible, e.g. extensive field measurements, statistical correlations of
discharges and pollutants detected in rivers, computer simulation models,
etc. The approach taken here is to present simple methods for assessing the
fate of toxicants.
The application of any method necessitates the use of judgment on the
part of those applying it. In almost every case, the user must estimate
many of the methods' input parameters on the basis of limited data.
Consequently, even the projections of detailed computer models such as RAMS
(Burns, et _al_., 1981) and PEST (Park, et aj_., 1980) are only as good as the
accuracy of the assumptions made by their developers and users. Thus, the
goal of the materials presented herein is twofold: to present simple
methods and to provide the background necessary to make knowledgeable
judgments.
Predicting aquatic fate of pollutants involves several steps. Tho
steps described in the remainder of this section include:
* Determination of Fate-Influencing Processes
t Delineation of Environmental Compartments
• Representation of Hydrologic Flow
• Mathematical Representation of Speciation Processes
• Mathematical Representation of Transport and Transformation
Processes
• Determination of Pollutant Load and Mode of Entry into the
Aquatic Environment
Prediction of the fate of toxic pollutants requires the user to know
which processes act on the toxicant. Figure II-2 illustrates the transport
arid transformation processes which are of potential importance in a lake or
27
-------
po
oo
INFLOW
OUTFLOW
volatilization
V
HA:? H++A-
Acid-Base Equilibria
Precipitation-Dissolution
H-0--
H
Hydrolysis
Biodegradation
Bioconcentration
Reduction-Oxidation
FIGURE 11-2 SPECIATION, TRANSPORT AND TRANSFORMATION PROCESSES IN THE AQUATIC
ENVIRONMENT
-------
other surface water body. The processes fall into four categories as
follows:
• Loading Processes
The rates at which waste discharges, atmospheric deposition,
and land runoff introduce toxicants into natural waters
influence resulting pollutant levels.
» Speciation Processes
Acid-Base Equilibria - The pH of a natural water determines
the fraction of an organic acid or base in neutral or ionic
states, and therefore influences volatility.
Sorption - Hydrophobic organic compounds sorb to suspended
matter; their subsequent fate is influenced by the fate of
the suspended matter.
t Transport Processes
Precipitation-Dij_sgJution - Solubility limits of both organic
and inorganic pollutants can cause a pure pollutant phase to
form restricting its availability to transport and
transformation processes or substantially changing the
transport route.
Advection - Hydraulic flows transport pollutants which are
dissolved or sorbed on suspended sediments into and out of
particular aquatic habitats.
Volatilization - Organic pollutants may enter the atmosphere
from a water body, thereby reducing aquatic concentrations.
Sedimentation - Deposition of suspended sediments containing
sorbed pollutants, as well as direct sorption onto or
desorption from bottom sediments can alter pollutant
concentrations.
29
-------
• Transformation Processes
Biodegradation - Microbial organisms metabolize pollutants,
altering their toxicity in the process.
Photolysis - The absorption of sunlight by pollutants causes
chemical reactions which affect their toxicity.
Hydrolysis - The reaction of a compound with water frequently
produces smaller, simpler organic products.
R e d u c t i o n - 0 x i d at ion - Reactions of organic pollutants and
metals which involve the subtraction or addition of electrons
strongly influence their environmental properties. For
orga.nics, nearly all significant redox reactions are
microbially mediated.
• Bioaccumulation
Bio coneen tration - Uptake of toxic pollutants into biota via
passive means, e.g. absorption through fish gills.
B jomagji i f i c at "[on - Uptake of toxicants into biota via
consumption of contaminated food.
Once the pertinent processes have been identified, the physical
compartments of the environment between which the transport processes act
must be delineated. For most water bodies, compartments representing the
atmosphere, bottom sediments, and one or more water elements are sufficient.
These methods are capable of representing transport of pollutants between
the atmosphere and a water body. But rather than calculating atmospheric
concentrations of a pollutant, these methods generally assume them to be
close to zero unless available data indicate otherwise. Bottom sediments,
however, frequently accumulate high levels of organic pollutants. Because
of the difficulty of modeling the behavior of toxicants in sediments,
usually assumptions which approximate only the removal or addition of a
pollutant to the water column are made. These approximations are presented
in the individual chapters on each water body.
30
-------
The next step in assessing the aquatic fate of toxic pollutants is to
represent the advection or flow of water. Figure 11-3 illustrates a
representation of rivers as a segregated flow system and lake layers as
completely mixed flow systems. Although these models are simple, they serve
as adequate first-approximations of real systems. Refinements and
limitations of these flow system models are considered in the individual
chapters on rivers, lakes, and estuaries.
The transport and transformation processes responsible for the removal
of a pollutant from the water column are considered next. First-order rate
expressions adequately represent all of the processes considered here. The
first-order decay of a pollutant by a process is represented as follows:
Rate of Pollutant Removal = k • C (H-^
i T ( '
where
k. = first-order rate constant for process i
C = total concentration of pollutant
The rate constant for a process is specific to both the chemical it acts
upon and the local environment in which it acts.
When all the first-order processes act independently, the total rate of
pollutant removal is:
Total Rate of Removal = k • C (H-4)
where
k = k +k+k+k+k (II-5)
T vm S B P H ;
k = specific mixed-body volatilization rate constant
vm
k = specific rate constant for removal to bottom sediment
O
31
-------
COMPLETELY MIXED FLOW
NATURAL SYSTEM:
LAKE
IDEALIZATION:
SEGREGATED FLOW
L
MIXED FLOW
NATURAL SYSTEM:
IDEALIZATIONS:
RIVER
Pi UG Fi ow
FLOW WITH
AXIAL DISPERSION
FIGURE 11-3 FLOW SYSTEM REPRESENTATIONS
32
-------
k = specific rate constant for biodegradation
B
k = specific rate constant for photolysis
P
k = specific rate constant for hydrolysis
H
The additivity of processes which are first-order with respect to pollutant
concentration is particularly convenient for analysis.
Many of the decay processes are influenced by the chemical state of the
toxicant. For example, sorbed pollutants cannot volatilize. Mathematical
representations of equilibria between two species of a chemical can be
reduced to the following type of equation. This type of equation serves
well at the low solute concentrations encountered in waste waters and
natural waters:
C = K C (H-6)
i ij J
where
C = concentration of form i
i
K = equilibrium constant
C = concentration of form j
j
It is also convenient to know the fraction of the total pollutant
concentration which is in a given state:
C,
a =
i C,
1 (II-7)
where
C = concentration in state i
i
33
-------
c = c + c
T S
C = total dissolved phase pollutant concentration
C = total sorbed phase pollutant concentration
o
To complete the assessment of the aquatic fate of a pollutant the mode
of entry into the aquatic environment must be considered. Many pollutants
enter in dissolved or sorbed form from a point source. In this case, a
simple mixing computation is sufficient to determine the initial
concentration of a pollutant in the water body. Other cases include spills,
non-point sources, and desorption from sediments. Chapter 4 presents
methods for dealing with these cases.
The user may now reckon the concentration of a pollutant in a given
water body. The equations which yield the desired results are specific to
each surface water type and are developed in the individual chapters on
lakes, rivers, and estuaries. An equation representative of those in each
chapter is presented in Table 11-10. The individual chapters go into
greater detail about factors influencing rate processes and interactions
with other important phenomena in each water body (See Sections 4.9, 5.6,
6.4.3, and 6.4.5).
2.2.2 Use of Assessment Techni ques as Screening_Tqol_s
2.2.2.1 Making Conservative Assumptions
With the computational methods presented in this document, the user
could produce a relatively complete analysis of the aquatic fate of a
pollutant. The goal of this screening method, however, is to
determine—with a minimum of effort—whether toxicants are likely to reach
problem levels in surface water bodies for either existing or projected
loading rates. The user can minimize the effort expended in screening a
pollutant by starting with a simple approach which incorporates conservative
assumptions about the fate of a pollutant. Conservative assumptions are
designed to yield higher calculated environmental concentrations than
34
-------
TABLE 11-10
EXPRESSIONS FOR TOXIC POLLUTANT LEVELS
IN VARIOUS WATER BODIES
Water Body
Expression for Steady-State
Pollutant Concentration
Rivers
(Chapter IV)
[-k1 -^r*K
. ..Y— . .^"^. »
1 + K S U
P J
(IV-115)
Impoundments
(Chapter V)
Estuaries
where x = distance downstream
U = river velocity
C = total dissolved phase concentration
C = Cin/(l + Tw x k) (V-47)
where T = hydraulic residence time
W
C = total dissolved and sediment phase
concentration
f.
(VI-33)
(VI-34)
where C.
f.
r.
t =
concentration in segment i
fraction of fresh water in segment i
segment i exchange ratio
time expressed in tidal cycles
35
-------
probably exists in the real system. If these higher concentrations are
below the water quality criterion under consideration, a violation of the
standard is unlikely. If the initial predictions are higher than the
standard, the user may successively refine the approach until it becomes
apparent that either the standard will be met or that a more detailed study
is necessary.
Three levels of refinement in assessing the aquatic fate of a pollutant
are considered here. In order of increasing complexity, they are:
1) Treating the pollutant as a conservative substance
2) Considering transport and speciation processes
3) Considering transformation, transport, and speciation
processes.
Each approach has advantages and limitations which the user should consider.
By following this sequence of refinements, the user should be able to
eliminate cases where water quality problems are unlikely with a minimum of
time and effort.
2.2.2.2 Treating the Pollutant as a Conservative Substance
The simplest approach to estimating the concentration of a toxic
pollutant is to assume it behaves conservatively (i.e. does not undergo
reaction):
k = 0
Unless an internal source of the pollutant exists, this approach will yield
the highest possible pollutant levels since pollutant decay and removal
processes are neglected. The obvious advantage of this approach is that it
requires no chemical or environmental data to evaluate rate and equilibrium
constants. The only data needed are pollutant loads and hydrological
parameters. Its major drawback is that it neglects the possibility of a
36
-------
compound accumulating in another environmental compartment, especially
bedded sediments. This could result in the underestimation of the duration
of the exposure of an aquatic habitat to a chemical. Although the duration
of exposure may be underestimated, water column concentrations would not
exceed the upper limits predicted by this approach at any time during the
exposure period. The fate of conservative pollutants in rivers,
impoundments, and esturaries is discussed in Sections 4.1.9, 5.6.1, and
6.4.
2.2.2.3 Considering Transport and Speciat ion Processes
This refinement incorporates those processes which influence pollutant
transport out of the aquatic environment but neglects those processes which
chemically alter the compound. Transport processes strongly depend upon
chemical speciation, which therefore must be included. The rate constant
for first-order pollutant attenuation in this approach is:
k = k + k (II-8)
T S vm
where
k = specific rate constant for removal to bottom sediment
k = specific mixed body volatilization rate constant.
vm
This approach requires more information on the properties of the toxicant
and the environment than when the pollutant is assumed to behave
conservatively, but the necessary data are much more readily available than
that required to characterize transformation processes. Nearly all the
chemical data necessary to characterize acid-base equilibria, sediment
sorption, solubility limitations, and volatilization for the organitza
priority pollutants are presented in tables in Sections 2.1.5, 2.3.1, and
2.4.2. The necessary environmental data can usually be obtained or
estimated with a minimal amount of effort. Because of the demonstrated
importance of transport processes and the relative simplicity of assessing
them, this is a good intermediate step between the simplest and most
complicated approaches.
37
-------
Transport and speciation processes are applied specifically to
rivers, impoundments, and estuaries in Sections 4.9, 5.6, 6.4.3, and 6.4.5.
2.2.2.4 Considering Transformation, Transport, and Speciatipn Processes
The most complex model which the user can employ using these screening
methods includes consideration of transformation, transport, and speciation
processes. With this approach, the rate constant for first-order
attenuation of a pollutant is:
k=k+k +k+k+k (II-9)
T S vm B P H
where
k = specific rate constant for biodegradation
B
k = specific rate constant for photolysis
P
k = specific rate constant for hydrolysis
H
The inclusion of the degradative processes (i.e. biodegradation,
photolysis, and hydrolysis), considerably increases the chemical and
environmental data required to model a compound's fate. Rather than
accurately determining all the constants for speciation, transport, and
transformation, the user should first ascertain which processes are the most
significant for a compound. As a first step the user should obtain data on
the properties of the chemical which influence its aquatic fate from this
document or other sources. From compound specific data, it is usually
possible to eliminate some processes from consideration. For organic
priority pollutants, consulting the ratings of the relative importance of
aquatic processes for the fate of each compound, Table 11-11, may aid the
user in eliminating unimportant processes. Once the most significant
processes have been identified, the user should collect the environmental
data necessary to determine site specific constants. These site specific
constants are then applied in the appropriate equation for each water body
type to obtain the best estimate of the actual pollutant concentrations in
38
-------
TABLE 11-11
RELATIVE IMPORTANCE OF PROCESSES INFLUENCING
AQUATIC FATE OF ORGANIC PRIORITY POLLUTANTS (After Callahan et al_., 1979)
Compound
Process
PESTICIDES
Acrolein
Aldrin
Chlordane
ODD
DDE
DDT
Dieldrin
Endosulfan and Endosulfan Sulfate
Endrin and Endrin Aldehyde
Heptachlor
Heptachlor Epoxide
Hexachlorocyclohexane (a,3,6 isomers)
-Hexachlorocyclohexane (Lindane)
Isophorone
TCDD
Toxaphene
PCBs and RELATED COMPOUNDS
Polychlorinated Biphenyls
2-Chloronaphthalene
HAL06ENATED ALIPHATIC HYDROCARBONS
Chloromethane (methyl" chloride)
Dichloromethane (methylene chloride)
Trichloromethane (chloroform)
Tetrachloromethane (carbon tetrachloride)
Chloroethane (ethyl chloride)
1,1-Dichloroethane (ethylidene chloride)
1,2-Dichloroethane (ethylene dichloride)
1,1,1-Trichloroethane (methyl chloroform)
1,1,2-Trichloroethane
1,1,2,2-Tetrachloroethane
Key to Symbols:
++ Predominant fate determining process - Not likely to be an important process
+ Could be an important fate process ? Importance of process uncertain or not
known
39
-------
TABLE 11-11 (continued)
Compound
Process
Hexachloroethane
Chloroethene (vinyl chloride)
1,1-Dichloroethene (vinylidene chloride)
1,2-j,rajns-Di chloroethene
Trichloroethene
Tetrachloroethene (perchloroethylene)
1,2-Dichloropropane
1,3-Dichloropropene
Hexachlorobutadiene
Hexach1orocyclopentad i ene
Bromomethane (methyl bromide)
Bromodichloromethane
Dibromochloromethane
Tribromomethane (bromoform)
Dichlorodifluoromethane
Trichlorofluoromethane
HALOGENATED ETHERS
Bis(choromethyl) ether
Bis(Z-chloroethyl) ether
Bis(2-chloroisopropyl) ether
2-Chloroethyl vinyl ether
4-Chlorophenyl phenyl ether
4-Bromophenyl phenyl ether
Bis(2-chloroethoxy) methane
MONOCYCLIC AROMATICS
Benzene
Chlorobenzene
1,2-Dichlorobenzene (^-dichlorobenzene)
1,3-DiChlorobenzene (m-dichlorobenzene)
1,4-Dichlorobenzene (£-diChlorobenzene)
1,2,4-TriChlorobenzene
Hexachlorobenzene
7
+
-f
+
4-
+
+
+
+
+
Key to Symbols:
++ Predominant fate determining process - Not likely to be an important process
+ Could be an important fate process ? Importance of process uncertain or nol
known
40
-------
TABLE 11-11 (continued)
Compound
Process
Ethylbenzene ?+?___
Nitrobenzene +__+__
Toluene ++?___
2,4-Dinitrotoluene + - - + - ?
2,6-Dinitrotoluene +_-+??
Phenol - + + + __
2-Chlorophenol --?+__
2,4-Dichlorophenol - ++
2,4,6-Trichlorophenol ? - ? ?
Pentachlorophenol + - + + _ +
2-Nitrophenol - ++b
4-Nitrophenol + ++
2,4-Dinitrophenol + ++b
2,4-Dimethyl phenol (2,4-xylenol) -_?+__
jD-chloro-m-cresol - - ? ++
4,6-Dinitro-o-cresol + - ++ ? ?
PHTHALATE ESTERS
Dimethyl phthalate +_+__+
Diethyl phthalate +.+__+
Di-n-butyl phthalate +_+__+
Di-n-octyl phthalate +_+__+
Bis(2-ethylhexyl) phthalate +_+__+
Butyl benzyl phthalate +_+__+
POLYCYCLIC AROMATIC HYDROCARBONS
Acenaphthene0+-++__
Acenaphthylene0 +-++__
Fluorenec +-++__
Naphthalene +-++__
Anthracene ++++__
Fluoranthene0 ++++__
Phenanthrene0 ++++__
Benzo(a)anthracene ++++__
Benzo(b)fluoranthene0 +-++__
Benzo(k)fluoranthene +-++__
Chrysene +-++__
Key to Symbols:
++ Predominant fate determining process - Not likely to be an important process
+ Could be an important fate process ? Importance of process uncertain or not
known
41
-------
TABLE 11-11 (continued)
Compound
Process
Pyrene
Benzo(ghi)perylenec
Benzo(a)pyrene
Dibenzo(a,h)anthracene
Indeno(l,2,3-cd)pyrene
NITROSAMINES AND MISC. COMPOUNDS
Dimethylnitrosamine
Diphenylnitrosamine
Di-n-propyl nitrosamine
Benzidine
3,3'-Dichlorobenzidine
1,2-Diphenylhydrazine (Hydrazobenzene)
Aery Ionitrile
Key to Symbols:
++ Predominate fate determining process - Not likely to be an important process
+ Could be an important fate process ? Importance of process uncertain or not
known
Notes
a Biodegradation is the only process known to transform polychlorinated biphenyls
under environmental conditions, and only the lighter compounds are measurably
biodegraded. There is experimental evidence that the heavier polychlorinated
biphenyls (five chlorine atoms or more per molecule) can be photolyzed by
ultraviolet light, but there are no data to indicate that this process is operative
in the environment.
based on information for 4-nitrophenol
c Based on information for PAH's as a group. Little or no information for these
compounds exists.
42
-------
the environment that these methods are capable of making. (See Sections
4.9, 5.6, 6.4.3, and 6.4.5).
Frequently, kinetic and equilibrium constants will depend on the values
of parameters which the user must estimate (e.g. pH). In such cases,
assuming conservative values is the best policy. However, calculations
using a range of values may identify processes for which a more careful
determination of the key environmental and chemical parameters is warranted.
Example II-l is an overall example for this chapter. It demonstrates
the initial steps a user would take in applying these methods to assess the
fate of a particular organic pollutant. The example follows the three level
analysis described above and also draws upon some of the procedures for
specific environmental processes which are developed later in this chapter.
This example can serve as a guide to evaluating the importance of the
various fate influencing processes for a particular pollutant.
EXAMPLE II-l
Pentachlorophenol in the Aurum Mirth Watershed
Pentachlorophenol enters the Aurum Mirth River from a continuous point
source. The river is the sole tributary to Lake Castile. After mixing at
the point of entry, the concentration of pentachlorophenol in the river is
20yg/l. The travel time from the point of contamination with
pentachlorophenol to Lake Castile is about 6 days. The mean hydraulic
residence time in Lake Castile is 10 days.
Use the screening methods to determine which chemical and environmental
parameters are of the greatest importance for predicting the fate of
pentachlorophenol in the watershed's surface waters.
43
-------
1) TREATING PENTACHLOROPHENOL AS A CONSERVATIVE SUBSTANCE
The first step in the screening method is to assess the fate of
pentachlorophenol treating it as a conservative substance. Sections 4.1.9,
5.6.1, and 6.4 discuss the fate of conservative pollutants in rivers, lakes,
and estuaries. In this case, we assume no further dilution of the
pentachlorophenol occurs in either the lake or the river. Consequently, the
conservative pollutant approach predicts a mean concentration in the river
and lake of 20 yg/1.
Table 11-2 lists a proposed water quality standard for
pentachlorophenol. The 24 hour mean concentration must be less than
6.2 yg/1. Since 20 yg/1, exceeds this standard, a second level assessment
is in order.
Prior to applying the next two levels of analysis it is worthwhile to
check Table 11-11 for the relative importance of the different
transformation and transport processes. Table 11-11 summarizes the
influence of the aquatic processes on pentachlorophenol as follows:
• Sorption - Important process
• Volatilization - Not an important process
• Biodegradation - Important process
t Direct Photolysis - Important process
• Hydrolysis - Not an important process
• Bioaccumulation - Important process
It will be instructive to compare these statements to the results of the
screening methodology.
44
-------
2) CONSIDERING TRANSPORT AND SPECIATION PROCESSES
To analyze transport and speciation processes, first examine each
process for its potential influence on the fate of pentachlorophenol.
Speciation Processes
Acid-Base Effects (Section 2.3.1)
The chemical and environmental parameter governing acid-base effects
are:
Chemical Parameters:
• pK or pK - acid or base equilibrium constants
a b
Environmental Parameters:
• pH - hydrogen ion concentrations
The pK of pentachlorophenol is 4.74, as shown in Table 11-13.
a
According to Table 11-12, at least 90 percent of the pentachlorophenol will
be in the anionic state at pH's greater than 5.74. As long as the pH in the
Aurum Mirth River and Lake Castile remain above 5.74, the properties of
pentachlorophenol as measured for neutral waters will remain unaffected.
But, because pH's below 5.74 could significantly alter the behavior of the
compound, it is important to determine actual surface water pH values.
Sorption (Section 2.3.2)
The key environmental and chemical parameters which influence sorption
are:
45
-------
Chemical parameters:
• K octanol-water coefficient
ow
• S - solubility in water
w
Environmental Properties:
• Suspended sediment concentration
• organic carbon content of the suspended sediment
Table II-8 lists the solubility and octanol-water coefficient of
pentachlorophenol as:
S = 14 mg/1
K = 105
ow
Assuming an organic carbon content of 2 percent for the suspended sediments,
calculate K using Equations 11-18 and 11-16:
P
5
K - (.02) (.63) (10 ) - 1300
P
According to Table 11-14, greater than 10 percent of the pentachlorophenol
will be in the sorbed state at suspended sediment concentrations exceeding
100 mg/1. The relatively strong sorption of pentachlorophenol dictates that
the suspended sediment concentration in the Aurum Mirth River and the
sediment trapping efficiency of Lake Castile be investigated further.
Sorption of pentachlorophenol potentially affects both its speciation and
its transport rates.
46
-------
Transport Processes
Solubi1 ity Limitations (Section 2.4.1)
The most important chemical and environmental factors which influence
solubility of a compound are:
Chemical Parameters:
• S - Aqueous equilibrium solubility
w
Environmental Parameters:
• T - Temperature
• Salinity
Table II-8 lists the solubility limit for pentachlorophenol as 14 mg/1
(14000 ug/1). At no point in the Aurum Mirth watershed should the
solubility of pentachlorophenol restrict the ability of the aqueous phase to
transport it.
Volatilization (Section 2.4.2)
The most significant chemical and environmental properties which
influence volatilization are:
Chemical Parameters:
• K - Henry's Law Contant
H
Environmental Parameters:
k - reaeration constant
a
47
-------
• V - wind speed
t Z - mixed depth of water body
It is possible to estimate the Henry's law constant for
pentachlorophenol from its vapor pressure and aqueous solubility using
Equation 11-32. However, it is simpler to rule out volatilization as a
significant transport process on the basis of the volatilization half-life
of 100 days given in Table II-8. Because laboratory volatilization
half-lives are shorter than the true environmental values, it is safe to
assume the environmental half-life will be much greater than 100 days.
Given a total system mean hydraulic residence time of only 16 days (6+10),
volatilization can be safely neglected.
Summary. Acid-base equilibria and sorption significantly influence the
transport and speciation of pentachlorophenol in the aquatic environment.
Acid-base effects do not influence the near-neutral volatilization and
photolysis rate constants presented in this document as long as pH's remain
above 5.7. Sorption is a potentially important speciation process.
Consequently, the pH values and suspended sediment concentrations should be
determined in order to accurately evaluate these processes.
The strong tendency of pentachlorophenol to sorb on sediments may
result in sedimentation serving as a significant removal process in Lake
Castile. The absence of net sediment deposition in the river implies that
transport processes do not reduce pentachlorophenol concentrations in the
Aurum Mirth. Thus, the second level analysis predicts a total concentration
of 20 yg/1 of pentachlorophenol in the Aurum Mirth River with lower levels
possible in the lake. Because the predicted river concentrations exceed the
standard, the third level model is necessary.
48
-------
3) CONSIDERING TRANSFORMATION, TRANSPORT, AND SPECIATION PROCESSES
To consider transformation, transport, and speciation processes, the
transformation processes which were neglected in the level two analysis must
be examined for their potential importance in influencing the rate of
pentachlorophenol degradation.
Transformation Processes
Biodegradation (Section 2.5.1)
The key chemical and environmental variables which influence
biodegradation are:
Chemical Parameters:
• Metabolic Pathway (growth or co-metabolism)
• k - Biodegradation rate constant
B
Environmental Parameters:
e Bacterial population size
t State of adaptation
• Inorganic nutrient concentrations - Phosphorus
• Dissolved oxygen
t Temperature
• Pollutant concentration
49
-------
According to Table 11-23, pentachlorophenol is potentially
biodegradable, although adaptation may be slow. The reported specific rate
constant values, 0.1 to 1.0 per day, in Table 11-24 are in the same range as
the 0.05 to 0.5 per day values suggested in Table 11-23. Although both rate
constants were determined under laboratory rather than environmental
conditions, they do indicate that pentachlorophenol can degrade very
rapidly.
Table 11-24 also indicates that pentachlorophenol is used by bacteria
as a growth substrate. Thus, the time required for adaptation is of primary
concern. The most important environmental factors for determining whether
microorganisms in the Aurum Mirth watershed will adapt to degrade
pentachlorophenol are previous exposure, time, and the actual concentrations
of pentachlorophenol in the surface waters (too low--no enzyme induction;
too high—may have toxic effect on microbiota).
Photolysis (Section 2.5.2)
The key chemical and environmental characteristics influencing the rate
of photolysis are:
Chemical Properties
• k Near-surface rate constant
do
or • eU) - Light absorption coefficient of pollutant, and
• Quantum yield
Environmental Properties;
• I - Solar radiant flux
50
-------
t Z - Mixed depth of water body
• K - Diffuse light attenuation coefficient
a) Z - Secchi disc depth
sd
b) C - Suspended sediment concentration
ss
C - Dissolved organic carbon concentration
DOC
C - Chlorophyll pigment concentration
a
According to Table 11-29, the near-surface photolysis rate constant for
pentachlorophenol is .46/day. The size of the rate constant implies that
photolysis would be an important factor if the water bodies are not too deep
or too turbid. Thus, it is important to gather information on the water
depths, and to estimate the light attenuation coefficients, and the solar
radiant flux in the Aurum Mirth watershed.
Hydrolysis (Section 2.5.3)
The important parameters influencing the rate of hydrolysis are:
Chemical Parameters:
• k , k , k - Acid, neutral, and base catalyzed hydrolysis rate
a n b
constants
Environmental Properties:
t pH - concentration of hydrogen ion in the water bodies.
51
-------
Table 11-31 gives acid and base hydrolysis rate constants for
4 -1 -1
pentachlorophenol of 1.1 x 10 and 3.3 liter mole day . The neutral rate
constant is 5.8 x 10 per day. The same table lists a half life of
100 days at pH = 7. Because the acid catalyzed rate constant is large,
significantly higher rates could occur at lower pH's. Using Equation 11-85,
the rate constant for pH = 5 is:
k = 1.1 x 104 (10~5) + 5.8 x 10"3 + 3.3 (10~9)
H
= .23 day"1
At this lower pH, degradation by abiotic hydrolysis would be very rapid.
Thus, determining the pH in the Aurum Mirth River and Lake Castile is
very important.
Summary
The consideration given to transformation, transport, and speciation
processes indicates the following processes are of potential importance to
the fate of pentachlorophenol in the Aurum Mirth watershed:
• Acid-Base Effects
t Sorption
• Biodegradation
• Photolysis
• Hydrolysis
Since the three transformation processes are potentially important, there is
a good possibility that the initial pentachlorophenol concentration of
20 yg/1 will be reduced below the 6.2 yg/1 standard. Therefore further
analysis as presented in the specific water body sections is warranted.
52
-------
The results of this example agree with the summary of rate processes
given in Table 11-11 except for the case of hydrolysis. This demonstrates
that the process summary table can serve as a useful guide but should be
supplemented with actual data whenever possible.
END OF EXAMPLE II-l
53
-------
2.3 SPECIATION PROCESSES
2.3.1 Acid-Base Effects
The fate of toxic organics which are either acids or bases can be
strongly affected by the concentration of hydrogen ions in a water body. It
is therefore necessary to have a means for estimating this influence. This
section will first present a brief review of acid-base equilibria and then
will give a technique for quantifying the influence of hydrogen ion
concentration on the behavior of toxicants.
2.3.1.1 Acid-Base Equilibria
+
Acids by definition donate hydrogen ions, H , to solution. Bases, by
definition, accept hydrogen ions from solution. 2-Nitrophenol, one of the
129 priority pollutants, is an acid and donates hydrogen ions as shown by
the following reaction:
OH
2-nitrophenol 2-nitrophenolate + hydrogen ion
(HP) (P-) (H+)
Acid-base reactions are extremely fast and can be represented by
equilibrium expressions. For the above reaction the expression would be:
[HP]
where
54
-------
[H ] = concentration of hydrogen ions, moles/liter
[P ] = concentration of nitrophenolate ions, moles/liter
[HP] = concentration of undissociated nitrophenol, moles/liter
K = an equilibrium constant for acid dissociation (also called
a
an acidity constant)
The extent to which any acid will donate hydrogen ions to the solution
depends on how many hydrogen ions are in solution (the concentration of
hydrogen ions) and on the strength of the acid.
The concentration of free hydrogen ions in natural waters can range
-4 -10
from about 10 to 10 moles per liter. Hydrogen ion concentrations are
normally expressed in pH units. In dilute solutions, such as natural
waters, pH is defined as the negative logarithm of the molar hydrogen ion
concentration (pH = -log [H ]). For the above two concentrations the pH
values are 4 and 10.
The strength of an acid is quantified by the equilibrium constant, K .
For very strong acids (those which most readily donate hydrogen ions) the
value of this constant is greater than unity. Included in this group are
strong acids such as hydrochloric and nitric acid. Toxic organic acids,
though, are generally weak acids and have K values between 10 and 10 .
a
K values are typically expressed in terms of negative base ten logarithms.
When this approach is used the equilibrium constants are called "pK "
(pK = -log K ).
a 10 a
When the pH of a solution is the same as the pK value of an acid
(i.e. pH = pK ), 50 percent of the acid will have donated its hydrogen ions
9
to the solution and will exist as a charged anionic species. For pH values
greater than the pK value by one or more units, the acid will have donated
3
essentially all of its hydrogen ions to the solution and will exist in the
anionic form (e.g. P ).
55
-------
The extent to which any base will extract hydrogen ions from solution
depends upon the concentration of hydrogen ions in solution (pH) and on the
strength of the base. The strength of a base is quantified by an
equilibrium constant, K . For very strong bases (those that most readily
b
extract hydrogen ions from solution) the value of K is of the order of 1.
Toxic organic bases are generally weak and have K values between 10 and
-10 b
10 . In a manner similar to acids, K is typically expressed in terms of
negative base ten logarithms and is called "pK " (pK = -log K ).
b b 10 b
Water itself can behave as a weak acid or a weak base:
H20 * H+ + OH~ (acidic behavior)
H20 + H+ * H30+ (basic behavior)
Note that [H ]-[OH~] = K
w
where
[OH ] = the concentration of hydroxide ion, moles/1
-14 o
K =10 , at 20 C
w
pK =14, at 20°C
w
When the pH of a solution equals the pK of a base, 50 percent of the
b
base has accepted hydrogen ions and will exist as a charged cationic species.
For pH values greater than one unit above the value of (pK -pK ),
w b
essentially all of the base will exist in electrically neutral form
(e.g. NH ). For pH values less than the value of (pK -pK, ) by 1 or more units
the base will essentially exist in the electrically charged cationic form
(e.g. NH*).
Table 11-12 summarizes the behavior described above for acids and
bases. Values for pK and pK for selected toxic organic acids and bases
a b
and values of pK are given in Table 11-13.
w
56
-------
en
TABLE 11-12
OCCURRENCE OF ACIDS AND BASES IN NEUTRAL AND CHARGED
FORMS AS A FUNCTION OF pH, pK , AND pK,
a o
Acids
Definition: Hydrogen
Example
HN03
General
HP -
^
pK +2
pK +1
P
pKfl-l
PKa-2
PKa-3
— +~ H + + NO"
Reaction:
•*- H+ + P"
Speciation :
Fraction in
Neutral Form
0.001
0.01
0.09
0.5
0.91
0.99
0.999
ion donors
Fraction in
Ionic Form
0.999
0.99
0.91
0.5
0.09
0.01
0.001
Bases
Definition: Hydrogen ion
Example:
3 4
General Reaction:
B + H+ — »- BH+
Speciation:
Fraction in
pH Neutral Form
pK -pK +3 0.999
pK -pK. +2 0.99
pK _pK +1 0.91
pK -pK 0.5
pK _pK _1 0.09
pKw-pKb-2 0.01
pKw"pl
-------
TABLE 11-13
pK AND pK. VALUES FOR SELECTED TOXIC ORGANIC
a b
ACIDS AND BASES AND VALUES OF pK FOR WATER
w
Acids
Phenol
2-Chlorophenol
2,4-Dichlorophenol
2,4,6-Trichlorophenol
Pentachlorophenol
2-Nitrophenol
4-Nitrophenol
2,4-Dinitrophenol
2,4-Dimethylphenol
4,6-Dinitro-o-cresol
Bases
Benzidine
10.0
8.52
7.85
5.99
4.74
7.21
7.15
4.09
10.6
4.35
pKb b
9.34, 10.43
Water
Freshwater
pK
w
14.63 at 5UC
14.53 at 10°C
14.35 at 15°C
14.17 at 20°C
14.00 at 25°C
13.82 at 30°C
Seawater
14.03 at 5un
13.81 at 10°C
13.60 at 15°C
13.40 at
13.20 at
20°C
25°C
13.00 at 30°C
Notes:
a
All pK values from Callahan et al (1979)
a
All pKb values from Weast and Astle (1980)
c pK values from Stumm and Morgan (1981) and from Dickson and Riley (1979)
W
58
-------
Since toxic organics almost always exist in very low concentrations and
are at best only weak acids or weak bases, they will have little influence,
if any, on the pH values of the water. The hydrogen ion concentration of
the water will, however, determine whether acids or bases exist in neutral
or ionic forms.
Values of pH for natural waters can be obtained from the USGS, the
U.S. EPA, and state and local agencies. Waters with low alkalinities
(e.g. < 50 mg/1 as CaCO , or 1 milliequivalent/liter) are quite susceptible
~ 3
to changes in pH due to natural processes such as photosynthesis and
respiration and even to relatively small additions of strong acid or base.
Selection of representative pH values for such waters will require more data
than for systems with higher alkalinities where less change in pH can be
anticipated.
2.3.1.2 Quantifying the Influence of pH on Toxicant Volatilization
Only electrically neutral species are directly volatile.
Volatilization rate expressions must therefore use as the concentration of
toxicant only that fraction which is electrically neutral (non-ionic). The
fraction of an acid or base which is in the non-ionic form can be determined
by use of the expressions given below:
For organic acids:
a = A°- = -L_ (11-11)
For organic bases:
aBo = "B~ = 'r~"—-"—rrrr 01-12)
where
59
-------
a = the decimal fraction of the organic acid which is in the
Ao
electrically neutral (non-ionic) form
a = the decimal fraction of the organic base which is in the
Bo
electrically neutral (non-ionic) form
A = the total dissolved concentrations of the toxic organic acid
(e.g. HP+P ), also called the analytical concentration of A
B = the total dissolved concentration of the toxic organic base
(e.g. BH + B), also called the analytical concentration of
B
The rate expressions then become in general form:
and
R = k a A (II-13a)
v A
o
R = k a B (II-13b)
v B
o
where
R = rate of volatilization
k = specific rate constants for volatilization
60
-------
EXAMPLE 11-2
2-nitrophenol has been detected in the Alehandra Estuary, which has a
pH of 8, at concentrations of 20 yg/1 (total dissolved form). Determine the
volatilization flux on a per unit area basis. Assume the volatilization
rate constant, k , is 2 cm/hr.
v
From Table 11-13, the pK of 2-nitrophenol is 7.21. The fraction
present in the electrically neutral (non-ionic) form is:
a. = l
l + 10(PH-PKa)
= 1 + lo'8-0-7-2
= 0.14
From Equation 11-13 the volatilization flux is:
R = 2 cm/hr (0.14) (^^) (1220J.) (T?^) = 56 yg hr'V2
v ' m*
END OF EXAMPLE 11-2
61
-------
2.3.2 Sorption on Suspended Sediments
2.3.2.1 Introduction
Sorption refers to the accumulation of a chemical in the boundary
region of a solid-liquid interface. Sorption occurs when the net
sorbate-sorbent attraction overcomes the solute-solvent attraction, where
solute and sorbate refer to the sorbing species in solution and sorbed at
the interface, respectively.
Sorption of chemicals in the natural environment is significant because
the fates of sorbates and solutes can be significantly different. Sorbates
are transported along with sediments, and can be deposited in river or lake
beds to remain indefinitely. Sorbates are in many ways protected from
transformation processes which would otherwise affect the solute. For
example:
• Microbial degradation rates can be reduced. Steen e_t aj_. (1978)
performed tests which showed that sorption of toxicants to
suspended sediments renders some compounds unavailable for
biodegradation in the adsorbed state.
• Volatilization is diminished. Since volatilization of a
chemical occurs from the dissolved phase, the sorbate is not
directly available for volatilization. Rather, the sorbate
first desorbs before it volatilizes. Example II-4 will show
the significant influence of sorption on volatilization.
• Direct photolysis of pollutants adsorbed on suspended particles
is inhibited in some cases. Further, suspended solids deposited
on the bed of a river, lake, or estuary, receive very little
radiation for photolytic reactions.
The net interaction between the surface of a solid and sorbate can
result from a variety of forces, including coulombic attraction,
Van der Waals forces, orientation energy, induction forces, hydrogen
62
-------
bonding, and chemical forces (Reinbold €|t aK, 1979). In the case of many
organic compounds, the solute-solvent interaction is often weak so that even
a weak sorbate-sorbent attraction can result in sorption. This type of
sorption is referred to as hydrophobia sorption because of the importance of
the weak solute-solvent attraction. Hydrophobic sorption will be the topic
of much of the following discussion, but it is preceded by brief discussions
of equilibrium isotherms and sorption kinetics.
2.3.2.2 Adsorption Isotherms
Adsorption isotherms describe the relationship between the amount of
chemical sorbed and the equilibrium solution concentration. The most
commonly used isotherms are:
• Langmuir Adsorption Isotherm. This equation was originally
developed to describe adsorption of a gas to a solid surface,
but has been used to describe solid-liquid sorption.
• Freundlich Adsorption Isotherm. This empirical equation is
based on surface-free energy and monolayer capacity.
• Linear Adsorption Isotherm. This equation assumes that there
is a linear relationship between the concentrations of solute
and sorbate at equilibrium. It is valid for dilute solutions.
Figure II-4 shows example comparisons between the three isotherms, and
includes the equations which describe each isotherm. The quantity X is the
amount of sorbed chemical per mass of sediment, and C,, is the amount of
w
dissolved chemical per volume of solution. The remaining variables are
unknown parameters required to define the relationship between X and Cw.
The linear isotherm has one unknown parameter (K ), while both the
p
Freundlich and Langmuir isotherms have two unknown parameters (k ,n and m,b,
respectively).
63
-------
5000
4500
4000
g 3500
E
^5
0>
S» 3000
ra
1 2500
u
I 2000
•a
re
-=• 1500
X
1000 -
500 -,
Linear Isotherm
X = kp-Cw
Freundlich Isotherm
X = kf-Cw1/n
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Cw (ug dissolved chemical/I! solution)
FIGURE 11-4 ISOTHERMS FOR ADSORPTION OF A HYDROPHOBIC POLLUTANT
ON SEDIMENTS
64
-------
For the purposes of this document, analyses will mostly deal with
dilute aqueous solution in the range where the linear isotherm is generally
valid. This approach has the advantage of requiring that one unknown
parameter (K ) be evaluated, rather than two, and of being easier to
p
manipulate mathematically. Sectior
predicting the unknown parameter K
p
manipulate mathematically. Section 2.3.2.4 will present methods of
2.3.2.3 Kinetics of Adsorption
Sorption of organic pollutants is often treated as a process which
achieves rapid equilibrium so that expressions of kinetics are not needed.
The equilibrium approach will be used in the remaining chapters of this
document. However, a brief introduction will be given of sorption kinetics.
Studies of sorption kinetics are apparently few, with the result that
parameters required in rate expressions are ill defined and applicable only
under a specific set of conditions. Under these constraints, kinetics
expressions become less attractive unless the user can determine values of
the rate constants which apply to the specific system being investigated.
Most typically, kinetics expressions for sorption and desorption are
chosen to be first order. Specifically,
3C
W = -k ,(C —) (11-14)
3t sd w K
P
expresses the kinetic expression for the solute and
9X
= -k (X - K C ) (H-15)
3t sd p w
for the sorbate. The concentrations X and C are not necessarily
w
equilibrium concentrations. In these two equations, the rate parameter k
is assumed to be the same whether adsorption or desorption is occurring.
However, different rates could be used for each process.
65
-------
Karickhoff (1979) investigated the sorption and desorption of organic
pollutants and found that a very rapid component of adsorption preceded a
much slower component of adsorption, and that first order kinetics were
obeyed during each of the two periods. For the fast process, the time
constant was found to range from 4 to 30 per hour, while for the slow
process the time constant ranged from 0.06 to 1.5 per hour. Approximately
half of the sorptive equilibrium was realized within minutes, while the
slower component required days or weeks to complete. The slower second
period was visualized as diffusive transfer to sorption sites that were
inaccessible directly to the bulk water. Thus, equilibrium conditions are
more likely to be rapidly attained when the number of easily accessible
surface sites exceeds the amount of available sorbate, e.g. when suspended
sediment concentrations are high.
2.3.2.4 Partition Coefficients for Organic Chemicals Obeying Linear
Isotherms
The single unknown parameter, K , which relates the sorbate and solute
p
for linear isotherms is called the partition coefficient. A number of
studies have been completed which develop empirical relationships for
partition coefficients in natural sediments. Several of these studies will
be summarized here. Theoretically based methods of estimating partition
coefficients exist, such as a thermodynamic approach described in
Pavlou (1979); however, these will not be discussed here.
Karickhoff jat al. (1979) examined the sorption of aromatic hydrocarbons
and chlorinated hydrocarbons on natural sediments. They found it convenient
to relate the partition coefficient directly to organic carbon content of
the sediments as follows:
K = K x (11-16)
p oc oc
where
K = partition coefficient expressed on an organic carbon basis
oc
66
-------
x = mass fraction of organic carbon in sediment.
oc
These workers were able to expand this relationship to segregate the
influence of particle size as follows:
K = K [0.2(l-f)xs + fxf ] (H-17)
p oc oc oc
where
f = mass fraction of fine sediments (d < 50 ym)
xs = organic carbon content of coarse sediment fraction
oc
x"f = organic carbon content of fine sediment fraction.
oc
Karickhoff et al. (1979) were able to relate K to the octanol-water
oc
partition coefficient and to the water solubility by the following
relationships:
K = 0.63 K (11-18)
oc ow
where
K = octanol-water partition coefficient (concentration of
ow
chemical in octanol divided by concentration of chemical in
water, at equilibrium)
and
K = -0.54 log S + 0.44 (11-19)
oc w
where
S = water solubility of sorbate, expressed as a mole fraction.
w
67
-------
The water solubilities of the compounds examined ranged from 1 ppb to
1000 ppm.
Hassett et al. (1980) found a similar relationship between K and K
oc ow
for organic energy-related pollutants. Figure II-5 shows the relationship.
Data from Karickhoff et al. are included in the plot for comparison.
Prior to the work of Karickhoff jet _aK, Chiou et jil_. (1977)
investigated the relationship between octanol-water partitioning and aqueous
solubilities for a wide variety of chemicals including aliphatic and
aromatic hydrocarbons, aromatic acids, organochlorine and organophosphate
pesticides, and polychlorinated biphenyls. Their results, shown in
Figure II-6, cover more than eight orders of magnitude in solubility
and six orders of magnitude in the octanol-water partition coefficient. The
regression equation based on this figure is:
(11-20)
log K = 5.00 - 0.670 log Sw
ow w
where
S = solubility, in ymol/1
w
Brown and Flagg (1981) have extended the work of Karickhoff _et aj_. by
developing an empirical relationship between K and K for nine
ow oc
chloro-s-triazine and dinitroaniline compounds. They plotted their results,
along with those of Karickhoff ^t a]^, as shown in Figure II-7. The combined
data set produces the following correlation:
log K = 0.937 log K - 0.006 (H-21)
oc ow
The linear correlation between K and K for the compounds studied by
?c ow
or of uncertainty than those studied by
Karickhoff e£ aj_.
The previous paragraphs have shown how the partition coefficient K can
be predicted for organic hydrophobic compounds which obey a linear isotherm
relationship. First, K is predicted based on either water solubility or
oc
68
-------
en
o
7
6
5
4
3
2
I
0
o
7
FIGURE 11-5 RELATIONSHIP BETWEEN KQC AND OCTANOL-WATER PARTITION
COEFFICIENT (1C..) OF ENERGY-RELATED ORGANIC POLLUTANTS
uw
REFERENCE: HASSETT EI AL, (1980)
-------
10
10
10
10
I IO 10 10 10
Solubility in Woter (/xmoles/l)
10
10'
FIGURE 11-6 CORRELATION OF AQUEOUS SOLUBILITY
WITH OCTANOL-WATER PARTITION
COEFFICIENT
REFERENCE: CHIOU EJ_ AL, (1977)
70
-------
BROWN AND FLAGG (1981)
KARICKHOFF ET AL, (1979)
_L
0 050 100 150 200 250 300 350 400 450 500 550 600 650
LOG Kow
Note: The actual error bands for this figure are probably
greater than indicated here due to error in the
measurement of K
ow
FIGURE 11-7 RELATIONSHIP BETWEEN Knr AND 1C.. FOR
UL (JW
COARSE SILT
71
-------
the octanol-water partition coefficient. Tables II-5 through II-9 shown
earlier contain K values for a number of compounds. Then based on an
ow
estimate of organic carbon fraction in the fine and coarse sediments, K can
p
be estimated from Equation 11-17.
2.3.2.5 Solute and Sorbate Fractions
The relative amount of pollutant sorbed and dissolved depends on both
the suspended sediment concentration and the partition coefficient, and at
equilibrium is given by:
Cw 1
°w ~ T.' = TTiTs" . (n-22)
where P
C = total dissolved phase concentration
w
C = C + C
T w S
CS = XS
K = partition coefficient
P
S = suspended sediment concentration, on a part per part basis
X = mass of sorbed pollutant per mass of suspended sediment.
Equation 11-22 can be illustrated more vividly by tabulating ranges of K
p
and S values. Table 11-14 shows this information. Partition coefficients
o 4
and suspended sediment concentrations range from 10 to 10 . For the lowest
value of the partition coefficient nearly all of the pollutant is present in
the dissolved form, regardless of the suspended sediment concentration.
Also, for low suspended sediment concentrations, nearly all of the pollutant
is dissolved, unless the partition coefficient is exiremely large. When
relatively high partition coefficients and sediment concentrations occur
simultaneously, then most of the pollutant present exists as sorbate. For
72
-------
TABLE 11-14
RELATIONSHIP OF DISSOLVED AND SORBED PHASE POLLUTANT
CONCENTRATIONS TO PARTITION COEFFICIENT AND
SEDIMENT CONCENTRATION
K
P
10°
101
102
103
104
S (ppm)
1
10
100
1000
10000
1
10
100
1000
10000
1
10
100
1000
10000
1
10
100
1000
10000
1
10
100
1000
10000
Cw/CT .
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.9
1.0
1.0
1.0
0.9
0.5
1.0
1.0
0.9
0.5
0.1
1.0
0.9
0.5
0.1
0.0
V
100.
100.
100.
100.
99.
100.
100.
99.9
99.0
90.9
100.
99.9
99.0
90.9
50.
99.9
99.0
90.9
50.
9.1
99.0
90.9
50.
9.1
1.0
If CT = 100 ppb
X =
100.
100.
100.
100.
99.
IxlO3
IxlO3
999.
990.
909.
IxlO4
IxlO4
9.9xl03
9. IxlO3
5xl03
IxlO5
9.9xl04
9. IxlO4
5xl04
9xl03
9.9xl05
5
9.1x10
5xl05
9. IxlO4
9.9xl03
CS =
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.1
1.0
9.1
0.0
0.1
1.0
9.1
50.
0.1
1.0
9.1
50.
90.9
1.0
9.1
50.
90.9
99.0
73
-------
all the cases shown, X is high which indicates that the mass sorbed per unit
mass of sediment present can be high while C is simultaneously low.
EXAMPLE 11-3
Determine the fraction of benzo(a)pyrene that is dissolved in a system
containing 300 ppm suspended solids. The suspended sediments are 70 percent
fines (d <50 ym) and the weight fraction of organic carbon is 10 percent of
the fines and 5 percent of the sand fraction.
From Table II-9, the solubility of benzo(a)pyrene is 0.0038 mg/1, and
6
the octanol-water partition coefficient is 10 . If, for the moment, the
octanol-water partition coefficient is ignored, Equation 11-20 can be used
to predict K based on solubility. The solubility of 0.0038 mg/1 must be
ow
converted toy mole/1:
- (0.0038 mg/1) (ICT g/.g) () (106
= 0.015 ymole/1
From Equation 11-20, the predicted octanol-water partition coefficient is:
log K = 5.00 - 0.670 log (.015) = 6.22
ow
6.22 6
so K =10 , which is acceptably close to the tabulated value of 10 .
ow
Using the tabulated K , , K is computed from Equation 11-18:
ow oc
6
K = 0.63 10 = 630,000
oc
From Equation 11-17, the partition coefficient becomes:
K = 630,000 [0.2 (1-.7) (.05) + 0.7 (.10)]
P = 46,000
74
-------
The suspended sediment concentration for the system is 300 ppm, or
-6
300-10 parts per part. Using Equation 11-22, the fraction of
benzo(a)pyrene which is dissolved is:
C.
w
CT 1 + 300 • 10"6-46,000
= 0.067 or about 7 percent
Consequently, most of the benzo(a)pyrene is present as sorbate.
END OF EXAMPLE 11-3
75
-------
2.4 TRANSPORT PROCESSES
2.4.1 Solubility Limits
The concentration of a compound in a natural water, and therefore the
rate of transport by that water, can be limited by its equilibrium
solubility. The aqueous solubility of organic compounds ranges widely:
o
Aqueous Solubility at 25 C
Compound (mass which will dissolve in 1 liter of water)
Sucrose 2,000,000 milligrams
Benzene 2,000 milligrams
Toxaphene 2 milligrams
Chrysene 0.002 milligrams
Non-polar compounds have limited solubilities in polar solvents such as
water. The solubility of toxic organic compounds is generally much lower
than for inorganic salts. Equilibrium solubilities for toxic organic
compounds are given in Tables II-5 through II-9. Solubility increases with
temperature for most organic compounds, typically by a factor of about 3
from 0 C to 30 C.
Organics are generally less soluble in sea water than in fresh water as
can be seen in the tabulations below (Rossi and Thomas, 1981):
Solubility at 25 C
Compound Distilled Hater Sea Water
Toluene
Acenapthene
Pyrene
507 yg/1
2.41 yg/1
0.13 yg/1
419 yg/1
1.84 yg/1
0.09 yg/1
In the absence of colloids or micelles, the maximum amount of a toxic
organic substance which can be held in the water column under equilibrium
conditions is just the aqueous equilibrium solubility S , plus the
equilibrium amount of solute sorbed on suspended matter:
76
-------
C = 5 + f (S ) (11-23)
T w s w
where C = total amount of compound which can be held in a
natural water at equilibrium conditions, jjg liter
-1
S = equilibrium aqueous solubility, yg liter
w
f (S ) = equilibrium amount of sorbate on suspended matter; a
S W
function of S . f is the sorption isotherm function.
w s
If a linear sorption isotherm is used, as is commonly the case for trace
constituents (see Section 2.3.2), the above expression becomes:
C
-------
2.4.2 Volatilization
2.4.2.1 Introduction
Volatilization is defined as the transfer of matter from the dissolved
to the gaseous phase. A considerable number of toxic substances volatilize
in the natural environment. Volatilization rates depend on the properties
of the toxicant and on the characteristics of the water body. If a toxicant
is "highly volatile", then obviously volatilization is an important process
affecting the fate of the toxicant. However, even for toxicants which are
considerably less volatile, volatilization cannot always be ignored. This
is because the fate of a toxicant is governed by a variety of processes. If
volatilization proceeds as fast as other competing mechanisims, even though
all the rates might be slow, then volatilization will influence the fate of
the toxicant.
Methods will be provided in this section to predict the volatilization
rate for toxic organic substances, which volatilize according to the
following relationship:
3C ~kv JP_ k- (C - -P- ) (II
3t Z K., v K>i|
where
C = concentration of toxicant in dissolved phase (concentration
of solute)
k = volatilization rate constant in units of length/time
v
k" = volatilization rate constant in mixed water body in units of
v -1
time
Z = mixed depth of water body
P = partial pressure of toxicant in atmosphere above the
waterbody being investigated.
78
-------
K = Henry's Law constant
H
For many applications the partial pressure of the compound in the atmosphere
is zero, so that Equation 11-25 simplifies to:
3C *.' r
It = ~ v (H-26)
An alternate form of Equation 11-26 is in terms of the total pollutant
concentration, C , and the cite specific volatilization rate constant, k :
3CT
T _ k p (11-27)
3T " "kvm CT ( '
where
a = fraction of toxicant present in dissolved phase
w
The following sections will illustrate how to predict the volatilization
rate for toxicants of either low or high volatility. But first, a brief
discussion of Henry's Law is required.
2.4.2.2 Henry's Law
Henry's Law is an expression which relates the concentration of a
chemical dissolved in the aqueous phase to the concentration (or pressure)
of the chemical in the gaseous phase when the two phases are at equilibrium
with each other. One method of commonly expressing Henry's Law is:
P = K C (11-29)
H w
where
P = equilibrium partial pressure of pollutant in atmosphere above
the water, atm
79
-------
C = equilibrium concentration of pollutant in the water, mole/m
w
3
K = Henry's Law constant, atm m /mole
H
Henry's Law in this form is valid for pollutants present in concentrations
up to 0.02 expressed as a mole fraction. For compounds with molecular
weights greater than 50 g/mole, a mole fraction of 0.02 represents a
concentration of at least 55,000 mg/1. Typically toxic pollutant levels
in the environment are present at levels far below this concentration.
Table 11-15 contains values of Henry's Law constants for a number of
selected hydrocarbons. In the table, Henry's Law constant is expressed in
units of atm m /mole. However, in the literature Henry's Law constant can
be defined in numerous ways. A second, widely used method of defining
Henry's Law constant is:
(11-30)
w
where
C = molar concentration in air, mole/m
a
K = alternate form of Henry's Law constant, dimensionless
H
Equations 11-29 and 11-30 are related as follows:
V
8.2 x
where
T = temperature of water, K.
This relationship is based on the ideal gas law. Equation 11-31 is useful
because of the frequent necessity to convert literature data from one set of
units to another.
80
-------
TABLE H-15
HENRY'S LAW CONSTANT FOR SELECTED HYDROCARBONS
Oleflns
and
Acetylenes
Ethene (g)
Propene (g)
1-Butene (9)
1-Pentene (I)
1-Hexene (I)
2-Heptene (t)
1-Octene U)
Propyne (?)
1-Butyne (g)
Cycloalkanes
Branched-Chain
Alkanes
Cycl opentane (i.)
Cyclohexane (i)
Methyl cycl opentane (i)
Kethylcyclohexane (t)
Propyl cycl opentane (£)
Isocutane (y)
Isopentane (i)
2-Methylpentane (t)
2-Kethylhexane (i)
2,2-Dimethylpentane (t)
3-*ethyl heptane (i)
2,2,4-Trimethylpentane (I)
4-Metnyloctane (i)
Polychlorinated
Biphenyls
Aroclor 1242
Aroclor J248
Aroclor 1254
Aroclor 1260
KH
Henrys Law
Constant
(atm-m-'/mole
0.214
0.232
0.268
0.398
0.412
0.418
0.905
0.0110
0.0194
0.187
0.196
0.362
0.428
0.893
1.24
1.364
1.73
3.42
3.15
3.71
3.04
9.936
5.7x10-'
3.5 xlO'1
2.8x10-'
7.1 x 10-'
*v (cm/hr)a
20.
20.
20.
20.
20.
20.
20.
19.8
ZOi
20.
20.
20.
20.
20.
20.
20.
20.
20.
20.
20.
20.
20.
15.6
18.9
18.9
19.6
Aromatics
Benzene (t)
Toluene (1)
Ethyl benzene (l)
o-Xylene U)
I sopropyl benzene (1)
Naphthalene (s)
Biphenyl (s)
Acenaphthene (s)
Fluorene (s)
Anthracene (s)
Phenanthrene (s)
n-Alkane
Methane (9)a
Ethane (g)
Propane (g)
n-Butane (g)
n-Pentane (i)
n-Hexane (t)
n-Heptane (i)
n-Octane (t)
n-Nonane (i)
Decane (t)
Dodecane (£)
Tetradecane (i)
Pesticides
DDT
Lindane
Dieldrin
Aldrin
Endrin
Heptachlor
Chlordane
Toxaphene
h
Henry's Law
Constant
(atm-mVmole
5.49x10-'
6.66x10-'
8.73x10''
5.27x10-'
1.45xlO-2
4.25X10-11
6. 36x10" *
2.28x10"*
2.35x10—
1.65x10"'
1.48x10-'
0.665
0.499
0.707
0.947
1.26
1.85
2.07
3.22
3.29
4.93
7.12
1.14
3.9xlO-5
4.9x10-'
2.0x10"'
1.4xlO-5
4.6x10-'
1.5x10-'
5x10' 5
0.1
kv 'cm/hr)a
19.4
19.5
19.6
19.4
19.8
14.5
16.0
11.7
11.9
18.2
9.6
20.
20.
20.
20.
20.
20.
20.
20.
20.
20.
20.
20.
20.
3.9
0.06
0.02
0.60
0.06
18.
4.8
19.8
These are estimated values based on k, = 20 cm/hr and k = 3000 cm/hr.
81
-------
Henry's Law constant can be estimated for slightly soluble compounds
(mole fraction ^0.02) by the following expression:
Ps X MW (11-32)
K (atr,ma/mole) = ____ (II 32)
where
P = saturation vapor pressure of pure compound in Torr
s
MW = molecular weight
S = solubility in water in ppm
w
Tables II-5 through II-9 presented vapor pressure and solubility data for
the organic priority pollutants, which can be used to predict Henry's Law
constant. Although Equation 11-32 is not valid for highly soluble
chemicals, generally the toxicants of interest here are only slightly
soluble, so that the expression is adequate. The dimensionless form of
Henry's Law constant is expressable as:
16.04 P x MW
' (11-33)
H Sw x T
where all variables have been previously defined.
EXAMPLE 11-4
Henry's Law Constant for Chloroform
Calculate Henry's Law constant in the two forms expressed by Equations
11-32 and 11-33. Chloroform (also called trichloromethane or CHC1 ) has the
O
following properties:
• vapor pressure = 150 Torr (from Table II-5)
• solubility = 8200 ppm at 20 C (from Table II-5)
82
-------
0 Molecular weight = 12 (carbon)
1 (hydrogen)
3 x 35.5 (chlorine)
Sum = 119
From Equation 11-32,
150x119 -3
K = = 2.86 x 10 atm-m3/mole
H 760x8200
From Equation 11-33, at 20 C (293 K):
16.04x150x119
KM = = 0.12
H 8200x293
Henry's constant, expressed as K", had been found experimentally to be 0.12,
the same as predicted here.
END OF EXAMPLE II-4
2.4.2.3 Two Film Theory of Volatilization
When a chemical volatilizes from water, the process can be visualized
as a mass transfer occurring over several distinct steps. Figure II-8
presents a schematic representation of the process. The concentration of
the chemical is C in the bulk liquid solution. As the chemical moves upward
in the bulk solution it moves through a thin "liquid film" where a
concentration gradient develops because the transfer rate is limited by diffusion.
The dissolved chemical then volatilizes and passes through a thin "gas
film", where again transfer may be limited, before reaching the bulk vapor
phase.
At the interface between the gas and liquid films the concentration in
the liquid (C ) and in the gas (P , expressed as partial pressure) are
i ci
assumed to be in equilibrium and to obey Henry's Law:
83
-------
Toxicant concentration
Vapor phase
Direction of movement
FIGURE II-3
SCHEMATIC REPRESENTATION OF VOLATILIZATION
FROM SOLUTION PHASE TO LIQUID PHASE
84
-------
P . = K C (11-34)
ci H i '
In the absence of net accumulation at the interface the mass flux from one
phase must equal the mass flux from the other, or
Fz=f (PC-PC1) =kl. (C-C.) (11-3
where
F = flux of chemical in z direction
z
k = mass transfer coefficient in the qas phase across "gas film"
9i
k = mass transfer coefficient in the liquid phase across "liquid
film"
and P , P ., C, C. are defined in Figure II-8. Since it is not convenient
L. C I 1
to measure the partial pressure and concentration at the interface, it is
worthwhile to develop expressions for bulk transfer coefficients, given by:
F = 1_V£ (p . p') - i, (r ci (11-36)
z RUT l c PcJ kvl (C - V
where
k = overall volatilization rate defined for the gaseous phase
k = overall volatilization rate defined for the liquid phase
S = saturation concentration of chemical in equilibrium with P
P c
P" = partial pressure in equilibrium with C
c
Combining Henry's Law equilibrium expressions with Equations 11-35 and 11-36
the overall volatilization rates become:
85
-------
k R T k,. k .
vg u IT gi
(11-37)
and
KHkgi
(11-38)
Of the two expressions, normally Equation 11-38 is more useful for the
purposes of this document because the pollutants being analyzed are in the
aqueous phase. To simplify terminology Equation 11-38 will be rewritten as:
or:
1 i R T
1 _ 1 u
kv kl KHkg
kv kl KH kg
(II-39a)
(II-39b)
where the second subscripts to each variable have been dropped. The
volatilization rate, k , is the same as shown earlier in Equation 11-25 and
v
depends on k , K , and k .
g H 1
There are two special cases of Equation 11-39, depending on the value
of Henry's Law constant. They are:
k,, for large 1C (liquid-phase limited)
_ H g
, for small 1C (gas-phase limited)
(II-40a)
(II-40b)
To make Equation 11-40 usable, "large" and "small" values of K" have to be
H
defined. For cases when the liquid phase is limiting the transfer rate, a
large fraction, R, of the total resistance exists in the liquid phase, or:
(11-41)
kl KHkg,
86
-------
Similarly when the gas phase is limiting:
1 / 1 \ /I 1
= R T- = R T-
(11-42)
Equations 11-41 and 11-42 can be rearranged to express Henry's Law constant
explicity:
k R
_L , for liquid-phase limited (II-43a)
kg 1-R
_L ilE. , for gas-phase limited (II-43b)
kg R
At this point values for R, k , and k must be specified. "Typical" values
of k and k for surface waters are in the range of 20 cm/hr and 3,000
cm/hr, respectively. For R values of 0.83, 0.90, and 0.95, the phase
limiting values of Henry's Law constants, converted to units of atm-ru /rcole
using Equation 11-31, are as follows:
Henry's Constant (atm-m3/mole)
J^ L_iquid-phase Limited Gas-phase Limited
0.83 7.8 x 10"" 3.3 x 10"5
0.90 1.4 x 10~3 1.8 x 10'5
0.95 3 x ID'3 8.4 x 10~6
-3 3
Hence, for Henry's Law constants larger than about 1.0 x 10 atm m /mole
most of the resistance to volatilization lies in the liquid phase, and for
-5 3
Henry's Law constants less than about 1.0 x 10 atm m /mole, most of the
resistance lies in the gas phase. When either of the two phases controls
the volatilization rate, then the simplified Equation 11-40 can be used in
lieu of Equation 11-39. The data in the tables presented earlier can be
used to predict Henry's Law constant and then to decide whether the gas or
liquid phase limits volatilization.
Based on the two-film model there are two methods which can be used to
estimate volatilization rates. One approach is considerably more simple
than the other. The simpler approach is based on the following reasoning.
Using "typical" values of k and k , k can be estimated based solely on K
1 g v H
87
-------
as the independent variable, where K is allowed to vary over its potential
H
range of values. As Table 11-16 shows, K can vary by at least seven orders
H
of magnitude. Based on this variability of Henry's Law constant, Table
11-17 presents the associated volatilization rates. As Henry's Law constant
increases, the volatilization rate approaches 20 cm/hr, the liquid phase
limiting rate. As Henry's Law constant decreases, so does the
volatilization rate, with the lower limit being zero.
The second method of predicting k is based on finding k and k
v g 1
individually, rather than assuming typical values. The gas-pnase transfer
rate can be found based on the evaporation rate of water as outlined in
Mills (1981). Mills showed that:
kg-700V
where
k ' = gas transfer rate for water vapor, cm/hr
V = wind speed, m/sec
This expression was derived from an empirical relationship shown in Linsley
et jil_., (1979) for the evaporation of water. Liss (1973) conducted
measurements in an experimental basin and found that:
k^ = 1000 V (n
where the units are the same in Equation 11-44. Considering that the
approaches used to develop Equations 11-44 and 11-45 are different, their
agreement is good.
The values of kg and kg are related by penetration theory (Bird e_t al_.,
1960) as follows:
-------
TABLE 11-16
HENRY'S LAW CONSTANTS FOR SELECTED COMPOUNDS
Compound
Henry's Law Constant (atm-mVmole)
Vinyl Chloride
Carbon Tetrachloride
Toluene
Aroclor 1254
Flourene
DDT
Dieldrin
3.7
2 x ID'2
6.7 x ID'3
2.8 x 10'3
2.4 x 10-"
3.9 x lO'5
2.0 x 10'7
TABLE 11-17
TYPICAL VALUES OF POLLUTANT VOLATILIZATION RATES
IN SURFACE WATERS
Ku(atm-m3/mole)
n
10°
lo-1
io-2
ID'3
10-"
io-5
io-6
io-7
1C (dimensionless)
H
41.6
4.2
4.2X10'1
4.2xlO-2
4.2xlO-3
4.2x10'"
4.2xlO-5
4.2xlO-6
k (cm/hr)*
20.
20.
19.7
17.3
7.7
1.2
0.1
0.01
using k = 3000 cm/hr
k, = 20 cm/hr
89
-------
where
D = diffusion coefficient of pollutant in air
a
D = diffusion coefficient of water vapor in air
wv
Diffusion coefficient data can be found in such references as Perry and
Chilton (1973), or estimated using the Wilke-Chang method, also in Perry and
Chilton. If an analytical method is used to estimate diffusion coefficients,
note that it is easier to predict the ratio of two diffusion coefficients
than to predict each coefficient individually because some of the required
information cancels out of the ratio, and consequently is not needed at all.
In many cases it is acceptable to approximate the ratio of diffusion
coefficients as follows:
wv
_Da (laV* (H-47)
D
where
MW = molecular weight of pollutant
Table 11-18 illustrates the difference between calculating the diffusion
coefficient ratio by using tabulated data from Perry and Chilton and by
using Equation 11-47. The percent differences between the ratios range
from 1 to 27 percent and average 15 percent. This agreement is acceptable for
screening purposes. Combining Equations 11-46, 11-44, and 11-47, the final
expression for k (in units of cm/hr) is:
9
(
^
MW
This expression is valid for rivers, lakes, and estuaries.
The liquid phase transfer coefficient k can be predicted based on the
reaeration rate, k , for the system. The relationship proposed by Smith
et al. (1981) is;9
90
-------
TABLE 11-18
COMPARISON OF TABULATED AND PREDICTED VALUES OF DIFFUSION
COEFFICIENTS FOR SELECTED POLLUTANTS
Pollutant
Chlorobenzene
Toluene
Chloroform
Naphthalene
Anthracene
Benzene
Molecular
Weight
113
92
119
128
178
78
Diffusion
Perry & Chil
(cm2/sec)
0.075
0.076
0.091
0.051
0.042
0.077
Coefficient
ton Predicted
(cm2/sec)
0.088
0.097
0.086
0.083
0.070
0.106
Perry & Chvlton
/DV«
w
.58
.59
.64
.48
.44
.59
Predicted:
l°\h
w
.63
.66
.63
.61
.56
.69
Percent
Difference
9
12
1
27
27
17
-------
'D \"
W
where
\ 2
k' , 0.5 < n < 1 (11-49)
a ~~ ~~
D = diffusion coefficient of pollutant in water
w
Dg = diffusion coefficient of dissolved oxygen in water
k" = surface transfer rate of dissolved oxygen, expressed in the
a
same units as k .
In other chapters of this report, the reaeration rate is presented as k ,
a
defined as:
k = k/Z ("
a a
where
Z = mixed depth of water body
For rivers the mixed depth is the total depth, while for estuaries the mixed
depth is the total depth only if the estuary is well mixed. Otherwise, it
is the depth to the pycnocline. Similarly for lakes, the mixed depth can be
less than the total depth, and can be chosen to be the depth of the
epilimnion.
The exponent n varies as a function of the theorical approach used to
develop Equation 11-49. If film theory is used, i.e. the film is considered
to be a laminar sublayer, then n = 1. If penetration or surface renewal
theory is used, n = 0.5. Using experimental approaches, researchers have
found n to vary from 0.5 to 1.0. Since the movement of water in natural
water bodies is generally turbulent, the parameter n can be chosen to be
0.5.
92
-------
Perry and Chilton (1973) provide data and methods to predict the
diffusion coefficient of a pollutant in water. The Othner-Thakor
relationship, described in Smith jrt al. (1981) can also be used. As an
approximate approach, by using the square root of the molecular weights the
following expression results:
(H-51)
A recent study (Rathbun and Tai, 1981) used a tracer technique to
predict the volatilization rates of four priority pollutants from 12
different rivers. That study provides an opportunity to compare, even if
only to a limited degree, some of the methods presented here against field
results. Table 11-19 briefly summarizes the results of Rathbun and Tai
(1981). As shown by the values of Henry's Law constant for the four
pollutants, each pollutant is liquid phase limited, since all Henry's Law
constants exceed 1.0 x 10 atm m /mole. The study results were unable to
predict differences in volatilization rates for the four pollutants, and
found that the best predictive expression was:
k = 0.655 k'
V a
Based on Equation 11-51 the screening methods predict:
k = 0.7kT to 0.8 k'
v ci a
where the range reflects the variability in molecular weight among the four
pollutants.
If the default value of 20 cm/hr, suggested earlier in this section
were used as a rough estimate of the volatilization rate for liquid phase
limited pollutants, this value would fall within the observed range of 1.5
to 24 cm/hr. It appears that the screening methods presented here generate
acceptable estimates of volatilization rates.
93
-------
TABLE H-19
RESULTS OF A STUDY TO DIRECTLY DETERMINE VOLATILIZATION RATES OF
SEVERAL PRIORITY POLLUTANTS IN RIVERS3
Henry's Constant
Pollutant (atm-rnVmole) Molecular Weight
Benzene 5.5 x 10"3 H9
Chloroform 2.9 x 10"3 78
Methylene Chloride 2.7 x 10"3 85
Toluene 6.7 x 10"3 92
Study results showed: k = 0.655 k"
Range of values for 12 rivers: 1.5 to 24 cm/hr
Screening method predicts: k = 0.7 k"L to 0.8 k_"
—. v a a
aRathbun, R.E. and D.Y. Tai. 1981. Techniques for Determining
the Volatilization Coefficients of Priority Pollutants in Streams
Water Research, Volume 15, pp. 243-250.
94
-------
2.4.2.4 Volatilization Half-Life
Numerous researchers have in the past calculated the volatilization
half-life of toxicants under controlled laboratory conditions. The result
of some of this work was shown earlier in Tables II-5 through II-9.
Typically researchers have used the following expression to calculate the
half-life:
0.693Z
where
t = half-life (time required for the concentration of the
}|
contaminant to decrease by half)
It is important to understand that the volatilization half-life of a
toxicant varies according to the environmental conditions. Under controlled
laboratory conditions, where the depth of water is extremely small, t can
%
be extremely small. If the water depth increases by 100 fold, for example,
so does t .
h
The volatilization half-life is affected by suspended solids in the
system. When suspended solids are present, Equation 11-52 should be
modified to:
- (1 + SKJ (H-53)
% KM $
where
S = suspended solids concentration
K = partition coefficient
P
The partition coefficient is the ratio of the sorbed pollutant concentration
to the dissolved phase concentration. A method to predict K was discussed
n
earlier in Section 2.3.2. Since the toxicant which sorbs to the sediments
is not directly available for volatilization, the total flux of volatilizing
95
-------
particles decreases. The following example illustrates how sorption can
influence the half-life.
EXAMPLE I1-5
The following data for hexachlorobenzene were obtained from Table 11-8:
• solubility = 20 yg/1
-5 o
• vapor pressure = 10 Torr at 20 C
6
• K = 10
ow
Under the conditions reported in the work of Mackay and Leinonen (1975)
L = 1 m
k = 8 cm/hr = 8 x 10~2 m/hr
Hence
0.693 x 1
t, = 2 = 8.7 hours
h 8 x ID'2
Note that the half-life is small even though the vapor pressure is only 10
Torr. The results indicate that the vapor pressure is, by itself, not
necessarily a good indicator of the importance of volatilization.
Now, consider the following conditions which might be encountered in a
r i ver:
• k (reaeration rate) = 0.5/day
a
• suspended sediment concentration = 550 ppm
96
-------
4
0 K = 5 x 10
P
• depth = 1 m
The expression of volatilization half-life modified to account for the
presence of the suspended solids is:
0.693 Z , .
From Equation 11-51, the liquid-phase transfer rate for hexachlorobenzene
is:
\
<:ob /
x 0.5 x 1 = 0.29 m/day = 0.01 m/hr = 1 cm/hr
Henry's Law constant can be estimated based on Equation 11-32.
Using the data presented earlier
K.. =
x 285 = 1.9 x 10"" atm-mVmole = 7.8 • 10"3, dimensionless
H 760 x .02
Using a default value of 3,000 cm/hr for k , the volatilization rate is:
9
So
k =1 cm/hr
The half-life becomes
V
A comparison of half-lives shows that:
• t = 8.7 hours under laboratory conditions
97
-------
• t = 75 days under instream conditions
This example illustrates that half-lives are not always extrapolatable from
one type of system to another due to the combined difference in sorption
effects and volatilization rates.
END OF EXAMPLE 11-5
2.4.2.5 Flux of Volatilizing Pollutants
The preceding sections have provided techniques for predicting
volatilization rates of pollutants. Obviously, if the volatilization rate
of one pollutant exceeds that of a second pollutant, then the first
pollutant is more volatile than the second. However, this criterion alone
does not determine whether volatilization is important in a specific
situation. The volatilization flux is the rate at which mass is transferred
to the gaseous phase from the liquid phase and is given by the following
expression:
Flux = k C - — (H-54)
KH
= kvC, when P = 0 (11-55)
where
C = concentration of pollutant in water as solute
P = partial pressure of pollutant in atmosphere.
Hence both the volatilization rate and the dissolved phase concentration
have to be considered jointly to predict the flux being volatilized. Table
11-20 illustrates these principals for several chemicals. The
volatilization rates for these pollutants, range from a high of 20 cm/hr for
carbon tetrachloride to a low of 0.02 cm/hr for dieldrin. Anthracene has a
volatilization rate constant of 18 cm/hr, 90 percent as high as the volatile
carbon tetrachloride. However, the solubility of anthracene in water is
much lower (0.06 ppm versus 785 ppm). Hence if each of these two chemicals
98
-------
TABLE 11-20
RELATIVE VOLATILIZATION MASS FLUXES OF SEVERAL CHEMICALS IN SATURATED SOLUTIONS
Chemical
Henry's Law Constant
(atm-niVmole)
Volatilization Rate Solubility
Constant (cm/hr) (ppm)
K
ow
Flux Ratio'
Carbon
DDT
Dieldri
Tetrachloride
n
Phenanthrene
2
3
2
1
.3
.9
.0
.5
x
x
X
X
io-2
io-5
10-"
10- 3
20
3
0
9
.
.9
.02
.6
785
.002-. 085
0.2
1.0
400
10"-106
-
29,000
5x10
4
2
1
"-2xl06
x 10b
x IO3
This is the ratio of volatilization flux of a saturated solution of carbon tetrachloride to the
volatilization of the specified chemical.
-------
were to volatilize from saturated solutions, the flux of carbon
tetrachloride would be 15,000 times as great. The same type of comparison
can be made for DDT and carbon tetrachloride. The volatilization rate
constant for DDT is relatively high (about 20 percent that of carbon
tetrachloride), but the solubility is so low that the ratio of
volatilization flux would be about 100,000:1.
These comparisons have not considered the relative differences in
sorption characteristics of the pollutants. Since only the solute
volatilizes, the volatilization flux of a pollutant which is mostly sorbed
to suspended material is lower than in the absence of suspended material,
all other factors remaining the same. Tables 11-5 through 11-9 show the
octanol-water partition coefficient, which provides a measure of relative
importance of sorption for the four pollutants. Because both DDT and
anthracene have higher octanol-water partition coefficients than does carbon
tetrachloride, the ratio of volatilization of mass fluxes is likely to be
even greater than calculated above for natural systems containing suspended
material.
100
-------
2.5 TRANSFORMATION PROCESSES
2.5.1 Biodegradation
2.5.1.1 Introduction
Microorganisms are ubiquitous in the aquatic environment.
Microbes are also very active chemically due to their ability to supply
energy for reactions through normal metabolic processes and to catalyze
reactions through enzymatic activity. Chemical reactions which proceed very
slowly or not at all in the absence of biota occur at rates up to eleven
orders of magnitude faster in the presence of biological enzymes. Some of
the reactions catalyzed by microorganisms transform or degrade organic
pollutants. Frequently, microbial degradation, or biodegradation, is the
most important, if not the only process which can decompose an organic
pollutant in the aquatic environment.
Although microbial communities catalyze countless reactions, many of
them fall into a few classes of important reactions. Oxidative reactions
make up one very important class of biochemical reactions. The
hydroxylation of aromatic compounds, such as benzene, is an example of an
oxidative reaction which generates polar compounds from non-polar ones:
OH
Enzyme
Benzene
Catechol
An extremely important oxidative reaction unique to microbial organisms is
aromatic ring fission:
CCOO"
COO"
Enzyme
Enzyme
Catechol
101
-------
Microbes also catalyze reductive reactions. A notorious example is the
dehydrochlorination of DDT to produce DDE:
ci
DDT DDE
Enzymes can catalyze otherwise slow hydrolytic reactions as well:
s s
II u
(CH30)2— P— S— CHCOOC2H5 - ^^^ - *- (CH.^ - P — S — CHCOOC^ + C^OH
CHCOOC2Hg CH2COOH
MalatMon Malathion S-monoacid
The term "biodegradation" encompasses these and other biologically
mediated processes which chemically alter a pollutant. Although each
reaction causes the disappearance or primary degradation of a compound,
different reactions affect the toxicity of a compound in markedly different
ways (Alexander, 1980).
"Mineralization" refers to the complete degradation of an organic
compound to inorganic products:
Toxic Organic Compound — *- CO + Inorganic Products (e.g. NOZ,
2 ^
P04~> SO*')
In many reactions, however, organic products remain.
"Detoxication" reactions produce innocuous metabolities from a toxic
substance:
Toxic Organic Compound —*- Innocuous Compounds
102
-------
In "activation" reactions microbes convert an innocuous compound into a
toxic compound:
Innocuous Compound » Toxic Compound
The "defusing" of potentially hazardous compounds occurs when biota produce
an innocuous compound before the parent compound's harmful form is
generated.
Potentially Toxic Compound » Innocuous Compound
Finally, a toxic compound may be transformed chemically but still retain its
toxicity. Figure II-9 illustrates some of the above types of reactions as
they occur among the phenoxy herbicides.
Because of the wide variety of toxicological effects metabolic
transformations may have, evaluating the impact of a compound on the
environment requires a knowledge of the potential products which form.
However, for the purposes of estimating the concentration of a pollutant in
a natural water body, the user may simply consider biodegradation to be a
decay process. Methods of estimating the rates of biodegradation constitute
the subject matter of the remainder of this section.
2.5.1.2 Rates of Biodegradation in the Environment
The rate at which a compound biodegrades in the aquatic environment depends
on its role in microbial metabolism. Some organic pollutants serve as
food sources which provide energy and carbon for growth and cell maintenance
when metabolized by a microorganism. In other cases, microorganisms
transform the pollutant, but are unable to derive energy for growth from the
reaction. These two metabolic patterns, growth metabolism and cometabolism,
exhibit distinct characteristics and rates of degradation. Because of the
important differences between these two types of biodegradation, they are
treated separately in the following discussion.
103
-------
MICROSIAL TRANSFORMATIONS OF TOXIC CHEMICALS
(Potential Toxin)
O(CH2)3COOH
(Less Toxic Substance)
OH
Cl
OCH2CH2OSO3H
r-CI
Cl
(Potential Toxin)
Cl
Cl
(Toxin)
Source: Alexander (1980)
FIGURE 11-9 MICROBIAL TRANSFORMATIONS OF PHENOXY HERBICIDES
104
-------
2.5.1.2.1 Metabolism of Growth Substances.
Heterotrophic bacteria degrade certain organic compounds to provide the
energy and carbon required for their growth. Many toxic substances function
as growth substrates for bacteria in a manner similar to naturally occurring
organic compounds. These growth substrates are identifiable by their
ability to serve as the sole carbon source for a bacterial culture. The
metabolic transformation of these growth substrates generally results in
relatively complete degradation or mineralization, thus detoxifying toxic
growth substrates. The detoxifying effect and relatively rapid rates of
growth metabolism imply that potential growth substrates pose1 a lesser
threat to the environment than compounds which cannot be used in this
way (Tiedje, 1980).
Before the utilization of a compound can begin, the microbial community
must adapt itself to the chemical. Investigations of biodegradation of a
compound to which the biota have not been recently exposed, both in the field
(Spain ert aJL 1980), and in the laboratory (Shamat and Maier, 1980), have shown
the existence of a lag time (lag phase) of 2 to 50 days before the microbial
community acclimates. Since the degradation of a growth substrate is
relatively rapid once a microbial population has adapted to it,
Tiedje (1980) has suggested that the primary concern in assessing
biodegradation of such substances should be the conditions and time period
required for adaptation or acclimation.
The lag time depends on several biological and environmental
constraints. The primary contraint is the development of a sufficiently
large bacterial population which is capable of utilizing the pollutant as a
growth substrate. Frequently, specific organisms with specific enzymes are
required to metabolize a pollutant. The processes of species selection and
enzyme induction by which a microbial community adapts itself to a pollutant
require time. The adaptation time is influenced both by prior exposure of
the community to a pollutant and the initial numbers of suitable species.
Spain et_ aJL (1980) have demonstrated that prior exposure to a compound
reduces or eliminates the adaptation period. Thus, lag times in pristine
environments should be much longer than in locations which have been chronically
exposed to a compound. In addition, Ward and Brock (1976) have shown that lag time
105
-------
preceding the onset of petroleum degradation depends on the initial size nf
the bacterial population. Water with larger microbial communities should
require relatively shorter times to develop a viable population of
degraders. High microbial biomass levels are associated with higher BOD
5
concentrations.
The presence of more easily degraded carbon sources may delay the
adaptation of a microbial community to the metabolism of a pollutant. Ward
and Brock (1976), found that microorganisms in lake water metabolized added
glucose completely before degrading hydrocarbons. This diauxic pattern may
result in longer lag times.
A final factor which influences lag time is the concentration of the
pollutant in the water. There may be concentration thresholds below which
adaptation does not take place. (For example, no adaptation for metabolism
of 4-nitrophenol occurred at concentrations below about 40 yg/1 (Spain j2t
al . , 1980). Too high a pollutant concentration, on the other hand, may be
toxic to the microbes (Tabak et a\_., 1981). The user should be aware of
these possibilities when extremely low or high concentrations are involved.
Once the microbial community has adapted to the organic pollutant, it
is of interest to know the rate at which biodegradation occurs. Kinetic
expressions for compounds used as a growth substrate can be relatively
complicated since both the substrate and bacterial concentrations change
with time. The Monod equation has been used to describe the degradation
rate of a compound which serves as a sole carbon source:
dC 1 dB ymax B • C
dt Yo K + c
s
where
C = pollutant concentration
B = bacterial concentration
Y = biomass produced per unit C consumed
y = maximum specific growth rate
max
K = half-saturation constant
s
106
-------
Frequently, the Monod equation is reduced to a second-order
biodegradation expression by assuming C «K , in which case:
(.1-57)
where
k = second-order biodegradation rate constant
82
s
Although Monod kinetics accurately describe some laboratory results,
they are inapplicable in the environment due to the presence of other carbon
sources. As a simple alternative, first order kinetics are frequently
applied:
-— = k • C (11-58)
dt B
where
k = first-order biodegradation rate constant.
B
This first-order expression is analagous to the equation commonly used for
the decay of BOD (see Chapter 4). Larson (1981) has shown that first-order
kinetics which include a lag phase (lag time) represent the degradation of
6
growth substrates reasonably well at initial bacterial concentration of 10
cells/ml or less, a condition which is usually met in the environment.
2.5.1.2.2 Cometabolism
Microorganisms also degrade compounds which they cannot use as a
nutrient or growth substrate through cometabolism. Cometabolism is thought
to occur when enzymes of low specificity alter a compound to form products
which the other enzymes in the organism cannot utilize. The metabolites
formed in the process are structurally similar to their parent molecules and
frequently retain their toxicity. In some cases, the product of
Cometabolism can be used as nutrients by other organisms, but often these
intermediate products accumulate (Alexander, 1980).
107
-------
The kinetics of microbial cometabolism differ significantly from that
of growth metabolism. Often no lag occurs before cometabolism begins. The
degradation rates, though, are generally slower than the fully adapted rates
of growth metabolism (Tiedje, 1980). Since cometabolism does not provide
the microbes with any energy, it has no effect on the population size. The
rate of cometabolism, however, is directly proportional to the size of the
microbial population. Paris et^ aiL (1981), showed that a second-order rate
law described microbially catalyzed hydrolytic reactions:
-£.I«K.B.C di-59)
Since the bacterial population, B, is independent of the rate of
cometabolism, it is possible to reduce Equation 11-59 to a first-order law
by making the following substitution:
kg = kB2 - B (11-60)
In order to use literature values of the second-order biodegradation
rate constant in Equation 11-60, it is necessary to make an estimate of the
size of the bacterial population. Since different techniques of bacterial
enumeration can yield results which vary over several orders of magnitude,
it is important to use estimates of B based on the same method used to
calculate k . Table 11-21 lists bacterial densities which are typical of
no
lakes and rivers. Obviously, large uncertainties in environmental rates of
cometabolism exist due to the wide range of possible bacterial densities.
Generally, the user should make conservative assumptions unless other data,
e.g. a high BOD, indicate larger bacterial densities.
2.5.1.2.3 Summary
Table 11-22 summarizes some of the major differences between growth
metabolism and cometabolism. Although the exceptions to the generalizations
about each process are numerous and some compounds can undergo both
processes, the distinction between the metabolic processes can serve a
useful function in a screening method. The generalization about each
108
-------
TABLE 11-21
SIZE OF TYPICAL BACTERIAL
POPULATIONS IN NATURAL WATERS
Water Body Type
Bacterial Numbers (cells/ml)
Ref.
Oligotrophic Lake
Mesotrophic Lake
Eutrophic Lake
Eutrophic Reservoir
Dystrophic Lake
Lake Surficial Sediments
40 Surface Waters
Stream Sediments
Rur River (winter)
50- 300
450- 1,400
2000-12,000
1000-58,000
400- 2,300
Q in
8xlOy - 5xl0lu cells/g dry wt
500-lxlO6
107-108 cells/g
3xl04
a
a
a
a
a
a
b
c
d
References:
aWetzel (1975). Enumeration techniques unclear
Paris et _§_]_. (1981). Bacterial enumeration using plate counts.
^>
Herbes & Schwall (1978). Bacterial enumeration using plate counts.
Larson et al_. (1981). Bacterial enumeration using plate counts.
109
-------
TABLE 11-22
SUMMARY OF THE CHARACTERISTICS OF THE TWO GENERAL TYPES OF
BIODEGRADATION: METABOLISM AND COMETABOLISM (After Tiedje, 1980)
Topics
Metabolism for Growth
Cometabolism*
Distinguishing characteristics
Degradation rates
Behavior at low pollutant
concentrations
Acclimation
Relation of degradation kinetics
to total biomass, e.g. decay
rate = kg2 • B • C
Extrapolation
Effect of added carbon
Organism will grow on suf^tance as sole C
source. Generally ultimate degradation.
High rates.
Possible anamalous behavior due to threshold
for enzyme induction.
Major effect: lag may be quite variable or
lengthy due to low initial density of
degraders, and perhaps starvation state
of organisms in natural sample.
Likely not valid, use first-order kinetics.
General: expect eventual degradation in nature.
Quantitative: difficult to be precise because of
growth kinetics and acclimation effects, but
may not be important problem because of
generally fast rates.
Diauxic pattern—more easily metabolized
substrates are used first.
Organism will not grow on substance as sole C source.
Accumulation of intermediate products likely.
Generally slow rates.
No anomalous behavior, rates are first order in
pollutant concentration.
Often no effect; rarely causes induction, may
increase tolerance to toxic chemical.
May be valid since activity of interest is often
proportional to general biomass.
Measure kinetic parameters accurately: because
of the generally slower rates, extrapolations
will be made over longer times, and thus measured
parameters need to be accurate. Also environ-
mental influence factors, e.g. temperature, pH,
play a more important role.
Generally effect is proportional to microbial
population unless specific carbon source happens
to induce or inhibit activity of interest.
*Alteration of a substrate, for purposes other than growth, e.g. for detoxification
-------
process suggest the following approaches when the user has some knowledge of
a compound's metabolic pathway:
• Cometabolism -
a) Find a second-order rate constant and estimate biomass
density. Apply Equations 11-59,60.
b) When a) is not possible, assume cometabolism is
negligible, i.e. k =0.
B
t Growth Metabolism
a) Find a first- or second-order rate constant.
b) Estimate a range of lag times as follows:
For chronically exposed water bodies, assume that no lag
time (t ) occurs. For water bodies not recently exposed
(within 200 days) proceed as follows:
1. Estimate lag time using available information. If no
information is available use a range of 2-20 days.
2. Assume adaptation occurs as follows:
a) Rivers - at travel times < t , k =0
L B
- at travel times > t , k ? 0
- L B
b) Lakes - for well mixed lakes, first determine
C at time = t , Ct due to all
processes except biodegradation.
Then
using Ct as C solve for Ct with a
modified time, t , (t = t-t ). (Use
mm L
equations in Section 5.6.1)
- for stratified lakes use only the
volume through which the inflow passes
(e.g. the hypolimnion volume) in
111
-------
calculating the hydraulic residence
time (T ). Then proceed as above.
w
c) Estuaries - consider all processes except
biodegradation through that downstream
segment for which T , as measured from
w
the injection point, becomes greater
than t . Thereafter include
biodegradation.
When no data on which metabolic pathway a compound follows are available,
the user should apply any available kinetic information and allow for the
possibility of a lag phase prior to the onset of degradation.
2.5.1.3 Chemical Properties Influencing Biodeqradation
The chemical properties of a compound determine whether microbes can
potentially utilize it as a growth substrate or not. Compounds which serve
as bacterial growth substrates usually decay more rapidly than those which
microbes cometabolize. Thus, significant differences in the aquatic fate of
pollutants can arise depending on which degradation process takes place.
Unfortunately, it is not possible at this time to predict whether a
toxic compound is a potential source of energy and carbon solely on the
basis of its chemical structure. Rather, the biodegradability of a compound
is usually investigated in laboratory tests (Gilbert and Lee, 1980).
Compounds which are growth substrates should be able to serve as sole carbon
sources for a microbial community. Compounds which cometabolize should
degrade only in the presence of another carbon source. A systematic study
of the metabolic pathways of the priority pollutants is desperately needed.
Table 11-23 contains the results of a preliminary degradation test on
the organic priority pollutants (Tabak jjt a\_., 1981). Because the
experimental conditions were so favorable for biodegradation, the tests
serve as a good indicator of a compound's potential biodegradability. Since
the pollutants were not the sole carbon sources, no conclusions can be
112
-------
POTENTIAL BIODEGRADABILITY OF ORGANIC POLLUTANTS
IN AN AEROBIC ENVIRONMENT
(After Tabak et. al_., 1981)
Pesticides
Test Compound
Al drin
Dieldrin
Chlordane
DDT p.p'
DDE p.p'
DOD p.p'
Endosul fan-al pha
Endosul fan-beta
Endosul fan sul fate
PCB-1016
PCB-1221
PCB-1232
PCB-1242
Chloroethanes
1 ,1-Dichloroethane
1 ,2- Dichl oroethane
1,1, 1-T rich! oroethane
1 ,1 , 2 -Trichl oroethane
1 , 1 ,2 ,2-Tetrachl oroethane
HexacMoroethane
Hal ^methanes
Mettiylene chloride
Bromochlorome thane
Carbon tetrachlon de
Chi oroform
0 i chlo rob romome thane
Bromoform
Chlo rod ibromo methane
Tn chlorofl uorome thane
Adaotat i on
Summary
N
N
N
N
N
N
N
N
N
N
D
D
N
A
B
B
C
N
D
D
D
D
A
A
A
N
N
Rate
Summary
0
0
0
0
0
0
0
0
0
PCB's and Related
0
2
2
0
Halogenated Aliphatic
1
1
1
1
0
2
2
2
2
2
1
1
0
0
Test Compound
Endrin
Heptachl or
Heptachlor epoxide
Hexachlorocycl ohexane
ti-BHC-alpha
Hexachlorocyclohexane
B-BHC-beta
Hexachlorocycl ohexane
4>-BHC-delta
Hexachlorocycl ohexane
X-BHC-gamma (lindane)
Acrolein
Compounds
PCB-1248
PCB-1254
PCB-1260
2-Chl orohaphthal ene
Hydrocarbons
Chloroethylenes
1,1-Dichloroethylene
1 ,2-Dichloroethylene-cis
1 .2-Dichloroethylene-trans
Trichloroethylene
Tetrachloroethylene
Chloropropanes
1,2-Di chlo ro propane
Chloropropylenes
1,3-Dichloropropylene
CMorobutadienes
Hexachloro-1 ,3-butadiene
Chi oroperttadienes
Hexachl orocycl open tad iene
Adaptation
Summary
N
N
N
N
N
N
N
D
N
N
N
D
A
B
B
A
A
A
A
D
D
Hate
Summary
0
0
0
0
0
0
0
2
0
D
0
2
2
1
1
1
1
1
1
2
2
Halogenated Ethers
Bis-(2-ch1oroettiyl) ether'
2-Chloroethyl vinyl ether
4-Chlorodi phenyl ether
Results of Tdbok et al.(i981)
conditions, ine test measures
duration = 28 days.
Key to Test Summary
D
D
N
2
2
0
using Bunch and Chambers screening test.
disappearance rather than mineralization of
4-Bromodi phenyl ether
Bis-(2-chloroethoxy) methane
Bis-(2-chloroisporopyl ) ether
Results reflect potential biodegradability
a compound. A domestic sewage innoculum
N
N
D
0
0
2
under favorable
was used. Test
N - Not significantly degraded under conditions of test method.
D - Significant degradation with rapid adaption; <7 days.
D* - Same as D except slower adaptation at higher pollutant concentration.
A - Significant degradation with gradual adaptation, 7-21 days.
A* - Same as A except no degradation evident at higher pollutant concentration.
B - Slow degradation.
C - Very slow degradation with long adaption period required, >28 days.
T - Significant degradation with gradual adaptation followed by deadaption (toxicity).
Key to Rate Summary
Very crude estimates of first-order biodegradation rate constants may be made from the information given in
0 - No significant degradation rate
-1
1 - .05 day"
2 - k
-.5 day , use .05 day"
, .5 day 'l , use .5 day"1
113
-------
TABLE 11-23 (continued)
Test Compound
Benzene
Chlorobenzene
1,2-Dichlorobenzene
1,3- Dichloro benzene
1,4- Dichloro benzene
1,2, 4-Trichloro benzene
Phenol
2-Chloro phenol
2,4-Dichloro phenol
2,4,6-Trichloro phenol
Pentachloro phenol
2 ,4 Dlmethylphenol
Dimethyl phthalate
Diethyl phthalate
Di-n-butyl phthalate
Napthalene
Acenapthene
Acenaphthylene
Anthracene
Phenanthrene
Adaptation
Summary
D
D
T
T
T
T
D
D
0
D
A
D
0
D
0
D
D
. D
A
D
Rate
Summar
2
2
1
1
1
1
2
2
2
2
1
2
2
2
2
I
2
2
1
2
Monocyclic Aromatics
y Test Compound
Hexachl orobenzene
Nitrobenzene
Ethyl benzene
Toluene
2,4-Dinitrotoluene
2,6-Dimtrotoleune
Phenolic Compounds
p-Chloro-m-cresol
2-N1tro phenol
4-Nitro phenol
2,4-Dinitro phenol
4,6-Dini tro-o-cresol
Phthalate Esters
Bis-(2-ethyl hexy! ) phthalate
Di-n-octyl phthalate
Butyl benzyl phthalate
Polycyclic Aromatic Hydrocarbons
Fluorene
Fluoranthene
1,2-Benzanthracene
Pyrene
Chrysene
Adaptation
Summary
N
D
D
D
T
T
D
D
D
D
N
A
A
D
A
A*
N
D*
A*
Rate
Sumtna ry
0
2
2
2
1
1
2
2
2
2
0
1
1
2
1
2
0
2
1
Nitroso Amines and Miscellaneous Compounds
Ni trosamines
N-Nitroso-dl-N-
propylamine
N-Nitrosodiphenyl amine
N
D
0
2
Substituted benzenes
Isophorone
1 ,2-Diphenylhydrazine
Acrylonitrile
D
T
D
2
1
2
Results of Tabak et a|. (1981) using Bunch and Chambers screening test. Results reflect potential biodegradabil ity under favorable
conditions. T~he test measures disappearance rather than mineralization of a compound. A domestic sewage innoculum was used. Test
duration * 28 days.
Key to Test Summary
N - Not significantly degraded under conditions of test method.
D - Significant degradation with rapid adaption; < 7 days.
D* - Same as D except slower adaptation at higher pollutant concentration.
A - Significant degradation with gradual adaptation, 7-21 days.
A* - Same as A except no degradation evident at higher pollutant concentration.
B - Slow degradation.
C - Very slow degradation with long adaption period required; > 28 days.
T - Significant degradation with gradual adaptation followed by deadaption (toxicity).
Key to Rate Summary
Very crude estimates of first-order biodegradation rate constants may be made from the information given In lab^k el_ <0 .
0 - No significant degradation rate
1 - .05 day" < k < .5 day" ; use .05 day~
2 - kB > .5 day"*; use .5 day"1
114
-------
reached about their metabolic pathways. Some information on the rates of
adaptation and decay, though, can be extracted from the results.
The adaptation summary results may be used as follows:
• Rapid Adaptation (D) - Use a range of adaptation times from
zero days upward depending upon conditions described above.
• Gradual Adaptation (A) - Use a range of adaptation times from
7 days to more than 20 depending upon the conditions described
above.
The rate summary results represent estimates of the biodegradation rate
constants assuming the compounds decay according to first-order kinetics.
General values presented at the bottom of the table are gross estimates and
should only be used if no better data are available. The rate constants
should represent an upper limit for biodegradation rates by adapted populations
observed in the environment.
Table 11-24 contains literature values of biodegradation rate
constants. Where possible, the likely metabolic pattern has been indicated.
Some of these constants were measured under environmentally relevant
conditions. In general, rate constants should be compared with those in
Table 11-24 before use.
2.5.1.4 Environmental Influences on Biodegradation Rates
Environmental conditions strongly influence the metabolic activity of a
microbial community. The environment affects the types of metabolic
reactions microbes are able to carry out, the availability of nutrients for
these reactions, and the rates at which these reactions occur. The
environmental variables which are responsible for these effects are
discussed in the following sections.
115
-------
TABLE 11-24
BIODEGRADATION RATE CONSTANTS UNDER AEROBIC CONDITIONS
Cc.npcund
Pesticides
2,4-D Butoxyethyl ester
Malathion
Chlorpropham
Furadan
Atrazine
1 Polychlorinated Blphenyls
Aroclor 1221
Aroclor 1016
Aroclor 1242
Aroclor 1254
Halogenated Ethers
4-Chlorophenyl phenyl ether
Monocyclic Aromatics
Nitrobenzene
2-Chlorotoluene
Phenolic Compounds
Phenol
2-Chlorophenol
2,4-Dlchlorophenol
Pentachlorophenol
2,4-Oimethylphenol
2,4-Dinitrophenol
2,4 ,6-Tri ni trophenol
Phthalate Esters
Dimethyl
Di-ethyl
Dl-n-butyl
Di-n^-octyl
Di-(2-ethylhexyl
Butyl Benzyl
k k
Second-Order First-Order
Rate Constant Rate Constant
(ml cell"' day"1) (I/day)
1.2xlO"5'3 1.3X10"2'1
1.1x10"'° l.lxlO"'(1
6.2X10"10'3 6.2xlO"7(1
2.4x10"' 2.4x10"'''
2.4x10"' 2.4xlO"!'4
.8<2
.2<2
.15<2
.1(2
.011-.01612
3.8<4
.7'2
6.5x10" (3 6.5x10"*
4.'2
6.<2
l.<2
.3
.5<2
.1(4
.1(2
I.'2
I.'2
.2<2
0
1.2x10"' .12(1
• 7.7x10"' 7.7X10"5'1
7.0xlO"7 7.0x10"*''
7.4x10"' 7.4X10"6'1
1.0x10"" l.OxlO"7
2.5xlO"2'4
>.35<4
t
'5 Reference
Half-Life Temperature
(days) («c)
53 20
6.3xl02 20
1.1x10* 20
3 x 10* ?
3 x 10* ?
.9 ?
3.5 ?
4.5 ?
7. ?
43-63 ?
.2 ?
1. 20
1.1x10 ?
.2 20
.1 ?
.7 20
2.3 ?
1.4 20
6 25
7 25
.7 25
.7 20
3.5 20
20
5.6 ?
9.0xlO! ?
1.0x10' ?
9.3x10* ?
6.9x10' ?
28 ?
<2 ?
Compound
Used as a
Growth Substrate? Experimental Conditions Ref.
' Natural surface water sanples
Yes Natural surface water sanples
? Natural surface water samples
7 7
7 7
? Acclimated activated sludge
? Acclimated activated sludge
? Acclimated activated sludge
? Acclimated activated sludge
? River water; Log = 5-13 days
? Activated sludge
Yes Adopted activated sludge, COD decay
? Natural surface water sample
Yes Adapted activated sludge, COD decay
? Polluted river water
Yes Adopted activated sludge
7 Soil suspension
Yes Adapted activated sludge; COD decay
7 Natural lake waters
Yes Unadapted; Nutrient Broth
Yes Adapted; Nutrient Broth
Yes Adapted activated sludge
Yes Adapted activated sludge
No Activated sludge
7 7
? ?
7 7
7 7
7 7
7 River Water
? River Water
a
a
a
b
b
c
c
c
c
c
c
d
e
d
c
d
c
d
c
f
f
d
d
d
g
g
g
g
g
c
c
116
-------
TABLE 11-24 (continued)
kB2 kB
Second -Order Fi rst-Order
Rate Constant Rate Constant
Compound (ml /cell /day) (I/day)
Polycyclic Aromatic Hydrocarbons
Napnthalene - - H
< 4.X10-
Arthracene - .0025
2.5x10"*
1.5
Benz( a)anthracene - l.xlO
A _ i n~ '
t . X 1U
Benz(a)pyrene - < 3x10
< 3x10" s
Phenanthrene 3.8x10"' 3.8x10"'"
t
Half-Life
(days)
5.0
1.7X101
2.8xl02
2.8xlOs
.5
6.9x10'
large
large
1.8x10*
T
o
Reference
Temperature
rc)
12
12
12
12
?
12
1 7
It
12
12
-j
Compound
Used as a
Growth Substrate7 Experimental Condi tons Pef
Yes Contaminated stream srdiF'ents h
? Pristine strean sediments h
Yes Contaminated stream sediments h
? Pristine stream sedinents h
? Contaminated stroarn c
Yes Contaminated stream secnments h
? Contaminated stream sedirr.ents h
? Pristine stream sediments h
? ? e
Notes' References•
1} First-order rate constant computed using Equation 11-60 and B = 103 cells/ml. a) Paris et_ al_ (1981)
2} First-order constant calculated from percent disappearance and elapsed time. b) Schnoor (1981}
3} Bacterial enumeration using plate count technique. c} Callahan ei a_l_. (1979)
4) First-order rate constant computed from reported half-life d) Fitter (1976)
e) Paris et aj (1930)
f) Kirsch and Ctzel (1973)
g) Wolfe et a_]_ (]9?0)
h) herbfs & Schwall (1978)
117
-------
2.5.1.4.1 Temperature
In general, a molecule must have an energy greater than a threshold or
activation level in order for it to react chemically. Since increasing the
temperature increases the number of molecules which have this minimum
energy, both biotic and abiotic reactions generally proceed more rapidly at
higher temperatures. However, the fact that enzymes catalyze most
biochemical reactions, and that microbial populations can adapt to changes
in ambient temperatures, complicate the temperature dependence of
microbially mediated reactions.
It is common practice to represent the temperature dependence of
biodegradation using the following empirical formula:
k (T) = k (T ) • 0 o (11-61)
B B o B
where
k (T) = specific biodegradation rate constant at temperature = T
k {T ) = specific biodegradation rate constant at temperature = T
Boo o
T = ambient temperature, C
o
T = reference temperature, C
o
0 = temperature coefficient for biodegradation
B
The results of Larson et al_. (1981) and Ward and Brock (1976) show that
the rates of nitrilotriocetate and hydrocarbon biodegradation increased
approximately two-fold over a ten degree temperature range (0 = 1.072).
Either this value or the typical value of 1.047 used for BOD decay is
adequate for screening purposes.
2.5.1.4.2 Nutrient Limitation
Microbes require nutrients such as nitrogen and phosphorus in order to
metabolize organic substrates. Several researchers have suggested that
inorganic nutrient limitation is a significant factor influencing
biodegradation rates in the aquatic environment (Ward and Brock, 1976;
118
-------
Roubel and Atlas, 1978; Herbes and Schwall, 1978). Ward and Brock (1976)
found a high correlation between hydrocarbon degradation rates and
phosphorous concentrations in natural waters. The data fit a saturation
relationship of the Michael is-Menten type:
where
k (C ) = specific biodegradation rate constant at dissolved
B p
inorganic phosphorus concentration, C
P
C - dissolved inorganic phosphorous concentration, ya/1
P
k (C *) = non-nutrient limited biodeqradation rate onstant
B p
This relationship should serve as a good indicator of possible
phosphorous limitation of biodegradation in the environment. Generally
surface waters downstream of domestic sewage treatment plants are not
limited in either nitrogen or phosphorus. Equation 11-62 should be applied
only when other nutrients such as carbon and nitrogen are not limitina.
2.5.1.3 Sorption of Substrates
Many organic pollutants adsorb strongly on sediments, (See
Section 2.3.2. The difference in the physical and chemical environments
between sorbed and dissolved pollutants is likely to influence their
availability to microbial organisms. Baughman et aj_. (1980) showed that the
dissolved fraction of the compounds studied was available to biota for
degradation while the sorbed fraction was not. In such cases, the rate of
disappearance of the pollutant is:
d CT
._! = k • C = a 'k -C (H-63)
dt B w w B T
119
-------
where
C = the pollutant concentration in the aqueous phase
w
a = the decimal fraction of the total analytical pollutant
W
concentration which is in the aqueous phase (a = 1 - fraction
sorbed).
It is well known, however, that bacteria grow very readily on surfaces
and that increasing available surface area in the form of clays and
sediments can increase rates of microbial metabolism. If specific
information regarding the effects of sorption on the rates of biodegradation
are not available for a compound, it is best to assume that sorption does
not change this rate.
2.5.1.4.4 Solubility
Wodzinski and Bertolini (1972) have shown that in the dissolved state,
naphthalene and biphenyl were degradable while in the pure crystalline state
they were not. Thus, sparingly soluble compounds could degrade slowly for
this reason alone. The extent to which this phenomenon applies to other
biodegradation reactions has not been established. The user may assume that
only dissolved and sorbed chemicals are degraded.
2.5.1.4.5 £H
The hydrogen ion concentration also influences rates of biodegradation.
Each bacterial species has a pH range for which it is best suited. Thus, at
different pH values, different species may exist, or a given species may
metabolize the pollutant at different rates. Hambrick ot_ aJN (1980) found
that the mineralization rate of naphthalene in oxidizing sediments varied in
the proportions 1:6:5 at pH 5, 6.5, and 8. The same study found that the
mineralization rates of octadecane varied in the proportions 4:5:7 at the
same three pH's. Until more general rules for predicting pH effects are
120
-------
available, the user should assume biodegradation rates are independent of pH
in the pH range of 5-9 and decrease outside this range.
2.5.1.4.6 Anoxic Conditions
As the concentration of dissolved oxygen in natural water is depleted,
metabolic pathways shift. When the dissolved oxygen concentration drops to
about 1 mg/1, the rate of biodegradation becomes dependent on oxygen
concentration in addition to substrate concentration and the rate of
degradation starts to decrease. At a dissolved oxygen concentration of
about 0.5 to 1.0 mg/1, nitrate begins to substitute for molecular oxygen as
an oxidant.
When oxygen is depleted anaerobic metabolism prevails with its
generally lower energy yields and lower growth rates. Most organic
substances are biodegraded more slowly under anaerobic conditions. Rate
constants derived for oxygenated systems are no longer appropriate; their
use may overpredict the amount of degradation.
Exceptions do exist to the rule of slower degradation under anoxic
conditions. Reactions such as dehydrochlorinations and reductive
dechlorinations lead to much higher degradation rates for many chlorinated
hydrocarbons. Example compounds include lindane, heptachlor,
pentachlorophenol, and some one and two carbon chlorinated alkanes.
121
-------
EXAMPLE 11-6
Blodegradability of Naphthalene
Evaluate the biodegradability of naphthalene discharged into the
Lepidoptera River by a point source just upstream from Northvilie's sewage
treatment plant. Assume the following water quality parameters at the
upstream discharge:
o
Temperature = 10 C
Suspended Sediment = 10 mg/1
Inorganic Phosphorus = 5 yg/1
Dissolved Oxygen = 5 mg/1
First, check the potential biodegradability of naphthalene in
Table 11-23. The table indicates that naphthalene degrades rapidly,
kB= .5 day , and that bacteria adapt quickly to it.
Next, examine Table 11-24 for further information on naphthalene's
biodegradability. Naphthalene is a potential growth substrate. In
addition, the data in this table concur with the rapid degradation rates
suggested by Table 11-23. In sediment, which had been previously exposed to
naphthalene, a biodegradation rate constant of 0.14 day was measured.
This is close to the rate constant in Table 11-24. As one would expect for
-4 -1
a growth substrate, degradation rates are much lower, e.g. kg < 4x10 day ,
in sites not previously exposed to naphthalene.
Since naphthalene is a growth substrate, estimating the adaptation time
in the Lepidoptera River is a primary issue. Because the point source
continuously discharges naphthalene into the Lepidoptera River, it is safe
to assume that the bacterial populations have adapted.
In a complete analysis, the user would check whether the oxygen is
depleted from the river. If so, degradation could be neglected until
dissolved oxygen levels exceed 1.0 mg/1 again.
122
-------
Sorption by suspended sediment could potentially reduce the rate at
which naphthalene biodeqrades. Table II-9 gives a K for naphthalene of
ow
2,300. Using Equations 11-17 and 11-19 and assuming a suspended sediment
organic carbon content of 2 percent, the partition coefficient is:
K = (.02) (.63) (2,300) = 29
P
At the suspended sediment levels in the Lepidoptera River 10 mg/1, Table
11-14 shows that sorption will not significantly reduce water column
concentrations of naphthalene. Although phosphorus levels are low, assume
carbon is the growth-limiting substrate.
Finally, the degradation rate is adjusted to the river water
temperature using Equation 11-61.
(10-12) -1
k = 0.14 • 1.072 ' = 0.12 day
END OF EXAMPLE 11-6
123
-------
2.5.2 Photolysis
2.5.2.1 Introduction
The sun provides the aquatic environment with a large supply of energy.
Substances which absorb sunlight transform much of its radiant energy into
thermal energy. But, molecules which absorb sunlight in the ultraviolet and
visible portion of the spectrum may gain sufficient energy to initiate a
chemical reaction. Plants use very specific photochemical reactions to
provide energy for the synthesis of sugar from carbon dioxide. In other
photochemical reactions, the absorption of light leads to the decomposition
of a molecule. The latter type of reaction, known as photolysis, strongly
influences the fate of certain pollutants in the aquatic environment.
Photolysis is truly a pollutant decay process since it irreversibly
alters the reacting molecule. However, the products of the photochemical
decomposition of a toxic compound may still be toxic. For example,
irradiated 2,4-D esters form 2,4-D acid, a priority pollutant, in aerated
waters (lepp et _al_., 1975). Upon irradiation, DDT reacts to form DDE, which
persists in the environment longer than DDT (Tinsley, 1979). Thus, even
though the methods in this section assume that pollutants irreversibly decay
through photolysis, the planner should remember that the decomposition of a
pollutant does not imply the detoxification of the environment.
The rate at which a pollutant photolyzes depends on numerous chemical
and environmental factors. The light absorption properties and reactivity
of a compound, the light transmission characteristics of natural waters, and
the intensity of solar radiation are some of the most important factors
influencing environmental photolysis. These factors will be covered by the
following discussion. Understanding these factors facilitates the
computation of rate constants and the identification of pollutants likely to
photolyze - the final two topics of this section.
124
-------
2.5.2.2 Factors Influencing Photolysis in the Aquatic Environment
2.5.2.2.1 Photochemical Reactions
All chemical reactions which occur at finite rates require the reacting
molecule to gain sufficient energy to become "activated" or form a reactive
intermediate. In dark or thermal reactions, the thermal energy of the
environment supplies the activation energy. In photochemical reactions, the
absorption of light provides the activation energy.
The "activated" molecules in photochemical reactions differ in
important respects from those of thermal reactions. Thermally activated
molecules usually remain in the normal or "ground" electronic energy state,
whereas photochemically activated molecules exist in higher, "excited"
electronic states. Because of the excess energy and the alteration of the
chemical bonds of photoactivated molecules, the range of potential reaction
products is much greater than that for thermally activated molecules.
The mechanism by which photoactivated molecules form and react is
divided into three steps: 1) the absorption of light to produce an
electronically excited molecule, 2) the "primary photochemical processes"
which transform or deexcite the excited molecule, and 3) the secondary or
"dark" thermal reactions which the intermediates produced in step 2 undergo
(Turro, 1978).
The mechanism of photochemical reactions provides a convenient structure
for a discussion of the factors which influence photolysis in the aquatic
environment. Environmental factors affecting the absorption of light, step
1, will be considered first. Then, the factors influencing the fate of
molecules which become excited by the absorption of light, steps 2 and 3,
are discussed.
125
-------
2.5.2.2.2 Light Absorption
"Only that light which is absorbed by a system can produce
chemical changes (Grotthaus-Draper Law)." (Glasstone, 1946)
As this "first law of photochemistry" implies, it is necessary to know
the rate at which reacting molecules absorb light in order to determine the
rate of a photochemical reaction in the environment. The following factors
which influence light absorption in the aquatic environment are discussed
here: 1) molecular absorption of light, 2) solar radiation, and 3) light
attenuation in natural waters.
2.5.2.2.2.1 Molecular Absorption of Light
Both light and molecules have quantized energies. Light interacts with
matter as quanta with energies inversely proportional to their wavelengths. A
molecule has quantized internal energy states associated with the
configuration of its electrons and the rotation and vibration of its
chemical bonds. Since a molecule can absorb light only as a whole photon,
light absorption is possible only if the energy of the photon corresponds to
the energy change of an allowed transition between the molecule's internal
energy states. Consequently, the probability of a photon being absorbed
varies strongly with wavelength of the light in a way that is unique to
every chemical species.
To initiate a chemical reaction, the absorbed light must be
sufficiently energetic to cause a change in the absorbing molecule's
electronic structure. Generally, radiation with wavelengths in the
ultraviolet-visible range, or shorter, has sufficient energy to initiate
photochemical reactions while radiation with wavelengths in the infrared
range, or longer, does not. Thus, the ultraviolet-visible light absorption
properties of a chemical are of primary interest in photochemistry.
126
-------
Photochemical reactions in the aquatic environmental depend on the rate
at which molecules in aqueous solution absorb light. According to Beer's
Law, the rate of light absorption by a single compound (I ) in a
a
cross-section of solution with infinitesimal thickness (Az) is proportional
to the concentration of the light absorbing species(C), i.e.:
I (z) = I(z) • 2.3 • c • C • Az (11-64)
a
where I (z) is the intensity of the light at a depth z in the solution and e
is the base 10 molar extinction coefficient, e reflects the probability of
the light being absorbed by the dissolved molecules and therefore varies
with the wavelength of the incident light as shown in Figure 11-10.
Absorption spectra, such as shown here, contain information necessary to
compute the rate at which pollutants absorb radiation available in the
environment.
2.5.2.2.2.2 Solar Radiation
The only radiant energy available for absorption by pollutants in the
aquatic environment comes from the sun. The sun emits radiation of nearly
constant intensity and spectral distribution. But, gases and particles in
the earth's atmosphere alter the incoming solar radiation through scattering
and absorption. Scattering of the direct solar beam creates the diffuse or
sky radiation visible at the earth's surface. Absorption of both diffuse
and direct radiation reduces the intensity of solar radiation reaching the
earth. Since the strength of absorption and scattering depends strongly on
the wavelength of the light involved, the interaction of sunlight with the
atmosphere alters the spectral distribution of solar radiation as well, as
Figure 11-11 shows.
The composition of the earth's atmosphere and the geometrical
relationship of the sun and earth change over time causing the solar
radiation incident upon the earth's surface to vary as well. A comparison
of the total solar irradiance under clear skies at various times, seasons,
and latitudes (Table 11-25) to the extra-atmospheric solar flux of
2800 langleys/day demonstrates the effects of changes in earth-sun geometry.
127
-------
ULTRAVIOLET ABSORPTION SPECTRUM OF NAPHTHACENE
106 r
182 200
250 300
Wavelength (nm)
400
500
Source: U.V. Atlas of Organic Compounds.
FIGURE 11-10 ULTRAVIOLET ABSORPTION SPECTRUM OF NAPHTHACENE
128
-------
>O —I
I—I I I—I—I I—I
1500 2000
a) Spectral distribution of sun's
radiation at edge of outer
atmosphere
b) Spectral distribution of sun's
radiation at earth's surface
Wavelength (nm)
Sources: (a) Weast and Astle (1980); (b) Moon (1940).
FIGURE 11-11 SPECTRAL DISTRIBUTION OF SOLAR ENERGY
(A) OUTSIDE THE EARTH'S ATMOSPHERE, AND
(B) AT THE EARTH'S SURFACE
129
-------
Table 11-25
CALCULATED SOLAR RADIANT ENERGY FLUX TO A HORIZONTAL SURFACE UNDER A CLEAR SKY
(langleys/day)
CO
o
Latitude
30°N
40°N
50°N
Time
Of Day
Mean1^
Mid-Day^'
Mean
Mid -Day
Mean
Mid- Day
Spring
680
2100
650
1900
590
1700
Season
Summer
750
2200
740
2100
710
1900
Fall
530
1700
440
1400
330
1000
Winter
440
1400
320
1000
190
650
Annual
Mean
600
1900
540
1600
460
1300
1) Mean values represent calculated seasonal means under a clear sky. These
should represent upper limits for solar radiant energy at sea level.
Reference: Weast and Astle (1980).
2) Mid-Day values represent mid-day flux extended over a 24-hour period. These
assume an atmospheric turbidity of 0, precipitable water content of 2 cm,
and an atmospheric ozone content of .34 cm NTP. Reference: Robinson (1966).
-------
The composition of the atmosphere differs greatly from place to place and is
the most difficult of the factors influencing the total solar flux to
accurately quantify. Historical records of the solar radiation, such as
shown in Figure 11-12, are the best way to estimate the mean solar energy
flux at a given locale. However, care should be taken to account for the
influence of riparian vegetation on incoming radiation. Section 4.4.3
discusses how to approximate the affects of shading.
Information concerning the variability of the spectral distribution of
solar energy incident upon the earth's surface is not as readily available.
It is known that the fraction of the solar energy in the ultraviolet region
decreases with increased attenuation of light by the atmosphere. The
fraction of the energy which is visible remains relatively constant. For
the purpose of this document, it is sufficiently accurate to assume that the
reduction in UV-visible radiation is proportional to the reduction in the
total flux.
2.5.2.2.2.3 Light Attenuation in Natural Waters
Just as the earth's atmosphere reduces the intensity of solar radiation
reaching the earth's surface, natural waters reduce the intensity of
radiation available for absorption by aquatic pollutants. The first process
which reduces the availability of light in the water column is reflection.
The surface of the water reflects less than 10 percent of solar radiation in
most cases (Zepp and Cline, 1977). Reflection also alters the solar
spectrum slightly. A calculated spectral distribution of solar radiation,
expressed in photons, immediately below the surface of a water body is
presented in Table 11-26.
As solar radiation penetrates deeper into natural waters, it is
absorbed and scattered by particulates, dissolved substances, and water
itself. Measurements of light attenuation in natural waters have been based
on the decrease of solar irradiance, which includes both collimated and
scattered light. Lambert's Law expresses the decrease in the irradiance,
I(z), i.e. the total flux incident upon dn element of surface divided by its
area, with depth, z, as follows:
131
-------
OJ
ro
EAN DAILY SOLAR RADIATION (Langleys), ANNUA
Ref: US Dept. Comm. (1963)
FIGURE 11-12 SOLAR RADIATION IN THE UNITED STATES
-------
Table 11-26
CALCULATED SOLAR IRRADIANCE IN A WATER BODY JUST BENEATH
THE SURFACE, ANNUAL MEAN AT 40°N
Wavelength
(nm)
300
310
320
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
480
490
500
510
520
550
600
650
700
750
800
Photon Spectral
W(X)C
(10 photons cm" secern1)
,00303
.0388
.113
.181
.211
.226
.241
.268
.294
.366
.526
.692
.712
.688
.814
.917
.927
.959
.983
.930
.949
.962
1.00
1.04
1.07
1.08
1.07
1.03
.988
Irradiance
W!(X)d
14 9ii
(10 photons cm sec!X nm1)
.0303
.388
1.13
1.81
2.11
2.26
2.41
2.68
2.94
3.66
5.26
6.92
7.12
6.88
8.14
9.17
9.27
9.59
9.83
9.30
9.49
9.6?
10.0
52.0
53.5
54.0
53.6
51.5
49.4
Estimated reference solar flux, I = 540 langleys/day. D =1.0
Centric wavelength of waveband X nm in width,
for 300
-------
- d~= K ' I(z) (11-65)
where K is the diffuse light attenuation coefficient. The diffuse
attenuation coefficient can be expressed as a sum of terms accounting for
absorption, a, and backward scattering of light, s (Smith and Tyler, 1976):
b
K = Da + s (11-66)
b
where D is the radiance distribution function. Usually, s is small
b
compared to the absorption term. The absorption term constitues part of the
beam attenuation coefficient, a, which can be measured in a
spectrophotometer:
a = a + s + s (11-67)
b f
where s is the forward scattering coefficient of the solution.
The inclusion of the distribution function, D, in Equation (11-66)
accounts for the difference in mean light pathlength of collimated and
diffuse light. Perfectly diffuse light has a mean path through an element
of water which is twice as long as that of a beam of light. The
distribution function, generally increases asymptotically with depth due to
the increasing fraction of the total light which is scattered. In water
bodies where scattering can be ignored, D has a value of 1.2. Miller and
Zepp (1979) reported that the mean value of D for six sediment laden waters
was 1.6.
The diffuse light attenuation coefficient of natural waters differs
greatly due to variations in the types and amounts of particles and
dissolved substances in the water. Miller and Zepp (1979), Zepp and
Schlotzhauer (1981), and Smith and Baker (1978) have investigated the
contributions of suspended sediments, dissolved organic carbon, and
chlorophyll pigments to the light attenuation coefficient. By using
Equation (11-66) to integrate the results of these investigations, and
assuming backscattering to be negligible, Burns ert _al_. (1981) derived the
following expression to estimate the diffuse light attenuation coefficient:
134
-------
K = D
(11-68)
where a
w
a
cfil a
aDOC
DOC
a + (a • chl a) + (a • DOC) + (a • SS'
w a - DOC ss
absorptivity of water
absorptivity of chlorophyl1-a pigment
concentration of chlorophyll-^, pigment
absorptivity of dissolved organic carbon
concentration of dissolved organic carbon
absorptivity of suspended sediments
concentration of suspended sediments
Each absorptivity term varies with the wavelength of light, as shown in
Table 11-27.
Diffuse light attenuation coefficients can also be estimated using
turbidity indicators such as Secchi disc depth. Empirical studies have
shown that the diffuse light attenuation coefficient is inversely
proportional to the Secchi disc depth, Z
sd
K =
_ R
"sd
(11-69)
The proportionality constant, R, has a value between 1.44 and 1.7 for
visible light, i.e. 400-800 nm. In the middle ultraviolet portion of the
spectrum, i.e. near 312 nm, R has a value of 9.15 (Zepp, 1980).
2.5.2.2.3 Fate of Excited Molecules
"Each molecule taking part in a chemical reaction which is a
direct result of the absorption of light takes up one quantum of
radiation (Stark-Einstein Law)." (Glasstone, 1946)
According to this "second law of photochemistry", the extent to which a
photochemical reaction progresses depends on the number of quanta of light
absorbed. Each absorbed photon produces an electronically excited molecule
which can undergo numerous processes, including reaction. Factors which
influence the fraction of excited molecules which undergo reaction, called
135
-------
Table 11-27
CONTRIBUTIONS TO LIGHT ATTENUATION COEFFICIENT
Waveband
Center
(nm)
300
310
320
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
480
490
500
510
520
550
600
650
700
750
800
Notes:
a a
w
(m-1)
.141
.105
.0844
.0678
.0561
.0463
.0379
.0300
.0220
.0191
.0171
.0162
.0153
.0144
.0145
.0145
.0156
.0156
.0176
.0196
.0257
.0357
.0477
.0638
.244
.349
.650
2.47
2.07
? Source: Smith and
Source: Smith and
ab
a
[(mg/i)"1
69.*
67.*
63.*
61'*
58.*
55.
55.
51.
46.
42.
41.
39.
38.
35.
32.
31.
28.
26.
24.
22.
20.
18.
16.
10.
6.
8.
3.
2.
0.
Baker (1981)
Baker (1978)
a c
aDOC
m'1] [(mg/1)-1 nT1
6.25
5.41
4.68
4.05
3.50
3.03
2.62
2.26
1.96
1.69
1.47
1.27
1.10
0.949
0.821
0.710
0.614
0.531
0.460
0.398
0.344
0.297
0.257
0.167
0.081
-
-
-
~
Calculated using
a d
ass
] [(mg/1)-1 m-1]
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
.35
aa = K2/D,
D = 1.2
cSource: Zepp and Schlotzhauer (1981)
dSource: Miller and Zepp (1979). Calculated using ass = KS/D.
Denotes extrapolated values.
136
-------
the quantum yield, comes first in the following discussion of the fate of
excited molecules. Then, the two major classes of enviromental photolysis
reactions, direct and sensitized, are discussed.
2.5.2.2.3.1 The Quantum Yield
Although all photochemical reactions are initiated by the absorption of
a photon, not every absorbed photon induces a chemical reaction. Besides
chemical reactions, possible processes which excited molecules may undergo
include the reemission of light through flourescence and phosphorescence,
the internal conversion of the photons' energy into heat, and the excitation
of other molecules, as shown in Figure 11-13. The fraction of absorbed
photons which cause the desired reaction(s) is termed the quantum yield, $:
moles of a given species formed or destroyed
<|> = (11-70)
moles of photons absorbed by the system
The quantum yields for photochemical reactions in the solution phase
exhibit two properties which greatly simplify their use:
• The quantum yield is less than or equal to one.
t The quantum yield is independent of the wavelength of the
absorbed photons.
Although exceptions to these rules exist, they are rare for photochemical
reactions in the aquatic environment.
Environmental conditions influence photolysis quantum yields.
Molecular oxygen acts as a quenching agent (see Figure 11-13) in some
photochemical reactions, reducing the quantum yields (Wolfe ei aj_., 1978).
In other cases, it has no effect or may even be a reactant. In any case,
rate constant and quantum yield measurements should be performed in water
with oxygen concentrations representative of environmental conditions.
137
-------
PHOTOCHEMICAL PATHWAYS OF AN EXCITED MOLECULE
A0 + heat
Internal
conversion
hv
MHHi
Absorption
A. + hi»'
0
Flourescence
Intersystem crossing
.Quenching
Chemical reaction
Chemical reaction
AQ — ground state of reactant molecule
A* — excited state
QQ — ground state of quenching molecule
Q* — excited state
FIGURE 11-13
PHOTOCHEMICAL PATHWAYS OF AN EXCITED
MOLECULE, EXCITED MOLECULES DO NOT
ALWAYS CHEMICALLY REACT,
138
-------
Suspended sediments also influence rates of photolysis. Not only do
suspended sediments increase light attenuation, but they change the
reactivity of compounds sorbed on them (Miller and Zepp, 1979). Sorption
may either increase or decrease a compound's reactivity depending on the
reaction it undergoes. This effect, however, is of secondary importance in
comparison to the increase in light attenuation by the suspended sediments
(Burns et a/L, 1981). Thus, the effects of sorption will be neglected.
Chemical speciation also effects rates of photolysis. Different forms
of an organic acid or base may have different quantum yields, as well as
absorptivities, causing the apparent photolysis rate of the compound to vary
with pH. The possibility of this should be kept in mind when the pK of a
a
photolyzing compound is 7 _+ 2. Except where stated otherwise, data
contained herein may be assumed independent of pH over the range of values
observed in natural waters.
Photochemicaly initiated reactions may show a temperature effect
depending upon the actual mechanisms involved. General methods for
predicting this effect have yet to be developed. Users of this screening
manual should assume thermal effects on photolysis to be negligible.
Quantum yields vary over several orders of magnitude depending on the
nature of the molecule which absorbs light and the nature of the reactions
it undergoes. The two major classes of photochemical reactions of interest
in the aquatic environment are direct and sensitized photolysis. A closer
examination of each reaction type follows.
2.5.2.2.3.2 Direct Photolysis
Direct photolysis occurs when the reacting molecule itself directly
absorbs light. The excited molecule can undergo various types of reactions,
including fragmentation, reduction, oxidation, hydrolysis, acid-base
reaction, addition, substitution, isomerization, polymerization, etc.
Figure 11-14 shows examples of the reactions undergone by three toxic
substances which directly photolyze.
139
-------
a)
OCH2COR
Cl
OCH2COR
b)
Sunlight
+ R - C
c)
Cl H
Sunlight
Cl
Cl
Cl
FIGURE
DIRECT PHOTOCHEMICAL REACTIONS OF (A) 2,4-D ESTER,
(B) BENZ(A)ANTHRACENE, AND (c) PENTACHLOROPHENOL,
140
-------
The quantum yield for the direct photolysis, , of a compound is a
d
constant defined as follows:
— 1
dt/
d dt ad
where C is the concentration of the compound and I is the rate at which
ad
the compound absorbs light. Table 11-28 lists several disappearance quantum
yields for direct photolysis of aquatic pollutants.
By comparing molecular absorption spectra with the spectral
distribution of sunlight, it is possible to determine whether or not a
compound may directly photolyze. Benzene, as shown in Figure II-15a, does
not directly photolyze because it does not absorb light above 275 nm.
Naphthacene, shown in Figure II-15b, does directly photolyze because of its
strong absorptivity in the sunlight region of the spectrum. Humic acids,
Figure II-15c, by virtue of their absorption of sunlight may initiate
indirect, or sensitized, photochemical reactions.
2.5.2.2.3.3 Sensitized Photolysis
Sunlight can cause the degradation of aquatic pollutants by means other
than direct photolysis. A light-absorbing molecule can transfer its excess
energy to an acceptor molecule causing the acceptor to react as if it had
absorbed the radiant energy directly. This reaction mechanism, known as
photosensitization, contributes to the degradation of aquatic pollutants
when suitable light absorbing substances, or photosensitizers, are present.
2,5-Dimethylfuran is an example of a compound which degrades by sensitized
photolysis. It does not react when exposed to sunlight in distilled water
but degrades rapidly in waters containing natural humic acids (Zepp ejtal.,
1981a).
Numerous substances, including humic acids, titanium dioxide, and
synthetic organic compounds, can sensitize photochemical reactions. But,
most potential sensitizers occur at such low environmental concentrations
that they have negligible effects on photolysis rates. Humic acids, the
naturally occurring by-products of plant matter decay, frequently attain
141
-------
Table 11-28
DISAPPEARANCE QUANTUM YIELDS, d FOR DIRECT PHOTOLYSIS
Compound
Polycyclic Aromatic Hydrocarbons
Naphthalene
1-Methyl naphthalene
2-Methyl naphthalene
Phenanthrene
Anthracene
9-Methyl anthracene
9, 10-Dimethyl anthracene
Pyrene
Fluoranthrene
Chrysene
Naphthacene
Benz(a)anthracene
Benz(a)pyrene
2,4-D Esters
Butoxyethyl ester
Methyl ester
Carbaryl
N-Nitrosoatrazine
Trifluralin
DMDE
(jjj Reference
.015
.018
.0053
.010
.0030
.0075
.0040
.0021
(313 nm) .00012
(366 nm) .000002
.0028
.013
.0033
.00089
.056
.031
.0055
.30
.0020
.30
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
c
d
d
d
References:
aZepp and Schlotzhauer (1979)
bZepp et al_. (1975)
Slolfe et al_. (1978)
dZepp and Cline (1977)
142
-------
CO
fD
-S
n>
3
O
ro
00 to CO
O T3 C
3- fD 3
3 O —i
c-t- "S U3
N CU 3-
fD » r+
-s
: en
-o
: fD
10 • n
fa »
(/)
CO
o c
-h -S
3
o >
kjQ jfD
0)|t-t-
_i.| Cu
O |—i
O
o -~-
3 i—
"O to
O OD
CO
fD
rc 3
C N
3 ro
— '• 3
O fD
CD QJ
O 3
_i. Q.
Q.
3
CO O)
-O -O
fD 3-
O rt
o
ro
3
fD
extinction coefficient (cm-1 gC//l"
S 8
Solar Irradiance (photons/sec'nmt
o
o
m
X ~T3 O
a: in o
•—02
o -H TJ
:E o o
z N
•— m
—I v
O
m
o
o
3:
T3
O •—
m
o
—l
i—i
o
CO
H Z O
m o n:
— no
z v_x m
O V)
33 O
m co «—i
O C= 33
H td m
co o
o
m
r~ o z: -*
-< o m co
2: o
-o T> 3=- 2
n: o co
o c: co o
H z o -n
O O 33
r~ -o GO
-< s: H o
N :r •— r~
m i—i o J>
s~^ GO •—i
CO O ~D 33
v~-x O m 33
m o >
> CO H C3
o n
~n m
CD
c
33
m
e, molar extinction coefficient (cm~'(mole/£)~t)
CO
c
Solar Irradiance photon s/sec/nm
(, molar extinction coefficient
Solar Irradiance tphotons/sec/nm)
-------
concentrations of 1-10 mg as carbon per liter in natural systems. Humic
acids strongly absorb sunlight with wavelengths shorter than 500 nm,
as the absorption coefficients for dissolved organic carbon, aDQC> in
Table 11-27 indicate.
The quantum yield for photosensitized reactions, , is defined in a
manner similar to the quantum yield for direct photolysis:
(H-72)
where C is the concentration of the pollutant and I is the rate of light
as
absorption by the sensitizing molecule. The quantum yield for sensitized
photolysis, however, is not constant but depends on the pollutant
concentration, such that:
4> = Qs" C (11-73)
where Q is a constant. This is due to the fact that the probability of the
sensitized molecule donating its energy to a pollutant molecule is
proportional to the concentration of the pollutant molecule. Published
values of Q are very rare. Zepp e_t cfl_. (1981b) report a Q of 19
(mol/1) for the photosensitized ozidation of 2,5-dimethylfuran.
2.5.2.3 Computing Environmental Photolysis Rates
The overall rate at which a pollutant photolyzes in the aquatic
environment is the sum of the rates of direct and sensitized photochemical
reactions. At the low pollutant concentrations observed in the environment,
the rates of both direct and sensitized photolysis are proportional to the
concentration of the pollutant. Thus, photolysis follows a first-order
rate law:
— = -k • C (11-74)
dt P
where
144
-------
k' = overall photolysis rate constant, day
= k + k
d s
k = direct photolysis rate constant, day
d
k = sensitized photolysis rate constant, day
Due to the complexity of the units for the parameters in the photolysis
section, it is essential that the user employ the specified units in each
equation. All resulting first-order photolysis rate constants have units of
day .
The determination of rate constants for direct and sensitized
photolysis is the subject of the remainder of this section.
Section 2.5.2.3.1 includes a derivation of the equations for k and k .
Sections 2.5.2.3.2 and 2.5.2.3.3 describe how to calculate these constants
on the basis of near surface rate constants or molecular absorption spectra.
2.5.2.3.1 Derivation of Rate Constant Equations
2.5.2.3.1.1 Direct Photolysis
The rate at which a compound directly photolyzes is proportional to the
rate at which it absorbs light. The rate of light absorption by a dissolved
substance in natural waters is (Miller & Zepp, 1979):
\
J 2.3-j-e(X) • C(z) • D(z)- W(X) -e'K(A) "z dAdz (H-75)
where
-1 -1
I = rate of light absorption, einstein 1 day
ad
145
-------
Z = mixed depth of water body, m
A = 500 nm
1
A £300 nm
o
-16 3 -1 -1
j = conversion factor = 1.43 x 10 mole-cm -sec-l -day
= base 10 molar extinction coefficient of pollutant,
1 mol cm
C = concentration of pollutant, mol/1
D = radiance distribution function
-2 -1 -1
W = photon irradiance near the surface, photons cm sec nm
K = diffuse light attenuation coefficient of the water, m
The expression for the direct photolysis rate constant, which can be
derived from Equations (11-71), (11-74), and (11-75) is:
r i P'K'Z
k. = 2.3-j-i. -D'/ e-W-ll* . dA (11-76)
/ K-Z
Equation (11-76) incorporates the assumption that C, K, and D are
independent of depth. Thus, it will overpredict photolysis rates if the
pollutant is not distributed evenly throughout the water column.
146
-------
2.5.2.3.1.2 Sensitized Photolysis
The rate at which a compound decays through sensitized photolysis is
proportional to the rate at which sensitizing molecules absorb light. The
rate at which sensitizers absorb light in the aquatic environment is:
Z X
1 r r (H-77)
las = IJ J J • a U) • C (z) • D(z) • W(X) • e "
dAdz
b b
A.
0
where
I = rate of light absorption by sensitizers, einstein 1 day
clS
a = base e absorption coefficient of the sensitizer,
s _i _i
e.g. 1 mg-DOC cm
C = concentration of sensitizer, e.g. mg-DOC/1
The rate constant for sensitized photolysis of a compound, k , is then:
s
Al
ks = J'-Cs ' D ' Qs' / as ' w • ] " e -dA (n
A K*Z
o
Equation (11-78) includes the assumptions that GS, K, and D are independent of
depth and that Q is independent of wavelength
2.5.23.2 Use of Near Surface Rate Constants
Experimental data for direct photolysis are generally reported as near
surface rate constants, as in Table 11-29. Near the surface of a water body
(K-z <0.2), the mean irradiance is approximately equal to the surface
irradiance. This fact permits Equation (11-76) to be simplified to the
following expression which defines the near surface rate constant, k :
do
147
-------
Table 11-29
NEAR-SURFACE DIRECT PHOTOLYSIS RATE CONSTANTS
Compound
Polycyclic Aromatic Hydrocarbons
Naphthalene
1 -Methyl naphthalene
2-Methylnaphthalene
Phenanthrene
Anthracene
9-Methyl anthracene
9, 10-Dimethylanthracene
Pyrene
Fluoranthrene
Chrysene
Naphthacene
Benzo(a)pyrene
Benzo( a)anthracene
Carbamate Pesticides
Carbaryl
Propham
Chlorpropham
Phthlate Esters
dimethyl ester
diethyl ester
di-n-butyl ester
di-n-octyl ester
di-(2-ethylhexyl) ester
2,4-D Esters
butoxyethyl ester
methyl ester
Hexachl orocycl opentadi ene
Pentachlorophenol (anion)
3,3 '-dichlorobenzi dine
N-ni trosoatrazine
Trifluralin
DMDE(l,l-bis(p-methylphenyl 1)-
2,2-dichloroethylene)
k l} 2) *3)
do , I i] \
1 °
(day" ) (lanqleys/day) (no) Ref.
9
.23
.76
.31
2.0
22.0
130.0
48.0
24.0
.79
3.8
490.0
31.0
28.0
.32
<.003
<.006
5x10"'
5x10" 3
5x10" '
5xlO"3
5xlO"3
.050
.030
94.
.46
670.
300.
30.
17.
2100
2100
2100
2100
2100
2100
2100
2100
2100
2100
2100
2100
2100
2100
740
740
600
600
600
600
600
420
420
540
600
2000
1800
1800
2200
310 a
312 a
320 a
323 a
360 a
380 &
400 a
330 a
a
320 a
440 a
380 a
340 a
313 b
c
c
d
d
d
d
d
e
e
f
318* f
280-330* f
9
g
g
References:
Notes:
1)
2)
3)
*
Parenthetic comments after name of compound indicate when the form
of the compound undergoing photolysis is something other than the
neutral form.
Estimated Solar Flux - usually high estimates to
photolysis rates.
Wavelength of maximum sunlight absorption.
Indicates the maximum of the absorption spectrum
give conservative
is used.
a) Zepp and Schlotzhauer (1979)
b) Zepp (1978)
c) Wolfe et al_. (1978)
d) Wolfe et a\_. (1980)
e) Zepp et al_. (1979)
f) Callahan e_t aj_. (1979)
g) Zepp and Cline (1977)
148
-------
kdo
l
o ' j 'f e • W * dA (H-79)
A
0
where
k = near-surface direct photolysis rate constant, day
do
D = radiance distribution near the surface (approximate value =
° 1.2)
According to Equation (11-79), the near surface rate constant is
independent of the properties of the water it is measured in, except for the
small variation in D . Thus, when the difference in solar irradiance
o
between the experimental and environmental conditions is accounted for, the
user can apply a near surface rate constant to other bodies of water using
the following expression:
k - k l D
d " do' '
where
I = total solar radiation (langleys/day)
I = total solar radiation under conditions at which k was
o do
measured (langleys/day)
A* = wavelength of maximum light absorption, i.e. wavelength where
the product e(A)-w(A) is greatest.
This approximate expression is valid if the following assumptions are
sufficiently accurate: 1) the solar irradiance at a wavelength is a
constant fraction of the total solar irradiance (Park ^t jfL , 1980) and
2) the light attenuation coefficient, K, is constant over the range of
wavelength that the compound absorbs solar radiation at high rates
(Burns et a!., 1981).
149
-------
Although it is possible to derive a similar expression for sensitized
photolysis, variation in the absorptivity and reactivity of natural humic
substances make extrapolations based on the concentration of dissolved
organic carbon subject to large errors. An approach taken by Zepp (1980)
was to correlate the sensitized photolysis rate constant with the absorbance
of a solution at 366nm. Such an empirical relationship was found for
2,5-dimethylfuran:
log k = .671og a - 1.15 (11-81)
y so 366
where
a = absorbance of solution at 366nm
366
k = near surface rate constant, day cm (I =1 langley/day).
so o
At present, data on sensitized photolysis are difficult to obtain. The
planner should be aware of its potential significance even if it is not
possible to estimate rates at this time.
2.5.2.3.3 Evaluation of Rate Constant Integrals
When both the absorption spectrum, e(A) or a (A), and the quantum
yields, or Q , are available, it is possible to evaluate the integrals in
d s
Equations 11-76 and 11-78 numerically, as shown below:
s . r - ts (n-83)
i
where
i = index of wavelength interval
W = W • AX
150
-------
The user may obtain information necessary to evauate these expressions
from the following sources:
• W - Table 11-26.
t K - Equation (11-68) and Table 11-27 or Equation (11-69).
• D - Assign a value between 1.2 and 2 as follows: 1.2 for
very clear waters, 1.6 for typical rivers, 2 for
extremely turbid waters.
• e - Spectroscopy reference works, e.g. Stadler U.V. Spectra
or U.V. Atlas of Organic Compounds.
• - Literature or Table 11-28.
d
• a - For reactions sensitized by humic acids, use an_- in
s
Table 11-27.
• Q - Literature (rarely available).
EXAMPLE II-7
Computation of Photolysis Rate Constants
Compute the mean annual photolysis rate constant for the pesticide
carbaryl in a hypothetical river near Fresno, California. Use both the
evaluation of integral and near surface rate constant methods described
above. Assume the following physical and chemical parameters apply to the
river:
Mean Depth = 2 m
Suspended Sediments - 10 mg/1
Humic Acid = 2 mg-DOC/1
Chlorophyll j. = 0 mg/1
151
-------
Zepp (1978) reported a quantum yield, ,, of .0060 and the following
absorptivities, e, for carbaryl :
Wavelength (nm) Absorptivity (M cm ) __
300 918
310 356
320 101
330 11
A. Near Surface Rate Constant Method
Table 11-29 contains the following information regarding carbaryl:
k = .32 day
d
I = 2100 langleys/day
^ = 313 nm
According to Figure 11-12, the mean annual solar irradiance at Fresno,
California is 450 langleys/day.
Assume that the radiance distribution function under reference, D ,
o
and environmental, D, conditions have values of 1.2 and 1.6 respectively.
To calculate the light attenuation coefficient at the wavelength of
maximum light absorption, 313 nm, we use Equation (11-68) and the data in
Table 11-27, at 310 nm;
K = 1.6 (.105 + 67«0 + 5.41 • 2 + .35 • 10) = 23.1m"
When the water absorbs nearly all of the incident radiation, i.e. kZ >_
3, the following approximation is valid:
i Q-KZ i
1 - e % 1
KZ % KZ
This approximation can be applied to Equation (11-80) and Equation (11-82).
It both simplifies the calculations and eliminates the dependence of the
rate constant on the radiance distribution function, D, in cases where the
152
-------
light attenuation coefficient is calculated from D, as in this example. In
such a case, the user's choice of a value of D does not affect the result.
Using this approximation in Equation (11-80), the mean photolysis rate
constant is computed to be:
k , = .32 day
-1 450 1.6
1
2100 1.2 23.1 • 2
- 2.0 x 10"3 day"1
This example demonstrates the significant difference, 100 fold in this
case, which may exist between near surface and mean photolysis rate
constants. The strong attenuation of light by the river water was the
primary cause of the reduction in rates.
B. Evaluation of Integrals
The absorption data for carbaryl indicate that we need to concern
ourselves only with light of wavelength 300-330 nm in order to determine a
mean rate constant.
First, we assume that D has the same value as above, 1.6. Then,
we compute the light attenuation coefficients using Equation (11-68) and
the data in Table 11-27.
A
(nm)
300
310
320
330
D
1.6
1.6
1.6
1.6
h
(irf1)
,141
.105
.0844
.0678
+ (a
V DOC
[(mg/1 TV1]
6.25
5.41
4.68
4.05
DOC/
(mg/1)
2
2
2
2
• (•„ •
[(mg/1 TV1]
.35
.35
.35
.35
(mg/1)
10
10
10
10
(m"1)
•*.. mimmmmim
25.8
23.1
20.7
18.7
153
-------
Table 11-26 lists the photon spectral irradiance, W, at a reference
total sloar flux, I , of 540 langleys/day. The local solar flux, as in part
o
A, is 450 langleys/day.
Next, evaluate the sum indicated in Equation (11-82).
Since KZ >3 for all wavelengths of interest, use the approximation
discussed in part A.
X e W x 10"14 ...
1 i ? c-W
(nm) (M cm"1) (photons/cnT/s) (K-Z) K Z
300 918 .0303 51.6 .539 x 10 14
310 356 .388 46.2 2.99 x 10 '14
320 101 1.13 41.4 2.76 x 10 14
330 11 1.81 37.4 .532 x 10 14
E = 6.82 x 10
i
14
Given that the quantum yield is .006, the mean photolysis rate constant
can be computed using Equation (11-82) and the above information:
k = 2.3 • 1.43 x 10"16 • — - .0060 • 1.6 • 6.82 x 1014
d 540
= 1.8 x 10"3 day"1
The small difference between the rate constants calculated in parts A
and B is due to the difference in the reference solar intensities. The
assumption made here that the spectral distribution of solar energy is
independent of intensity is only approximately true. Consequently, the
greater the discrepancy between the reference and local solar intensities,
the greater the error in rate constants that can be expected. When the
local exceeds the reference intensity, the actual rate constant is probably
higher than the calculated value. When the reference exceeds the local
intensity, the actual rate constant is probably lov.'er than calculated.
rjD OF EXAMPLE 11-7 —
154
-------
2.5.3 Hydrolysis
Some toxic compounds can be altered by direct reaction with water. The
chemical reaction of a compound with water is called hydrolysis. Typically
in hydrolysis reactions hydroxide replaces another chemical group.
An example hydrolysis reaction for a toxic organic compound is given
below:
Carbaryl
H20
Water
OH'
+ H2 NCH3 + C0:
a-Naphthanol + Methyl ami ne + Carbon
Dioxide
Generalized hydrolytic reactions of organic compounds are presented in
Table 11-30.
Hydrolysis reactions alter the reacting molecules but do not always
produce less noxious products. For example the more toxic 2,4-D acid is
produced from the hydrolysis of certain 2,4-D esters. Alternatively the
hydrolysis of carbaryl (shown above) produces less toxic products,
i.e. a-naphthanol and methylamine.
Hydrolysis products may be more or less volatile than the original
compound. Hydrolysis products which ionize may have essentially zero
volatility depending upon pH. Hydrolysis products are generally more
readily biodegraded than the parent compounds, although there are some
exceptions.
Hydrolysis reactions are commonly catalyzed by hydrogen or hydroxide
ions. This produces the strong pH dependence often observed for hydrolysis
reactions. Examples of this dependency are shown in Figure 11-16, where the
logarithms of reaction rate constants (k ) are plotted versus pH. The
H
155
-------
TABLE 11-30
GENERALIZED HYDROLYTIC REACTIONS OF ORGANIC COMPOUNDS
REACTANT REACTION CONDITIONS
PRODUCTS
CARBOXYLIC ACID ESTERS ACIDIC, NEUTRAL,
Q BASIC
R-ct
AMIDES ACIDIC, BASIC
R-c(
XN-R'
1
H
CARBAMATES ACIDIC, BASIC
H
R-N
C-O-R'
0
ORGANOPHOSPHATES BASIC (Acioic,
(AND DERIVATES) NEUTRAL)
Q
RO-P— OR
i
OR
HALOGENATED ALKANES NEUTRAL, BASIC
R
i
,C~\
R' R-
CARBOXYLIC ACID + ALCOHOL
o
R-C7 + R'OH
XOH
CARBOXYLIC ACID + AMINE
/ H\
R-cf + A\
XOH ,/ V
AMINE + ALCOHOL + CARBON DIOXIDE
R_N/H R'OH C02
\
PHOSPHATE DIESTER + ALCOHOL
0
RO — P — OH ROH
i
OR
ALCOHOL + HALIDE ION
R _
R'— C— OH X
i
1
R"
SOURCE: i.J. TINSLEY, CHEMICAL CONCEPTS IN POLLUTANT BEHAVIOR, J. WILEY, NEW YORK (1979).
156
-------
2 V0-CH2-CH2-0-CH
O Parathion
0 Carbaryl
O Chloromethane
A 2.4-D (2-butoxyethyl
ester)
PH
FIGURE 11-16 pH DEPENDENCE OF HYDROLYSIS RATE CONSTANTS
157
-------
hydrolysis rate of carbaryl can be seen to increase logarithmically with pH.
The rate at pH = 8 is ten times that at pH = 7 and 100 times that at pH = 6.
The hydrolysis rate of parathion is high at low pH values, reaches a minimum
at pH = 6, and then increases with increasing pH. The hydrolysis rate of
chloromethane shows minimal dependence on pH over the range presented.
Adsorption can also influence hydrolysis rates. Adsorption of an
organic molecule protects it from acid or base catalyzed hydrolysis (Wolfe,
1981). The amount of adsorption can be predicted using the principles
presented in Section 2.3.2.
Microbially mediated hydrolysis reactions are responsible for the
breakdown of many complex molecules, including natural polymers such as
cellulose. Microorganisms catalyze hydrolysis reactions in the process of
using organic compounds as energy and/or carbon sources. In cometabolism
microbes may hydrolyze toxic organic compounds to hasten their removal from
cell protoplasm. Microbially mediated processes are covered under the
general heading of biodegradation in Section 2.5.1. Here only abiotic
hydrolysis is treated.
Abiotic hydrolysis reactions are represented by rate expressions which
are first order in the concentration of the compound being hydrolyzed:
R = — = -k C (11-84)
at H T
-1 -1
,_. jlvsis. mole 1itet
where R = the rate of hydrolysis, mole liter sec or
yg liter sec
k = specific hydrolysis rate constant, sec"1
H
C = the dissolved plus sorbed phased concentration of compound C,
mole liter or yg liter
In the literature k is typically defined as:
H
158
-------
k = k + k [H ]+ k [Oh"] (11-85)
H n a b
In this document the specific hydrolysis rate constant, k,,, is defined
h
to include the effects of adsorption:
4- - \~1
(11-86)
r / + - \i
k = k + a (k [H ]+ k [OH ]}
HLn w \ a b /_
where k = the neutral hydrolysis rate constant, sec
n
a = the decimal fraction of the total amount of compound C
w
which is dissolved (Calculation procedures in Section 2.3.2)
k = the acid catalyzed hydrolysis rate constant, liter
a -1-1
mole sec
[H 1 = the molar concentration of hydrogen ion, mole liter
k = the base catalyzed hydrolysis rate constant, liter
mole sec
[OH ] = the concentration of hydroxide ion, mole liter
[of] - io(pH-'V -- io°pH-14)
Equation 11-86 is a convenient definition of k because specific rates
constants which act on the dissolved and total concentrations do not have to
be used separately.
Values for the three rate constants k , ka, kb for selected compounds
are presented in Table 11-31. Additional values can be found in the
literature (e.g. Mabey and Mill, 1978). The three constants can also be
determined by simple laboratory tests.
Water body pH values must be obtained for hydrolysis reactions which
are pH dependent (i.e. those for which k 4 0 and/or k ^ 0). It should be
noted that in poorly buffered waters (alkalinity £50 mg/1 as CaCO ), pH
values may change by 1-2 units daily due to natural processes alone. In
159
-------
TABLE 11-31
HYDROLYSIS RATE PARAMETERS AND ESTIMATED ENVIRONMENTAL
HYDROLYSIS RATES
Compound
Pesticides
Endnsul fon
Hept-ichlor
C-if baryl
Prophan
Chlorpropham
2,4-D(2-Butoxyethyl ester)
2,4-D(Methyl ester)
Parathion
Phosmet
Dial i for
Malathion
Captan
Atrazine
Methoxychlor
Haloqenated Hydrocarbons
Chloromethane
Bromomethane
Chloroethane
Dlchloromethane
Tr ichloromethane
Brouiodi chl oromethane
D ibromochloromethane
Tri bromome thane
Hexachlorocyclopentadiene
Haloqenated Ethers
Bis{chloromethyl) ether
2-Chloroethyl vinyl ether
Phthalate Esters
Dimethyl ester
Oiethyl ester
Ci-n-butyl ester
Di-n-octyl ester
Di(2-ethylhexyl ) ester
Monocyc'Hc Aromatics
Pentachl orophenol
ka(M"1day"1) k^day'1) kb(M"1day"1)
3.3xl05
7 7 ?
4.3xl05
.66
1.7
1.7 - 2.6xl06
1.5xl06
1.3xl02 3.6xlO"3 2.46xl03
7 ? ?
777
7 77
1.6 4.9xl07
3.4 6.6
2.6xlO"3 31.
2.1xlO"3 .53
3.5xlO"2 12.
l.BxlO"2
2.8xlO"6 1.8xlO"3
6.0
1.4xl03
69.
28.
4.8xlO"2
1.6xl03
3.8xl02
1. - 6.0xl03
1. - 1.9xl03
1. - 9.1xl02
1. - 1.4xl03
1. - 9.6
l.lxlO4 5.8xlO"3 3.3
U n v i ro
Hydro lysi
kH(day"')
3.5xlO"2
.7
4.3xlO"2
6.6xlO"8
1.7xlO"7
.26
.15
3.9xlO"3
2.3
1.2
6.6xlO"2
5.6
6.6
2.6xlO"3
2.1xlO"3
3.5xlO"2
l.BxlO"2
2.8xlO"6
e.cxio"7
1.4x10"°
6.9xlO"6
2.8xlO"6
4.8xlO"2
1.6xl03
3.8xlO"5
6.0xlO"4
1.9xlO"4
9.1xlO"5
1.4xlO"4
9 SxlO"7
6.9xlO"3
nmental
s Rates (pH=7)
todays)
21
1.
16.
l.lxlO7
4.0xl06
2.7
4.6
l.SxlO2
.30
58
11.
.13
.10
2.7xl02
3.4xl02
20.
38.
2.6xl05
I.SxlO6
5.0xl04
l.OxlO5
2.5xl05
14.
4.5xlO"4
I.SxlO4
1.2xl03
3.7xl03
7.6xl03
4.9xl03
7.2xl05
l.OxlO2
Ref.
Temp.
27
30
27
27
27
28
28
7
20
20
20
27
25
25
25
25
25
25
25
25
25
25
25
20
25
30
30
30
30
30
7
Ref.
a
a
b
b
b
c
c
d
e
e
e
f
f
f
f
f
f
f
f
f
f
f
a
a
a
9
9
9
9
9
d
"?" denotes rate parameter not given and not estimable from data in reference
"-" denotes zero or very small rate parameter
References:
* Callahan e_t al. (1979)
Wolfe e_t JKT1978)
5 Zepp et al_. (1975)
a Park et, a]_. U980)
! Tinsley 11979)
Habey and Mill (1978)
9 Wotfe et al. (1980)
160
-------
these cases either additional data must be gathered to characterize the
system's pH regime or conservatively low values of k must be used.
H
EXAMPLE 11-8
A biodegradation rate constant, k for the fungicide Captan has been
given as 0.5 per day. Compare this with the abiotic hydrolysis rate
constant, k , at pH = 8.4, a temperature of 25 C, and with 90 percent of the
compound adsorbed on suspended matter. Values for k , k , and k can be
found in Table 11-31.
k = 0
a 7-1
k = 4.9 x 10 day
b -1
k = 1.6 day
n
f / + -VI
k = a k [H ] + k [OH ] + k
HlwVa b /I n
[OH'] -- !0PH-14 - 108'4-14 . ID'5-6 - 2.51 x 10-6
thus
kH =[(1.0-0.9)-(4.9 x 10 x 2.5 x 10~6)1
= 12.3+ 1.6 = 13.9 day
x 10 )| + 1.6
-1
Comparing k to k ,
H B
13.9
A t = 27.8
Comparison of k with k for the above situation shows that the abiotic
H B
hydrolysis rate is about 28 times faster than the biodegradation rate.
Biodegradation could be neglected here with minimal effect on the results.
— END OF EXAMPLE II-8—
161
-------
REFERENCES
Alexander, M. 1980. Biodegradation of Toxic Chemicals in Water and Soil.
In Dynamics, Exposure, and Hazard Assessment of Toxic Chemicals
~[R. Haque, editor) Ann Arbor Science. Ann Arbor, Michigan.
Baughman, G.L., D.F. Paris, and W.C. Steen. 1980. Quantitative Expression
of Biodegradation Rate. _I_n Biotransformation and Fate of Chemicals in
the Environment, A.W. Maki, K.L. Dickson, and J. Cairns, Jr. (editors)
American Soc. Microbiol., Washington, DC. pp. 105-111.
Bird, R.B., W.E. Stewart, and E.N. Lightfoot. 1960. Transport Phenomena.
John Wiley and Sons.
Brown D.S. and E.W. Flagg. 1981. Journal Environmental Quality.
10(3):382-386.
Burns, L.A., D.M. Cline, and R.R. Lassiter. 1981. Exposure Analysis
Modeling System (EXAMS): User Manual and System Documentation. Draft.
Environmental Research Laboratory, U.S. EPA, Athens, Georgia. 443 pp.
Callahan, M.A., M.W. Slimak, N.W. Gable, I.P. May, C.F. Fowler, J.R. Freed,
P. Jennings, R.L. Durfee, F.C. Whitmore, B. Maestri, W.R. Mabey,
B.R. Holt, and C. Gould. 1979. Water-Related Environmental Fate of
129 Priority Pollutants. Volumes I and II. Prepared for EPA by
Versar, Inc., Springfield, Virginia. Available from NTIS.
PB80-204373.
Chiou, C.T., V.H. Freed, D.W. Schmedding, and R.L. Kohnert. 1977.
Partition Coefficient and Bioaccumulation of Selected Organic
Compounds. Environ. Sci. Tech. 11:475-478.
Dickson, A.G. and J.P. Riley. 1979. The Estimation of Acid Dissociation
Constants in Seawater Media from Potentiometric Titrations with Strong
Base. I. The Ionic Product of Water -K . Mar. Chem. 7:89-99
w
Dilling, W.L., N.B. Terfertiller, and G.J. Kallos. 1975. Evaporation Rates
and Reactivities of Methylene Chloride, Chloroform,
1,1,1-Trichloroethane, Trichloroethylene, Tetrachloroethylene and Other
Chlorinated Compounds in Dilute Aqueous Solutions. Environ. Sci.
Tech. 9:833-838.
Gilbert, P.A. and C.M. Lee. 1980. Biodegradation Tests: Use and Value.
j_n Biotransformation and Fate of Chemicals in the Aquatic Environment
A.W. Maki, K.L. Dickson and J. Cairns, Jr. (editors). American
Society for Microbiology, Washington, D.C. pp. 34-45
Glasstone S. 1946.
New York.
The Elements of Physical Chemistry. D. Van Nostrand,
Hambrick, G.A., R.D. DeLaune and W.H. Patrick, Jr. 1980. Effect of
Estuarine Sediment pH and Oxidation-Reduction Potential on Microbial
Hydrocarbon Degradation. Applied and Environmental Microbiology,
Volume 40, No. 2, pp. 365-369.
162
-------
Hague, R. 1980. Dynamics, Exposure and Hazard Assessment of Toxic
Chemicals in the Environment: An Introduction. Jjn Dynamics, Exposure
and Hazard Assessment of Toxic Chemicals (R. Haque, editor) Ann Arbor
Science. Ann Arbor, Michigan.
Hassett, J.J., J.C. Means, W.L. Bonwart, and S.G. Wood. 1980. Sorptiori
Properties of Sediments and Energy-Related Pollutants.
U.S. Environmental Protection Agency, Athens, Georgia.
EPA-600/3-80-041.
Herbes, S.E. and L.R. Schwall. 1978. Microbial Transformation of
Polycyclic Aromatic Hydrocarbons in Pristine and Petroleum-Contaminated
Sediments. Applied and Environmental Microbiology, Volume 35, No. 2,
pp. 306-316.
Karickhoff, S.W., D.S, Brown, and T.A. Scott. 1979. Sorption of
Hydrophobic Pollutants on Natural Sediments. Water Res. 13:241-248.
Keith, L.H. and W.A. Tel 1iard. 1979. Priority Pollutants. I. A
Perspective View. Env. Sci. Tech. 13(4):416-423.
Kirsch, E.J. and J.E. Etzel. 1973. Microbial Decomposition of
Pentachlorophenol. Journal Water Pollution Control Federation,
Volume 45, No. 2, pp. 359-364.
Larson, R.J. 1980. Role of Biodegradation Kinetics in Predicting
Environmental Fate, jji Biotransformation and Fate of Chemicals in the
Aquatic Environment, A.W. Maki, K.L. Dickson and J. Cairns, Jr.
(editors). American Society for Microbiology, Washington, D.C.
pp. 67-86.
Larson, R.J., G.G. Clinckemaillie, and L. VanBelle. 1981. Effect of
Temperature and Dissolved Oxygen on Biodegradation of
Nitrilotriacetate. Water Research, Volume 15, pp. 615-620.
Linsley, R.K., M.A. Kohler, and J.L.H. Paulus. 1979. Hydrology for
Engineers. McGraw-Hill.
Liss, P.S. 1973. Deep-Sea Research, Volume 20, pp. 221-238.
Mabey, W. and T. Mill. 1978. Critical Review of Hydrolysis of Organic
Compounds in Water Under Environmental Conditions. J. Phys. Chem.
Ref. Data, Volume 7, No. 2, pp. 383-415.
Mackay, D. and P.J. Leinonen. 1975. Rate of Volatilization of
Low-Solubility Contaminants from Water to Atmosphere. Environ. Sci.
Tech. 9:1178-1180.
Miller, G.C. and R.G. Zepp. 1979. Effects of Suspended Sediments on
Photolysis Rates of Dissolved Pollutants. Water Research 13:453-459.
Mills, W.B. 1981. Workshop Notes: Screening Analysis of Toxic and
Conventional Pollutants in Rivers. Washington, D.C.
163
-------
Moon, P. 1940. Proposed Standard Solar-Radiation Curves for Engineering
Use. J.F.I., Volume 230, No. 1379-23.
Paris, D.F., W.C. Steen, J.T. Barnett, and E.H. Bates. 1980. Kinetics of
Degradation of Xenobiotics by Microorganisms. ACS, Division of
Environmental Chemistry, Volume 20, No. 2, pp. 55-56.
Paris, D.F., W.C. Steen, G.L. Baughman, and J.T. Barnett, Jr. 1981.
Second-order Model to Predict Microbial Degradation of Organic
Compounds in Natural Waters. Applied and Environmental Microbiology,
Volume 41, No. 3, pp. 603-609.
Park, R.A., C.I. Connolly, J.R. Albanese, L.S. Clesceri, 6.W. Heitzman,
H.H. Herbrandson, B.H. Indyke, J.R. Loehe, S. Ross, D.D. Sharma, and
W.W. Shuster. 1980. Modeling Transport and Behavior of Pesticides and
Other Toxic Organic Materials in Aquatic Environments (PEST). Report
No. 7. Center for Ecological Modeling, Rensselaer Polytechnic
Institute, Troy, New York. 163 pp.
Pavlou, S.P. 1980. Thermodynamic Aspects of Equilibrium Sorption of
Persistent Organic Molecules at the Sediment-Seawater Interface: A
Framework for Predicting Distributions in the Aquatic Environment. Jji
Contaminants and Sediments, Volume 2, R.A. Baker (editor). Ann Arbor
Science, Ann Arbor, Michigan, pp. 323-332.
Perry, R.H. and C.H. Chi Hon. 1973. Chemical Engineers Handbook.
McGraw-Hill.
Pitter, P. 1976. Determination of Biological Degradability of Organic
Substances, Water Research, Volume 10, pp. 231-235.
Rathbun, R.E. and D.Y. Tai. 1981. Technique for Determining the
Volatilization Coefficients of Priority Pollutants in Streams. Water
Research. Volume 15, No. 2.
Reinbold, K.A., J.J. Hassett, J.C. Means, and W.L. Banwart. 1979.
Adsorption of Energy-Related Organic Pollutants: A Literature Review.
Environmental Research Laboratory, U.S. Environmental Protection
Agency, Athens, Georgia. EPA-600/3-79-086. 178 pp.
Robinson, N. (editor). 1966. Solar Radiation. Elsevier Publishing
Company, Amsterdam, London and New York. 347 pp.
Rossi, S.S. and W.H. Thomas. 1981. Solubility Behavior of Three Aromatic
Hydrocarbons in Distilled Water and Natural Seawater. Environmental
Science and Technology. Volume 15, No. 6, pp. 715-716.
Roubal, G. and R.M. Atlas. 1978. Distribution of Hydrocarbon-Utilizing
Microorganisms and Hydrocarbon Biodegradation Potentials in Alaskan
Continental Shelf Areas. Applied and Environmental Microbiology,
Volume 35, No. 5, pp. 897-905.
164
-------
Schnitzer, M. 1971. J_n Soil Biochemistry, Volume 2. Marcel Dekker, Inc.,
New York.
Schnoor, J.L. 1981. Assessment of the Exposure, Fate and Persistence of
Toxic Organic Chemicals to Aquatic Ecosystems in Stream and Lake
Environments. Part I. Lecture notes Water Quality Assessment of
Toxics and Conventional Pollutants in Lakes and Streams, a workshop
sponsored by U.S. Environmental Protection Agency, June 23-25, 1981 in
Arlington, Virginia. Civil and Environmental Engineering, Energy
Engineering Division, University of Iowa, Iowa City, Iowa. 31 pp.
Shamat, N.A. and W.J. Maier. 1980. Kinetics of Biodegradation of
Chlorinated Organics, Journal Water Pollution Control Federation,
Volume 52, No. 8, pp. 2158-2166.
Singer, S.J. and G.L. Nicholson. 1972. Science 175:723.
Smith, J.H., D.C. Bomberger, Jr., and D.L. Haynes. 1981. Volatilization
Rates of Intermediate and Low Volatility Chemicals from Water.
Chemosphere. 19:281-289.
Smith, R.C. and K.S. Baker. 1978. Optical Classification of Natural
Waters, Limnology and Oceanography, Volume 32, No. 2, pp. 260-267.
Smith, R.C. and K.S. Baker. 1981. Optical Properties of the Clearest
Natural Waters (200-800 nm). Applied Optics, Volume 20, No. 2,
pp. 177-184.
Smith, R.C. and J.E. Tyler. 1976. Transmission of Solar Radiation into
Natural Waters, Photochemical and Photobiological Reviews, Volume 1,
K.C. Smith, editor. Plenum Press, New York. pp. 117-155.
Spain, J.C., P.M. Pritchard, and A.W. Bourquin. 1980. Effect of Adaptation
on Biodegradation Rates in Sediment/Water Cores from Estuarine and
Freshwater Environments. Applied and Environmental Microbiology,
Volume 40, No. 4, pp. 726-734.
Steen, W.C., D.F. Paris, and G.L. Baughman. 1980. Effect of Sediment
Sorption on Microbial Degradation of Toxic Substances. lr\_ Contaminants
and Sediments, Volume 1, R.A. Baker (editor). Ann Arbor Science, Ann
Arbor, Michigan, pp. 477-482.
Stumm, W. and J.J. Morgan. 1981. Aquatic Chemistry, Second Edition.
John Wiley and Sons, New York. 780 pp.
Tabak, H.H., S.A. Quave, C.I. Mashni, and E.F. Barth. 1981.
Biodegradability Studies with Organic Priority Pollutant Compounds.
Journal Water Pollution Control Federation, Volume 53, No. 10,
pp. 1503-1518.
Tiedje, J.M. 1980. Fate of Chemicals in the Aquatic Environment: Case
Studies, Biotransformation and Fate of Chemicals in the Aquatic
Environment, A.W. Maki, K.L. Dickson, and J. Cairns, Jr. (editors).
American Soceity for Microbiology, Washington, D.C. pp. 114-119.
165
-------
Tinsley, I.J. 1979. Chemical Concepts in Pollutant Behavior. A Wiley -
Interscience Publication. John Wiley and Sons, New York, New York.
265 pp.
Turro, N.J. 1978. Modern Molecular Photochemistry. The Benjamin/Cummings
Publishing Company, Menlo Park, California. 628. pp.
U.S. Department of Commerce. 1968. Climatic Atlas of the United States.
U.S. Department of Commerce, Environmental Sciences Services
Administration, Environmental Data Service, Washinton, D.C.
U.S. Environmental Protection Agency. 1976. Quality Criteria for Water.
U.V. Atlas of Organic Compounds, Volumes 1-4. 1966-1971. Collaboration of
Photoelectric Spectrometry Group, London and Institut Fur Spektrochemie
und Angewandte Spektroskopie, Dortmund. Plenum Press, New York.
Ward, D.M. and T.D. Brock. 1976. Environmental Factors Influencing the
Rate of Hydrocarbon Oxidation in Temperate lakes, Applied and
Environmental Microbiology, Volume 31, No. 5, pp. 764-772.
Weast, R.C. and M.J. Astle (editors). 1980. CRC Handbook of Chemistry and
Physics. CRC Press. Boca Raton, Florida.
Wetzel, R.6. 1975. Limnology. W.B. Saunders. Philadelphia. 743 pp.
Wodzinski, R.S. and D. Bertolini. 1972. Physical State in Which
Naphthalene and Bibenzyl are Utilized by Bacteria. Applied
Microbiology, Volume 23, No. 6, pp. 1077-1081.
Wolfe, N.L., R.6. Zepp, and D.F. Paris. 1978. Carbaryl, Propham, and
Chloropropham: A Comparison of the Rates of Hydrolysis and Photolysis
with the Rate of Biolysis. Water Research, Volume 12, pp. 565-571.
Wolfe, N.L. 1980. Determining the Role of Hydrolysis in the Fate of
Organics in Natural Waters. In Dynamics, Exposure, and Hazard
Assessment of Toxic ChemicalsTR. Hague, editor). Ann Arbor Science.
Ann Arbor, Michigan.
Wolfe, N.L. 1981. Personal Communication.
Zepp, R.G. 1978. Quantum Yields for Reaction of Pollutants in Dilute
Aqueous Solution. Environmental Science and Technology, Volume 12,
No. 3, pp. 327-329.
Zepp, R.G. 1980. Assessing the Photochemistry of Organic Pollutants in
Aquatic Enviroments. J_n Dynamics, Exposure, and Hazard Assessment of
Toxic Chemicals (R. Haque, editor). Ann Arbor Science. Ann Arbor.
pp. 69-110.
Zepp, R.G., G.L. Baughman, and P.F. Schlotzhauer. 1981a. Comparison of
Photochemical Behavior of Various Humic Substances in Water: I.
Sunlight Induced Reactions of Aquatic Pollutants Photosensitized by
Humid Substances. Chemosphere. 10:109-117.
166
-------
Zepp, R.G., G.L. Baughman, and P.P. Schlotzhauer. 1981b. Comparison of
Photochemical Behavior of Various Humic Substances in Water: II.
Photosensitized Reactions. Chemosphere. 10:119-126.
Zepp, R.G. and D.M. Cline. 1977. Rates of Direct Photolysis in Aquatic
Environment. Environ. Sci. Tech. 11:359-366.
Zepp, R.G. and P.P. Schlotzhauer. 1979. Photoreactivity of Selcted
Aromatic Hydrocarbons in Water. j_n Polynuclear Aromatic Hydrocarbons,
P.W. Jones and P. Leber (editors). Ann Arbor Science Publishers, Inc.,
Ann Arbor, Michigan, pp. 141-158.
Zepp, R.G. and P.P. Schlotzhauer. 1981. Comparison of Photochemical
Behavior of Various Humic Substances in Water: III. Spectroscopic
Properties of Humic Substances, Chemosphere. Volume 10, No. 5,
pp. 479-486.
Zepp, R.G., N.L. Wolfe, J.A. Gordon, and G.L. Baughman. 1975. Dynamics of
2,4-D Esters in Surface Waters. Environmental Science and Technology,
Volume 9, No. 13, pp. 1144-1150.
167
-------
CHAPTER 3
WASTE LOADING CALCULATIONS
3.1 INTRODUCTION
This chapter outlines basic procedures that can be used to generate
estimates of diffuse (nonpoint) and point source loads. Loading functions
for the following pollutants will be considered for nonpoint sources:
• Sediment
• Nutrients (phosphorus and nitrogen)
• Organic matter
• Salinity in irrigation return flow
• Toxic organic pollutants
• Metals
Both cultivated and noncultivated agricultural land as well as urban areas
are addressed in this report. For agricultural lands, long term sediment
loads are calculated with the Universal Soil Loss Equation (USLE). This
method has been adopted for a number of reasons, principal among them being
the large data base that exists for the terms in the USLE. The modified
USLE is presented as a method for estimating single event conventional
pollutant loads. The loading of both nutrients and organic matter can be
quantitatively related to sediment loading. Thus, the discussion of
nutrients and organic matter logically follows the sediment loading
calculations. These selected water quality parameters were chosen for
inclusion in this report because they represent commonly occurring problems
of major concern to planners.
Salinity in irrigation return flow is important in many areas in the
arid western states. Considerable data are included in this report
especially for the Colorado River basin and the irrigated regions in
California (Section 3.2.8). Procedures are included for determining
electrical conductivity and sodium adsorption ratio (SAR).
168
-------
For urban areas two procedures are presented; the URS Urban Water
Quality Management procedure and the SWMM Level One Screening procedure. In
the former, solids loading rates are first calculated and then the loading
rates of other pollutants are related to them. Pollutants considered in
this section include BOD, phosphorus, nitrogen, coliforms, and heavy metals.
In the SWMM procedure separate and combined sewers are considered as well as
street sweeping efficiency. Single event procedures are also provided for
urban areas.
Next, typical point source pollutant loads for municipal and industrial
discharges are discussed. Whenever possible, however, local data should
always be used, if available, in lieu of the "typical" loadings given here.
Within each major section a subsection on toxic organic pollutants is
included. These sections cover the various ways in which toxicants
accumulate on watershed surfaces and provide procedures to estimate washoff
of toxicants in the sediment and water phases.
Each of the major divisions on nonpoint source calculations
(agricultural and urban areas) in this chapter is essentially independent of
every other. Accordingly, they can be used in any order. Within each
section, the calculations performed can be used in two different ways.
First, the magnitude of loadings can be compared for various alternatives
(e.g., different land use schemes) to ascertain the significance of the
changes. Second, the loadings can be used in calculations presented in
Chapter 4 to assess the water quality impacts of nonpoint source pollutants
on rivers and streams. These data can then be used to determine input of
nonpoint source pollutants to impoundments (Chapter 5) and estuaries
(Chapter 6), as appropriate.
In writing this chapter the following sources have been heavily and
freely drawn upon: "Loading Functions for Assessment of Water Pollution
from Nonpoint Source (McElroy _et jTL, 1976), "Water Quality Management
Planning for Urban Runoff" (Amy, et aj_., 1974), "Storm Water Management
Model Level I, Prel iminary Screening Procedures" (Heaney, et _al_., 1976),
"Predicting Rainfall Erosion Losses: A Guide to Conservation Planning"
(Wischmeier and Smith, 1978) and "A Mathematical Model for Estimating
169
-------
Pesticide Losses in Runoff" (Haith, 1980). Users should refer to these
references for further details concerning the methodologies.
3.2 NONURBAN NONPOINT SOURCE LOADS
3.2.1 Annual Sediment Loads
Sediment loading is defined in this report as the quantity of soil
material that is eroded and transported into the watercourse. Sediment
loading is dependent on (a) on-site erosion, and (b) delivery, or the
ability of runoff to carry the eroded material into the receiving waters.
The sediment loading function is based on the mechanisms of gross
erosion and sediment delivery. The Universal Soil Loss Equation (Wischmeier
and Smith, 1965) has been chosen to predict on-site surface (including sheet
and rill) erosion, for the following reasons:
1. This equation is applicable to a wide variety of land uses and
climatic conditions.
2. Data have been collected nationwide for factors included in
the equation.
The sediment loading function has the form:
Y(S)£ =1 [A. -(R-K-L.S.C-P).Sd] (III-l)
where
Y(S) = sediment loading from surface erosion, (tons/year,
tonnes/year)
n = number of subareas in the area
170
-------
A. = acreage of subarea i, (acres, ha)
R = the rainfall factor, expressing the erosion potential of
average annual rainfall in the locality.
K = the soil-erodibility factor, commonly expressed in tons
per acre per R unit
L = the slope-length factor, dimensionless ratio
S = the slope-steepness factor, dimensionless ratio
C = the cover factor, dimensionless ratio
P = the erosion control practice factor, dimensionless ratio
and
S, = the sediment delivery ratio, dimensionless.
d
Equation III-l can be used to predict sediment loading resulting from
sheet and rill erosion from cultivated and non-cultivated lands. Parameter
values for silviculture, construction, and mining are less well documented
than for agriculture, however. The user will thus find it relatively easy
to use Equation III-l for agriculture, and substantially more difficult for
other sources. The equation does not predict sediment contributions from
gully erosion, streambank erosion, or mass soil movement.
Estimation of surface erosion should be made for each relatively
homogeneous land-use type. For a given land-use type, if 90 percent or more
of the area is made up of one soil type, one may calculate soil loss for
each soil type that makes up at least 10 percent of the land use, and then
obtain a weighted average for the entire land-use area (U.S. Dept. of
Agriculture, 1974). There is no limitation on how finely the watershed can
be broken down into subwatersheds. This determination should be made based
on information density, time and monetary restrictions, and level of
accuracy desired in the results. Figure III-l is a flow diagram showing the
usage of Equation III-l for predicting annual average sediment loads.
171
-------
tvs
Local Drainage
Density Soil
Texture, and
Figure m-9,
or Eq. (HI-2)
Sediment
Delivery
Ratio 84
Land Use Acreage, A,- 's
.
t
Soil Names,
or Soil
Properties
.
Soil Erodibility
Factor K, from
Published Lists
or Nomographs
i
Land Use Types,
Dates of Cropstages,
Canopy, ground
Cover Density
1
Cover Factor C,
from Table HI -2
to HI -5, or Others
-
Types of
Conservation
Practice
•
-
Practice
Factor P,
from
Table m-6
»
n r~ ~~i
r(s)E=£ A| • (R-K • LS -c- p • sd )| 1
Slope Lengths,
and Slope
Gradients
*
Topographic
Factor LS,
from Figure
m-7 to m-8
Local Rainfall Erosivity
Factor R from Iso-erodent
Maps or by Calculation
FIGURE III-l FLOW DIAGRAM FOR CALCULATING SEDIMENT LOADING FROM SURFACE EROSION
-------
3.2.1.1 Data Requirements
The following should be obtained:
• total area, and land use acres in the area: cropland,
pastureland, and woodland, etc.
• soil characteristic information (e.g. soil texture) for each
land use.
• canopy and ground cover condition for each land use.
• good topographic maps
• the type and extent of conservation practices.
3.2.1.2 Determination of LISLE Factors
3.2.1.2.1 The Rainfall Factor (R)
R is a factor expressing the erosion potential of precipitation in a
locality. It is also called index of erosivity, erosion index, etc. It is
the summation of the individual storm products of the kinetic energy of
rainfall (denoted by E), and the maximum 30 minute rainfall intensity
(denoted by I) for all significant storms within the period under
consideration. The product El reflects the combined potential of raindrop
impact and runoff turbulence to transport dislodged soil particles from the
site (Wischmeier and Smith, 1965).
Values of average annual rainfal1-erosivity index, R, are shown in
Figure III-2 for the continental U.S. and Figure III-3 for islands of
Hawaii. On these maps, the lines joining points with the same erosion index
value are called isoerodents. Points lying between the indicated
isoerodents may be approximated by linear interpolation.
173
-------
35
FIGURE 111-2 AVERAGE ANNUAL VALUES OF THE RAINFALL-EROSIVITY FACTOR, R (EPA, 1975)
-------
0 5 10 Miles
OAHU
KAUAI
LANAI
MAUI
MOLOKAI
OAHU
FIGURE 111-3 MEAN ANNUAL VALUES OF EROSION INDEX FOR HAWAII (I'.SAA,, 1974)
-------
Interpolation for values of R factors in the mountainous areas,
particularly those west of the 104th meridian may not be appropriate because
of the sporadic rainfall pattern. Values of the erosion index at specific
areas can be computed from local recording rain gage records with the help
of a rainfall-energy table and the computation procedure presented by
Wischmeier and Smith (1978).
The USDA has recommended that 350 be the maximum used in the Gulf and
southeastern states, shown in Figure III-2, until further research can
validate values higher than 350.
In the northwestern United States, runoff from snowmelt contributes
significantly to surface erosion. The annual index of R for some portions
of this region is the combined effect of rainfall and snowmelt designated by
R and R , respectively. The snowmelt factor (R ) is important in Areas
I j »5
A-l, B-l, and C on Figure III-4 (also refer to Table III-l). The map values
in the shaded region of the Northwest (see Figure III-2) represent values
for the rainfall effect (R ) only, and must be added with appropriate R.
r b
values to account for the effect of runoff from thaw and snowmelt.
Interim procedures for calculating annual R values, which include both
R and R , for the northwestern U.S. are described in Conservation Agronomy
Technical Note No. 32, USDA/SCS, Portland, Oregon (1974), and are briefly
presented below.
The annual R factor is obtained by using as a base the two year, six
hour rainfall (2-6 rainfall). Relationships between R and 2-6 rainfall
vary to conform to specific local climatic characteristics. These
relationships are designated as Type I, Type IA, and Type II, and are shown
in Figure III-5. Specific areas applicable to these curves are shown in
Figure III-6. Type I curve is for the central valley and coastal mountains
and valleys of southern California. Type IA curve applies to the coastal
side of the Cascades in Oregon and Washington, the coastal side of the
Sierra Nevada Mountains in northern California, and the coastal regions of
Alaska. Type II curve applies to the remainder of the region. For 2-6
rainfall data, refer to Technical Paper No. 40, U.S. Department of Commerce,
176
-------
0 100 200 MILES
FIGURE 111-4 SOIL MOISTURE-SOIL TEMPERATURE REGIMES OF THE
WESTERN UNITED STATES (U.S.P.A,, 197*0
177
-------
TABLE III-l
APPLICABILITY OF Rp AND Rg FACTORS IN THE AREAS
WEST OF THE ROCKY MOUNTAINS (U.S. DEPT. OF AGRICULTURE, 1974a)
Areas (see
Figure III-4)
A-l
A-2
A-3
A-4
B-l
B-2
Typical Locations
Washington, Idaho, Nevada,
California, western Utah
Cascades, Sierra, Tetons of
Idaho, Wasatch Mountains
West of Cascades, San Joaquin
Valley, west of Sierras
Areas of southern California,
east of Santa Anas, southern
Nevada, intermountain Nevada,
Salt Lake area, Utah
Western Montana, Colorado,
eastern Utah, high elevations
of Arizona
Great plains area of eastern
Montana, Wyoming, Colorado
(includes gently sloping
mesas and upland at lower
elevations of Monticello,
Utah area)
Rainfall during summer is
high; high elevations
xi/
X
X
X
b/
a/ X needed
b/ - not needed
178
-------
2.0
3.0
4.0
2-Year, 6-Hour Rainfall, cm
5.0 6.0 7.0 8.0 9.0
II.0 12.0
600 -
1.5
2.0 2.5 3.0 3.5
2-Year, 6-Hour Rainfall,
4.0
4.5
5.0
FIGURE 111-5 RELATIONSHIPS BETWEEN ANNUAL AVERAGE RAINFALL FROSIVITY INDEX AND THE
?-YEAR, 6 HOUR RAINFALL DEPTH FOR 3 RAINFALL TYPES IN WESTERN ?',$,
'!! c n A 107/1)
. u i O 11' i n i j -L3/ ./
! I
I
-------
LEGEND- Storm Distribution
1 J TYPE IA
TYPE I
TYPE
FIGURE III-6 STORM DISTRIBUTION REGIONS IN V/ESTERN n,S,
(U.S.TU,, 1974)
180
-------
Weather Bureau, Washington, D.C. (1961), or other suitable rainfall
frequency analysis reports.
To obtain the annual R factor for a given location, obtain the average
annual total precipitation by snowfall (in inches of water depth) and
multiply it by the constant 1.5.
There are numerous sources of snowfall data for the United States.
Some of the major sources are:
• The 1941 Yearbook of Agriculture, USDA, Washington, D.C.,
• "Climates of the States," Water Information Center, Inc., Port
Washington, New York (1974), and
• Data resulting from the Western Federal-State-Private
Cooperative Snow Surveys, coordinated by SCS/USDA,
Portland, Oregon.
Data on snow density is necessary to convert depth of snow to depth of
meltwater. Snow at the time of fall may have a density as low as 0.01 and
as high as 0.15 g/ml. The average snow density for the United States is
taken to be 0.10 (Garstka, 1964). If snowfall is recorded as inches of
precipitation, no conversion is required.
The monthly distribution of the erosion index for the 37 states east of
the Rocky Mountains has been reported in USDA-ARS Agriculture Handbook
No. 282 (Wischmeier and Smith, 1965). Average monthly erosion index values
are expressed as percentages of average annual values and plotted
cumulatively against time in Appendix A.
The monthly distribution of erosion index for the islands of Hawaii
also has been developed (U.S. Dept. of Agriculture, 1974b). These curves
are shown in Appendix A. If monthly or seasonal sediment yields are
required the annual R value can be factored using the percentages from these
figures.
181
-------
For the areas west of the Rockies in the continental United States, the
monthly distribution of erosion index R is the summation of R and R .
r s
Where RS values are not needed, the R and R curves are the same.
The R factor in equation III-l can be expressed in metric units
((hundreds of metric tons/ha-cm) multiplied by (maximum 30-minute intensity,
cm/hr)) by multiplying the English R values by 1.735.
3.2.1.2.2 The Soi1-Erodibility Factor (K)
K factor is a quantitative measure of the rate at which a soil will
erode, expressed as the soil loss (tons) per acre per unit of R, for a plot
with 9 percent slope, 72.6 ft. long under continuous cultivated fallow. K
factors for topsoils, as well as subsoils, for most soil series have been
developed. Values of K for soils studied thus far vary from 0.12 to 0.70
tons/acre per unit R. Values can usually be obtained from the regional or
state offices of the Soil Conservation Service.
K values of soils can also be predicted from soil properties. In
Appendix B of this report, two methods are presented from which K values may
be determined for topsoils and subsoils when the governing soil properties
are known. The factor for conversion of K in English units to metric-tons
per hectare per metric R unit is 1.292 (Wischmeier, 1972).
3.2.1.2.3 The Topographic Factor (IS)
Soil loss is affected by both length (L) and steepness of slope (S).
These factors affect the capability of runoff to detach and transport soil
materi al.
The slope length factor is the ratio of soil loss from a specific
length of slope to that length (72.6 ft) specified for the K factor in the
USLE. Slope length is defined as the distance from the point of origin of
overland flow to either of the following: the point where the slope
decreases to the extent that deposition begins or the point where runoff
182
-------
enters a well-defined channel. Slope length can be determined accurately by
on-site inspection of a field, or by measurements from topographic maps.
When the land is terraced, the terrace spacing should be used.
The slope gradient or percent slope factor is the ratio of soil loss
from a specific percent slope to that slope (9 percent) specified for the K
value in the USLE. A 9 percent slope has a factor value of 1. Slope data
may be obtained from topographic maps, engineering or land level surveys,
and other sources. A widely used method is to estimate slope from soil
survey maps in which the soils have been mapped by slope range.
The slope length (L) and slope gradient (S) are usually combined in the
USLE into a single dimensionless topographic factor, LS, which can be
evaluated using a slope-effect chart.
The slope-effect chart in Figure III-7 is designed for the following
areas shown in Figure III-4: A-l in Washington, Oregon, and Idaho; and all
of A-3 (U.S. Dept. of Agriculture, 1974a). For the remainder of the U.S.,
the slope-effect chart, Figure III-8, is to be used (U.S. Dept. of
Agriculture, 1974a).
Slope-effect charts in Figure III-7 and III-8 can be used when uniform
slopes are assumed. The following steps are to be used for obtaining LS
values from these charts:
1. Enter the chart on the horizontal axis with the appropriate
value of slope length.
2. Follow the vertical line for that slope length to where it
intersects the curve for the appropriate percent slope.
3. Read across the point of intersection to the vertical axis.
The number on the vertical axis is the LS value.
183
-------
Slope Length, Meters
20 30 40 60 80IOO ISO 200 30O 4OO 60O 8OO
40.0
20.0
10.0
_J
- 6.0
O
O 4.0
O
u_
a.
o
2.0
0>
CX 1.0
o
0.6
0.4
0.2
O.I
(Slope %)
60
, 50
45
'40
•35
•30
25
,20
18
• 16
•14
• 12
. 10
8
•0.5
70 100 200 4OO 600 1000 2000
Slope Length, Feet
FIGURE 111-7
SLOPE EFFECT CHART APPLICABLE TO AREAS A-l
IN WASHINGTON, OREGON, AND IDAHO AND ALL OF
A-3 A^B/ (U,S, DEPT, OF AGRICULTURE, 1974)
A/ SEE FIGURE II1-4,
B/ DASHED LINES ARE EXTENSIONS OF LS FORMULAE BEYOND VALUES
TESTED IN STUDIES,
184
-------
20.0
3.5 6.0
10
Slope Length, Meters
20 40 60 100
200
400 600
1 ' ' >*
O.I I*
10 20 40 60 100 200 400 600 1000 2000
Slope Length, Feet
FIGURE 111-8 SLOPE—EFFECT CHART FOR AREAS WHERE FIGURE 111-7
is NOT APPLICABLE A/ (U.S. DEPT. OF AGRICULTURE)
A/ THE DASHED LINES REPRESENT ESTIMATES FOR SLOPE DIMENSIONS
BEYOND THE RANGE OF LENGTHS AND STEEPNESSES FOR WHICH DATA
ARE AVAILABLE,
185
-------
3.2.1.2.4 The Cover Management Factor (C)
In the USLE, the factor C represents the ratio of soil qualtity eroded
from land that is cropped or treated under a specified condition to that
which is eroded from clean-tilled fallow under identical slope and rainfall
conditions. C ranges in value from near zero for excellent sod or a
well-developed forest to 1.0 for continuous fallow, construction areas, or
other extensively disturbed soil.
Values of factor C for croplands are highly variable with planting
dates, type of vegetative cover, seeding method, soil tillage, disposition
of residues, and general management level. Generalized C values for various
types of crop management systems are listed in Table III-2. The reader is
advised to consult with state conservation agronomists of SCS for
appropriate C values for crops in the local area. The reader is also
referred to USDA-ARS Agriculture Handbook No. 537 (Wischmeier and
Smith, 1978) for a listing of approximated C values for various crops at
each crop stage, as well as a working table for derivation of average C
value for periods of crop rotation.
C values typical of permanent pasture, range, and idle lands, with
varying cover and canopy conditions, are given in Table III-3. These values
were developed by Wischmeier (1972). Wischmeier (1972) has also estimated C
values for some woodland situations (Table III-4).
For urban and road areas, as well as construction sites, the factor C
represents the effect of land cover or treatment that may be used to protect
soil from being eroded. Table III-5 (Water Resources Administration, 1973)
lists values of the C factors for various soil covers and treatments.
3.2.1.2.5 The Practice Factor (P)
The factor P accounts for control practices that reduce the erosion
potential of runoff by their influence on drainage patterns, runoff
concentration, and runoff velocity.
186
-------
GENERALIZED VALUES OF THE COVER AND MANAGEMENT FACTOR, C,
IN THE 37 STATES EAST OF THE ROCKY MOUNTAINS1
Line , , -\
Clop, rotation, and management
no.
Base value: continuous fallow, tilled up and down slope
CORN
1 C, RdR, fallTP, conv(l)
2 C, RdR, spring TP, conv(l)
3 C, RdL. fall TP, conv(l)
4 C, RdR, we seeding, spring TP, conv (1 )
5 C, RdL, standing, spring TP, conv (1)
6 C, fall shred stalks, spring TP, conv (1 )
7 C(silage)-W(RdL, fall TP) (2)
8 C, RdL, fall chisel, spring disk, 40-30% re (1)
9 C(silagc), W we seeding, no-till pi m c-k W (1 )
10 C(RdL)-W(RdL, spring TP) (2)
1 1 C, fall shred stalks, chisel pi , 40-30% re (1 )
1 2 C-C-C-W-M, RdL, TP for C, disk for W (5)
1 3 C, RdL, strip till row zones, 55-40'7r. re (1 )
14 C-C-C-W-M-M, RdL, TP for C, disk for W (6)
15 C-C-W-M, RdL, TP for C, disk for W (4)
16 C, fall shred, no-till pi , 70-50',; re ( 1 )
1 7 C-C-W-M-M, RdL, TP for C disk tor W (5)
18 C-C-C-W-M, RdL, no-till pi 2d & 3rd C(5)
19 C-C-W-M, RdL, no-till pi 2d C (4)
20 C, no-till pi in e-k wheat, 90-70'/ re ( 1 )
21 C-C-C-W-M-M, no-till pi 2d & 3rd C (6)
22 C-W-M, RdL, TP for C, disk for \V (3)
23 C-C-W-M-M, RdL, no-till pi 2d C(5)
24 C-W-M-M, RdL, TP for C, disk tor W (4)
25 C-W-M-M-M, RdL, TP for C, disk for W (5)
26 C, no-till pi in c-k sod, 95-80% re (1 )
COTTON4
27 Cot, eonv (Western Plains) (1 )
28 Cot, conv (South) (1)
MFADOW
29 Grass & Legume mix
30 Alfalfa, lespedc/.a or Scricia
31 Sweet clover
SORGHUM, GRAIN (Western Plains)4
32 RdL, spring TP, eonv (1)
33 No-till pi in shredded 70-50' < re
Productivity level"
High
Mod.
C value
1.00
0.54
.50
.42
.40
.38
.35
.31
.24
.20
.20
.19
.17
16
14
.12
.11
087
.076
.068
.062
061
.055
051
.039
.032
017
0.42
.34
0.004
020
.025
0 43
11
1.00
0.62
59
.52
49
.48
44
.35
30
24
28
26
.23
24
20
17
18
14
13
11
14
1 1
095
094
074
061
.053
0 49
.40
001
0 53
IK
187
-------
TABLE 111-2 (Continued)
GENERALIZED VALUES OF THE COVER AND MANAGEMENT FACTOR, C,
IN THE 37 STATES EAST OF THE ROCKY MOUNTAINS1- continued
Line
no.
Crop, rotation, and management
SOYBEANS4
34
35
36
37
WHF.AT
38
39
40
41
42
43
44
45
46
47
48
49
B, RdL, spring TP, conv (1)
C-B, TP annually, conv (2)
B, no-till pi
C-B, no-till pi, fall shred C stalks (2)
W-F, fall TP after W (2)
W-F, stubble mulch, 500 Ibs re (2)
W-F, stubble mulch, 1000 Ibs re (2)
Spring W, RdL, Sept TP, conv (N & S Dak) (1 )
Winter W, RdL. Aug TP, conv (Kans) (1)
Spring W, stubble mulch, 750 Ibs re (1 )
Spring W, stubble mulch, 1250 Ibs re (1)
Winter W, stubble mulch, 750 Ibs re (1 )
Winter W, stubble mulch, 1 250 Ibs re (1 )
W-M, conv (2)
W-M-M, conv (3)
W-M-M-M, conv (4)
Productivity level2
High
Mod.
C value
0.48
.43
.22
.18
0.38
.32
.21
.23
.19
.15
.12
.11
.10
.054
.026
.021
0.54
.51
.28
.22
1 This table is for illustrative purposes only and is not a complete list of cropping systems or potential practices. Values of C differ
with rainfall pattern and planting dates. These generalized values show approximately the relative erosion-reducing effectiveness of
various crop systems, but locationally derived C values should be used for conservation planning at the field level. Tables of local
values are available from the Soil Conservation Service.
2 High level is exemplified by long-term yield averages greater than 75 bu. corn or 3 tons grass-and-legume hay; or cotton manage-
ment that regularly provides good stands and growth.
3 Numbers in parentheses indicate number of years in the rotation cycle. No. (1) designates a continuous one-crop system.
Gram sorghum, soybeans, or cotton may be substituted for com in lines 12,14,15, 17-19, 21-25 to estimate Cvalues for sod-
based rotations.
Abbreviations defined:
B - soybeans
C - corn
c-k - chemically killed
conv - conventional
cot - cotton
F - fallow
M - grass & legume hay
pi - plant
W -wheat
we - winter cover
Ibs re - pounds of crop residue per acre remaining on surface after new crop seeding
% re - percentage of soil surface covered by residue mulch after new crop seeding
70-50% re - 70% cover for C values in first column; 50% for second column
RdR - residues (corn stover, straw, etc.) removed or burned
RdL - all residues left on field (on surface or incorporated)
TP - turn plowed (upper 5 or more inches of soil inverted, covering residues)
Source: U.S. EPA, 1975
188
-------
TABLE 111-3
00
"C" VALUES FOR PERMANENT PASTURE, RANGELAND, AND IDLE LAND (WISCHMEIER, 1972)-7
Vegetal Canopy
Type and Height , ,
of Raised Canopy-
Column No.
No appreciable canopy
Canopy of tall weeds
or short brush
(0.5 m fall height)
Appreciable brush
or brushes
(2 m fall height)
Canopy-
Cover—' ,,
(%) Type—
2 3
G
W
25 G
W
50 G
W
75 G
W
25 G
w
50 G
W
75 G
W
Cover that Contacts the Surface
0
4
0.45
0.45
0.36
0.36
0.26
0.26
0.17
0.17
0.40
0.40
0.34
0.34
0.28
0.28
20
5
0.20
0.24
0.17
0.20
0.13
0.16
0.10
0.12
0.18
0.22
0.16
0.19
0.14
0.17
Percent
40
6
0.10
0.15
0.09
0.13
0.07
0.11
0.06
0.09
0.09
0.14
0.085
0.13
0.08
0.12
Ground Cover
60
7
0.042
0.090
0.038
0.082
0.035
0.075
0.031
0.067
0.040
0.085
0.038
0.081
0.036
0.077
80
8
0.013
0.043
0.012
0.041
0.012
0.039
0.011
0.038
0.013
0.042
0.012
0.041
0.012
0.040
95-100
9
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
0.003
0.011
(continued)
-------
TABLE 111-3 (Continued)
Vegetal Canopy
Type and Height , ,
of Raised Canopy-
Column No.
Trees but no appreci-
able low brush
(4 m fall height)
Canopy
Cover- ..
(%) Type—
2 3
25 G
W
50 G
W
75 G
W
Cover
that Contacts the
Percent
0
4
0.42
0.42
0.39
0.39
0.36
0.36
0
0
0
0
0
0
20
5
.19
.23
.18
.21
.17
.20
0
0
0
0
0
0
40
6
.10
.14
.09
.14
.09
.13
Surface
Ground Cover
0
0
0
0
0
0
60
7
.041
.087
.040
.085
.039
.083
0
0
0
0
0
0
80
8
.013
.042
.013
.042
.012
.041
95-100
9
0.003
0.011
0.003
0.011
0.003
0.011
a_/ All values shown assume: (1) random distribution of mulch or vegetation, and (2) mulch of
appreciable depth where it exists.
b/ Average fall height of waterdrops from canopy to soil surface: m = meters.
c_/ Portion of total-area surface that would be hidden from view by canopy in a vertical
projection (a bird's-eye view).
d_/ G: Cover at surface is grass, grasslike plants, decaying compacted duff, or litter
at least 5 cm (2 in.) deep.
W: Cover at surface is mostly broadleaf herbaceous plants (as weeds) with little
lateral-root network near the surface and/or undecayed residue.
-------
TABLE 111-4
'C" VALUES FOR WOODLAND (WISCHMEIER, 1972)
Stand Condition
Well stocked
Medium stocked
Poorly stocked
Tree Canopy
Percent of
Area^-7
100-75
70-40
35-20
Forest
Litter
Percent of
Area^
100-90
85-75
70-40
Undergrowth—
Managed—
Unmanaged—
Managed
Unmanaged
Managed
Unmanaged
"C" Factor
0.001
0.003-0. Oil
0.002-0.004
0.01-0.04
0.003-0.009
0.02-0.09-7
a/ When tree canopy is less than 20%, the area will be considered as
grassland or cropland for estimating soil loss.
b/ Forest litter is assumed to be at least 2-in. deep over the percent
ground surface area covered.
c/ Undergrowth is defined as shrubs, weeds, grasses, vines, etc., on
the surface area not protected by forest litter. Usually found
under canopy openings.
d_/ Managed - grazing and fires are controlled.
Unmanaged - stands that are overgrazed or subjected to repeated
burning.
e/ For unmanaged woodland with litter cover of less than 75%, C values
should be derived by taking 0.7 of the appropriate values in
Table III-3. The factor of 0.7 adjusts for the much higher soil
organic matter on -permanent woodland.
191
-------
TABLE III-5
"C" VALUES FOR CONSTRUCTION SITES
(WATER RESOURCES ADMINISTRATION, 1973)
Type of Cover
C Value
None (fallow) 1.00
Temporary seedings
First 60 days 0.40
After 60 days 0.05
Permanent seedings
First 60 days 0.40
After 60 days 0.05
After 1 year 0.01
Sod (laid immediately) 0.01
Rate of Application
Mulch
Hay or straw
Stone or gravel
Chemical mulches
In Metric Tons In Tons
Per Hectare Per Acre
1/2
1
1-1/2
2
14
55
120
220
1/2
1
1-1/2
2
15
60
135
240
C Value
0.34
0.20
0.10
0.05
0.80
0.20
0.10
0.05
Maximum Allowable
Slope Length
TO
20
30
40
50
15
80
175
200
6
9
12
15
5
24
53
61
First 90 days
After 90 days
Woodchips
2
4
6
11
18
23
a/
a/
2
4
7
12
20
25
0.50
1 .00
0.80
0.30
0.20
0.10
0.06
0.05
50
50
25
50
75
100
150
200
15
15
8
15
23
30
46
61
a/ As recommended by manufacturer.
192
-------
For croplands, control practices refer to contour tillage, cross-slope
farming, and contour strip-cropping. The practice value P is the ratio of
soil loss from a specified conservation practice to the soil loss occurring
with up-and-downhill tillage, when other conditions remain constant. Table
III-6 (USEPA, 1975) shows P values currently in common usage.
3.2.1.2.6 Sediment Delivery Ratio ($d)
The sediment-delivery ratio, in this report, is defined as the fraction
of the gross erosion which is delivered to some point in the stream system
downstream of the source area. The classical method for determining an
average delivery ratio is by comparing the magnitude of the sediment yield
at a given point in a watershed (generally at a reservoir or a stream
sediment measuring station), and the total amount of erosion. The
quantities of gross erosion from sloping uplands are computed by erosion
prediction equations for surface erosion, and estimated by various
procedures for gullies, stream channels, and other sources. The sedimenb
yield at a given downstream point is obtained through direct measurements.
Estimates of the delivery ratio for some specific watersheds, particularly
in the humid sections of the country, can be obtained from the Soil
Conservation Service, USDA.
Many delivery-ratio studies have been aimed at finding measurable
influencing factors that can be related to sediment-delivery ratio. A
popular means of developing such information is by statistical analysis
using the sediment-delivery ratio as the dependent variable and measurable
watershed factors as the independent, or controlling variables. Many
physical and hydrologic factors may influence sediment-delivery ratios.
Empirical relationships for delivery ratios have been proposed and some are
presented below. Estimates of sediment loading can be made through the use
of these relationships, but such estimates should be tempered with judgment
and consideration of other influencing factors not included in the
quantitative expressions. The user is encouraged to consult with local
experts and should use local data when available.
193
-------
TABLE 111-6
PRACTICE FACTORS (P) USED IN SEDIMENT
LOADING EQUATION
Practice
Contouring (Pt)
Contour strip cropping (Pst->
R-R-M-M1
R-W-M-M
R-R-W-M
R-W
R-O
Contour listing or ridge planting
(Pel)
Contour terracing (Pj)
No support practice
Land slope (percent)
1.1-2
2.1-7
7.1-12
12.1-18
18.1-24
(1-actor P)
0.60
0.30
0.30
0.45
0.52
0.60
0.30
3 0.6/V^
1.0
0.50
0.25
0.25
0.38
0.44
0.50
0.25
0.5/Vn"
1.0
0.60
0.30
0.30
0.45
0.52
0.60
0.30
0.6 A/T
1.0
0.80
0.40
0.40
0.60
0.70
0.80
0.40
0.8/V^T
1.0
0.90
0.45
0.45
0.68
0.90
0.90
0.45
0.9/VrT
1.0
R » rowcrop, W * fall-seeded grain, O » spring-seeded grain. M = meadow. The crops are grown in rotation and \o arranged on
the field that rowcrop strips arc always separated by a meadow or winter-grain strip.
These Pt values estimate the amount of .soil eroded to the terrace channels and arc used for conservation planning. Tor prediction
of off-field sediment, the P| values are multiplied by 0.2.
n = number of approximately equal-length intervals into which the field slope is divided by the terraces. Tillage operations must
be parallel to the terraces.
Source: U.S. EPA, 1975.
194
-------
The MITRE Corporation (1974) reported that the sediment delivery ratio
for construction sites can be approximated by a function of the overland
distance between the construction site and the receiving water:
Sd = D"°'22 (HI-2)
where
D = overland distance between the erosion site and the receiving
water, (ft).
The above equation was empirically derived from available data. The
data base for the derivations includes values of D from 0 to 800 ft. MITRE
suggests that this function should be further tested, particularly in areas
of the Midwest and Central U.S.
For mining sites, logging roads and fire lanes, sediment delivery ratio
relationships have not yet been established due to lack of systematically
measured data. It is suggested that the delivery ratio developed by MITRE
and expressed in Equation III-2 be used as the first approximation for these
sites. This should be verified when appropriate data become available.
Sediment delivery ratios have been evaluated in many areas of the
country, particularly the eastern half of the United States. The delivery
ratio usually depicts a general trend in basins that are relatively
homogeneous with respect to soils, land cover, climate, and topography. The
Soil Conservation Service (1973) analyzed data from stream and reservoir
sediment surveys from widely scattered areas.
This analysis shows that sediment delivery ratios vary inversely with
"drainage basin size". It also indicates the effect of soil texture of
upland soil on the sediment delivery ratio.
The delivery ratio relationships reported by SCS (1973) were used by an
MRI study group in developing delivery ratios for sediment loading to
watercourses. The result is shown in Figure III-9. The horizontal scale of
the figure is the reciprocal of drainage density which is defined as the
195
-------
01
IOO
I/Drainage Density, Kilometers
i.o 10
100 200 500
Silty Clay
Predominantly Silt
0.02
1.0 10
I/Drainage Density, Miles
100
SWa
400
FIGURE 111-9 SEDIMENT DELIVERY RATIO FOR RELATIVELY HOMOGENEOUS BASINS
(MLROY, ET AL,, 1976)
-------
ratio of total channel-segment lengths (accumulated for all orders within a
basin) to the basin area. The reciprocal of drainage density may be thought
of as an expression of the closeness of spacing of channels, or the average
distance for soil particles to travel from the erosion site to the receiving
water. The drainage density is found by dividing the total length of
perennial streams in the waters and by the area of the watershed.
The delivery ratio relationship shown in Figure III-9 also takes into
account the effect of soil texture. For example, if soil texture of upland
soil is essentially silt or clay, the sediment delivery ratio will be higher
than when the soil texture is coarse.
The following steps are to be used to obtain the delivery ratio (S,)
from Figure III-9.
1. Enter the figure on the horizontal axis with the value of the
reciprocal of drainage density (1/DD).
2. Move vertically from the value of 1/DD to where it intersects
the curve for the appropriate soil texture.
3. Read across from the point of intersection to the vertical
axis. That number represents the delivery ratio, S,.
A great range of values of drainage density exists in the United
States, from 2 km/km2 (3 miles/mile2) for the Appalachian Plateau Province
(Smith, 1950) to 500 km/km2 (800 miles/mile2) in Badlands at Perth Amboy,
New Jersey (Schumm, 1956). In general, according to Strahler (1964), low
drainage density is found in regions of highly resistant or highly permeable
subsoil materials, under dense vegetative cover, and where relief is low.
High drainage density is favored in regions of weak or impermeable
materials, sparse vegetation, and mountainous relief.
Some typical values of drainage density for various locales in the U.S.
are given in Table III-7. Local drainage density figures may be obtained
from agencies such as the U.S. Geological Survey and the Army Corps of
Engineers.
197
-------
TABLE II1-7
TYPICAL VALUES OF DRAINAGE DENSITY
Location
Drainage density
2 2
km/ km mile/mile
Reference
Appalachian Plateau
Province
Central and eastern
United States
Dry Areas of the Rocky
Mountain Region
The Rocky Mountain Region
(except the above)
Coastal ranges of
Southern California
Badlands in South Dakota
Badlands in New Jersey
1.9-2.5
5-10
31-62
3.0-4.0
Smith (1950)
8.0-16.0 Strahler (1952)
50-100
5-10
12-25
125-250
183-510
8.0-16
20-40
200-400
310-820
Melton (1957)
Melton (1957)
Smith (1950)
Melton (1957)
Maxwell (1960)
Smith (1958)
Schumn (1956)
198
-------
Measurements of drainage density can be made from a topographic map
with a planimeter and chartometer. Care must be taken to include all
permanent stream channels to their upper ends by checking in the field or
with aerial photographs to verify topographic maps. A rapid approximation
method for determining drainage density is suggested by Carlston and
Langbein (1960).
3.2.1.3 Limitations and Accuracy of Sediment Loading Equation
The USLE predicts soil losses from sheet and rill erosion. It does not
predict sediment from gullies, streambank erosion, landslides, road ditches,
irrigation, or from wind erosion. The USLE was developed primarily for
croplands, and has been chiefly based upon experimental plot data from the
areas east of the Rocky Mountains. The loading function therefore is best
defined for these areas of use. For croplands in the western United States
and sources outside agriculture such as silviculture, construction, and
mining, the factors have not been systematically developed, which seriously
affects the ease of using the USLE for such sources.
The USLE was developed primarily as a means of predicting annual
erosion losses. Although several methods have been proposed for using it to
calculate losses from individual events the MUSLE (Williams, 1975) presented
later in this chapter is recommended.
The loading function (Equation III-l) and supporting data in tables and
figures were designed to predict longterm average loadings for specific
conditions. Sediment loading for a specific year may be substantially
greater or smaller than the annual averages because of differences in
number, size, and timing of erosive rainstorms, and in other weather
parameters. Table II of USDA Agriculture Handbook 282 (1965) contains a
listing of 50, 20, and 5 percent probability values of R factor at 181 key
locations in the area east of the Rocky Mountains. These may be used for
further characterization of soil-loss hazards.
199
-------
Because of year to year variations in climate and management factors,
the average soil loss predicted by the USLE may be different from that
occurring in any given year. Onstad et _al_. (1979) evaluated the precision
of the USLE factors. The results (Table 111-8) show that C is the most
variable parameter followed by K, R, L and S. When the coefficients of
variation for each parameter where used in an error propagation exercise, it
was found that the coefficient of variation of soil loss prediction was
roughly 107 percent.
EXAMPLE III-1<
Assessing Sediment Loading from Surface Erosion
The watershed of interest has an area of 830 acres. It is located in
Parke County, Indiana. Compute sediment loading from the watershed from
sheet and rill erosion in terms of average annual loading.
Basic Information
Land use types:
• Cropland
• Pasture
• Woodland
Delivery ratio: 60 percent
Land information: (Cropland - 180 acres)
• Continuous corn on contours
• Conventional tillage, average yield, 40 to 45 bu
• Cornstalks are removed after harvest, winter crop seeded
t Spring turn plowed
• Soil - Fayette silt loam
• Slope - 6 percent
• Slope length - 250 ft
200
-------
ISJ
o
TABLE III-8
EROSION EQUATION FACTOR PRECISION ERROR
Factor
Rainfall factor (R)
Soil credibility (K)
Cropping management
factor (C)
Slope steepness (s)
Slope steepness factor (S)
Slope length (1)
Slope length factor (L)
Number
Reps
22 yr
4 to 20 loc.
4 to 20
single plots
3
3
of
Treatments
42 loc.
14 soils
5 rotations
29
6
Range of
means
57-231
0.11-0.41
0.01-1.40
1.0-46.3
0.02-16.1
50-148
1.5-1.6
Coefficient
of Variation
I
34
39
92
4
5
38
19
Source: Onstad, et al., 1979
-------
Pasture: (220 acres)
• No appreciable canopy
• Cover at surface - grass and grass!ike plants
• Percent of surface or ground cover - 80 percent
• Soil - Fayette silt loam
• Slope - 6 percent
• Slope length - 200 ft
Woodland: (430 acres)
• Medium stocked
• Percent of area covered by tree canopy - 50 percent
• Percent of area covered by litter - 80 percent
• Undergrowth - managed
• Soil - Bates silt loam
• Slope - 12 percent
• Slope length - 150 ft
Solution:
Cropland:
R =
K =
LS =
C =
P
S =
200 (Figure III-2)
0.37 (USDA-SCS)
1.08 (Figure III-8)
0.49 (Table III-2)
0.50 (Table III-6)
0.60
Calculate average annual loading per acre.
Y(S) , = 200 x 0.37 x 1.08 x 0.49 x 0.50 x 0.6
x 'annual
Pasture:
R =
K =
LS =
C =
P =
= 11.7 tons/acre-year
200
0.37
0.95
0.013 (Table III-3)
1.0
202
-------
= 0.60
Y(S)
annual
Woodland:
R =
K =
LS =
C =
P =
S =
= 200 x 0.37 x 0.95 x 0.013 x 1.0 x 0.6
- 0.548 tons/acre-year = 1,100 Ib/acre-year
200
0.32
2.75
0.003 (Table III-4)
1.0
0.60
Y(S)
annual
= 200 x 0.32 x 2.75 x 0.003 x 1.0 x 0.60
= 0.3168 tons/acre-year
Calculations of Gross Loading
Average annual:
Cropland - 180 acres x 11.7 tons/acre-year
Pasture - 220 acres x 0.55 tons/acre-year
Woodland - 430 acres x 0.32 tons/acre-year
Total
Y(S)[
= 2114 tons/year
= 121 tons/year
= 138 tons/year
= 2374 tons/year
END OF EXAMPLE III-l
3.2.2 Single Event Sediment Loads
Soil erosion is governed by two processes: detachment of soil fines
and transport of soil fines to the receiving water body. Detachment of soil
fines can be accomplished by erosive rainfall or by shear forces at the soil
surface created by surface runoff. One of the shortcomings of the USLE is
that it does not explicitly consider transport nor detachment by runoff.
Williams (1975) modified the USLE by replacing the "R" factor with a "runoff
203
-------
energy" factor to provide more accurate sediment loss prediction for single
storm events. In addition this modification eliminates the need for the
sediment delivery ratio. The form of this modification is
Y(S)£ * 95 (Vq )°-56 KLSCP (HI-3)
where
Y(S)F = sediment yield in tons per event
V = volume of runoff in acre-feet and
q = peak flow rate in cubic feet per second
Other terms are as previously defined.
Alternatively the modified USLE (MUSLE) can be expressed
in metric units as
Y(S)E = 11.8 (Vqp)0'56 KLSCP (III'
where
Y(S)r = sediment yield in tonnes
V = volume of runoff (m3)and
q = peak runoff rate (m3/sec)
In order to use MUSLE the terms V and qp must be evaluated. This is
most conveniently accomplished using the Soil Conservation Service runoff
curve number method (Mockus, 1972).
3.2.2.1 Evaluation of Total Storm Runoff
The depth of runoff from the watershed area is estimated by the
following Equation:
Q = (R-0.2S)2/(R+0.8S) (III-5)
204
-------
where
Q = the depth of runoff from the watershed area (in, cm)
R = the total storm rainfall (in, cm) and
S = water retention parameter (in, cm).
The storm runoff volume can be calculated by
W
where
V = total storm runoff volume (acre-ft, m3)
a = a units conversion, 0.083 English, 100 metric
A = watershed area (acres, ha)
w
Q = depth of runoff (in, cm)
S, the watershed retention parameter, is calculated using
s , (l°°° . 10) . a (ni-7)
where
CN = the SCS Runoff Curve Number (dimensionless) and
a = 1.0 English, 2.54 metric.
Runoff curve numbers are dependent upon antecedent soil water
conditions, the relative permeability of the soil and vegetation cover and
management factors. Table III-9 gives runoff curve numbers for various
combinations of the above factors. The table is used by first determining
the hydrologic soil group. Descriptions of each group are located at the
bottom of the table. Next, move to the proper row for the crop/management
scheme. Within each crop/management scheme a subrow labeled Hydrologic
Condition is found. The qualifiers "good", "fair", or "poor" indicate
relative management conditions. For instance, under the crop management
scenario "small grains", "contoured", a "poor" hydrologic condition would be
a poor stand of vegetation with breakthroughs in the contours both of which
would increase surface runoff.
205
-------
TABLE III-9
RUNOFF CURVE NUBERS FOR HYDROLOGIC SOIL-COVER COMPLEXES
(FOR ANTECEDENT RAINFALL CONDITION II)
Hydro! ogic
Land Use
or Cover
Fallow
Row crops
Small grain
Close-seeded
legumes or
rotation
meadow
Pasture or
range
Meadow
(permanent)
Woods
(farm woodlots)
Farmsteads
Roads and
Treatment
or Practice
Straight row
Straight row
Straight row
Contoured
Contoured
Terraced
Terraced
Straight row
Straight row
Contoured
Contoured
Terraced
Terraced
Straight row
Straight row
Contoured
Contoured
Terraced
Terraced
Contoured
Contoured
Contoured
Hydrologic
Condition
—
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Good
Poor
Fair
Good
Poor
Fair
Good
Good
Poor
Fair
Good
—
—
A
77
72
67
70
65
66
62
65
63
63
61
61
59
66
58
64
55
63 /
51
68
49
39
47
25
6
30
45
36
25
59
74
B
86
81
78
79
75
74
71
76
75
74
73
72
70
77
72
75
69
73
67
79
69
61
67
59
35
58
66
60
55
74
84
Soil Group
C
91
88
85
84
82
80
78
84
83
82
81
79
78
t
85
81
83
78
80
76
86
79
74
81
75
70
71
77
73
70
82
90
D
94
91
89
88
86
82
81
88
87
85
84
82
81
89
85
85
83
83
80
89
84
80
88
83
79
78
83
79
77
86
92
right-of-v/ay
(hard surface)
*Soil Group Description
A Lowest Runoff Potential: Includes deep sands with very little silt and clay,
also deep, rapidly permeable loess.
B Moderately Low Runoff Potential: Mostely sandy soils less deep than A, and
loess less deep or less aggregated than A, but the group as a whole has above-
average infiltration after thorough wetting.
C Moderately High Runoff Potential: Comprises shallow soils and soils containing
considerable clay and colloids, though less than those of group D. The group
has below-average infiltration after presaturation.
D Highest Runoff Potential: Includes mostly clays of high swelling per cent, but
the group also includes some shallow soils with nearly irperr.eable subhorizons
near the surface.
Source: Scf~>w3b et ^1. , 1955
206
-------
The intersection of the crop/management/condition row with the
hydrologic soil group column is the curve number for this watershed. This
table is, however, for antecedent soil moisture condition II. To account
for very wet or very dry antecedent conditions the curve number is
multiplied by the appropriate correction found in Table 111-10.
3.2.2.2 Estimation of Peak Storm Runoff
In order to estimate peak storm runoff a hydrograph shape must be
assumed. Commonly, a triangular or trapezoidal shape is used. A
trapezoidal hydrograph is used here. The equation for the peak runoff rate
is
a A, R Q
q = ___w _ (IH-8)
P Tr (R-0.2S)
where
q = the peak runoff rate (ft3/sec, m3/sec)
A = watershed area (acres, ha)
R = total storm rainfall (in, cm)
Q = storm runoff depth (in, cm)
Tr = storm duration (hr)
a = a units conversion constant 1.01 English, 0.028 metric.
For the development of the above equation see Haith (1980).
EXAMPLE III-2
Assessing Single Event Sediment Loads and Storm Runoff
The 180 acres (72.8 ha) of cropland in the previous example will be
considered here. The USLE parameters are the same. Assume a 4.5 cm rain
falls in a 3 hour period in early June. The soil hydrologic condition is
good and the soils are in hydrologic group A. This particular storm was
preceded by 4 cm (1.6 in) in the previous five-day period. Calculate total
storm runoff and sediment yield.
207
-------
TABLE 111-10
ANTECEDENT RAINFALL CONDITIONS AND CURVE
NUMBERS (FOR I =0.2S)
a
Curve Number
of
Condition II
Factor to Convert Curve Number
for Condition II to
Condition I
Condition III
10
20
30
40
50
60
70
80
90
100
Condition
I
0.40
0.45
0.50
0.55
0.62
0.67
0.73
0.79
0.87
1.00
General Description
Optimum soil condition
2.22
1.85
1.67
1.50
1.40
1.30
1.21
1.14
1.07
1.00
5-Day Antecedent Rainfall
in inches
Dormant Growing
Season Season
<0.5 <1.4
II
III
from about lower plastic
limit to wilting point
Average value for
annual floods
Heavy rainfall or light
rainfall and low tem-
perature within 5 day
prior to the given storm
0.5-1.1
1.4-2.1
Source: Schwab et. al., 1956
208
-------
Solution:
First the total storm runoff volume will be estimated. The curve
number from Table III-9 is (row crops, contoured, good, A) 65 since
antecedent rainfall (growing season) is between 1.4 and 2.1 inches the
antecedent condition is II and no correction is needed. Therefore using
Equation III-7,
Oj • 2.!
- 10 • 2.54 = 13.68
\ UJ
Using Equation III-5
Q =[4.5 - 0.2 (13.68)]2/[4.5 + 0.8 (13.68)]
= 0.20
The storm runoff volume is (Eqn. III-6)
V = 100 (72.8) (.2)
= 1456 m3
Now the peak runoff is estimated from Equation III-8
q = 0.028 (72.8) (4.5 }_(Q.2Q]_
(TTT475 - 0.2 T13.68J]
= 0.35 m3/sec
The watershed sediment yield for the storm is (Eqn. III-4)
Y(S)E = 11.8 [1456 (.35)]0'56 (.37) (1.08) (.49) (.60)
= 45.5 tonnes - sediment
. END OF EXAMPLE 111-2-
209
-------
3.2.3 NITROGEN LOADING FUNCTION
While the complex interactions in soil, air, water and plants are
reasonably well understood, methods for quantifying movements within the
system are still in the research stage. Methods which are suitable for
general use often oversimplify the problem. They must be used with
discretion and may be inadequate in certain cases. For instance, it is not
presently possible to describe leaching processes for soluble forms of
nitrogen in a simplified manner. The nitrogen loading function assumes that
erosion is the primary N source for cultivated land. The loading functions
exclude leaching losses, and predict the amount of total nitrogen that is
released to surface waters by runoff and erosion. For predicting N and P
loads from forested areas see Section 3.2.7.
3.2.3.1 Nitrogen Loading Function for Erosion Loss
The nitrogen loading due to erosion is computed as:
Y(NT)E = a • Y(S)E • C$(NT) • rN (III-9)
where
Y(NT^ = total nitrogen loading due to erosion, (kg/yr or event,
Ib/year or event)
a = dimensional constant (10 for metric units, 20 for English
units)
C (NT) = total nitrogen concentration in soil, g/100 g
O
Y(Sl = sediment loading from surface erosion, (MT/year or event,
tons/year or event)
r = nitrogen enrichment ratio
Available nitrogen can be obtained by using a fraction fN which is the
ratio of available N to total N loss predicted by erosion. Thus, the
available nitrogen load is
-------
Y(NA)E = Y(NT)£ . fN (111-10)
3.2.3.2 Evaluation of Parameters in the Nitrogen Loading Function
In the erosion nitrogen loading function three parameters must be
evaluated. The value of Y(SL comes from the USLE or MUSLE depending upon
whether long term or single event-loads are being estimated.
Values of the nitrogen enrichment ratio r., are shown in Table 111-12.
These values range from roughly 2.0 to 5.0. It is generally higher on sandy
soils and lower on finely divided highly credible soils. The enrichment
ratio represents the effects of several processes which cause the nitrogen
content of the eroded soil to be higher than the source soil back in the
watershed. This is mainly due to preferential detachment and transport of
smaller soil particles which have higher N associated with them.
The user may want to use a low and a high value for K. to evaluate a
range of erosion nitrogen losses. The enrichment ratio is quite variable
with regard to management practices even for the same soil type.
The enrichment ratio also varies from storm to storm and within storms.
Menzel (1980) found that r., varied logarithmically with sediment loss. His
equation is
In [rN] = 2.0 + 0.20 ln[Y(S)£] (III-11)
where
Y is the sediment loss in (Kg/ha).
This equation can be used only on a per event basis.
The value of CS(NT) in the plowed layer of soil is variable from
location to location and from season to season. Estimates of native soil
nitrogen in the U.S. indicate a range between 0.02 and 0.4 percent (Jenny,
1930). Parker e_t at. (1946) published a map showing the nitrogen content
in the top 1-ft layer in the U.S. (Figure 111-10). Data in this figure
211
-------
TABLE 111-12
ENRICHMENT RATIOS FOR NITROGEN
Soil
Collington
Sandy Loam
Almena Silt Loam
Fayette silt Loam
Silt loam soils
Conditions Slope
Prevailing Practice
Conservative Practice
Control, Total N 3.5%
Cover crop
Manure
Cover & manure
Total N
Total N 3.0%
Total N 11.0%
rN
1.74
2.08
3.88
4.09
4.28
3.35
5.0
1.34
1.08
2.7
Reference
Stoltenburg &
White (1953)
Knoblauch,
Ko Today and
Brill (1942)
Neal (1944)
Massey, Jackson
& Hays (1953)
Massey & Jackson
(1952)
212
-------
ro
i—•
CO
NITROGEN
Percent N
Highly Diverse
Insufficient Data
Under 0.05
0.05-0.09
I'vfl 0.10-0.19
0.20 and Over
FIGURE 111-10 PERCENTAGE NITROGEN (N) IN SURFACE FOOT OF SOIL (PARKER, EJ_ AL,, 1946)
-------
should be viewed in general terms; for specific sites, local sources such
as ASCS and SCS Soil Survey should be consulted.
Precipitation also contributes to the soil nitrogen. Atmospheric
nitrogen extracted by soil microbes becomes incorporated into soil organic
matter. Animal manures, crop residues, and other wastes contribute
significant amounts of nitrogen to the soil. Jenny (1930) expressed the
nitrogen content of the soil in terms of temperature, T, and a humidity
factor, H. Jenny's equation is:
CS(NT) = 0.55/0-081" d-e-0'005") (111-12)
where
P = precipitation, mm/year
CS(NT) = concentration of soil nitrogen, g/100 g
T = annual average temperature, C
RH = relative humidity, percent
SVP. = saturated vapor pressure at given temperature,
mm of Hg
Equation 111-14 shows the relation between SV?t and T (Gladstone, 1946).
SVPt = 10 [9-2992 -2360/(273 + T)] (III
The solution of Equation 111-12 is shown graphically in Figure III-ll.
The value of humidity factor, H, can be determined from Equations 111-13 and
111-14. A nomograph solution of H is shown in Figure 111-12. For given
values of precipitation, relative humidity and temperature, the value of H
can be quickly and accurately established from Figure 111-12. For example,
given P, = 500 mm/year (19.7 in/year), RH , = 60 percent, and
T. = 5 t (41 F), the value of H factor can be determined as follows: using
a straight-edge ruler, align Pj and RH ^ to intersect on the index line at
"A" as shown on the inset of Figure 111-12. Align "a" with T on the
214
-------
0.01
100 200 3OO 400 500
H. Humidity Factor
600 700
FIGURE III-ll SOIL NITROGEN vs, HUMIDITY FACTOR AND TEMPERATURE
215
-------
ro
80 -t- 2000
70-
60-
50-
40-
30 i
•
20 -=
15-
i.
^ K>-
£ 9 -
o 6-
5- 5~"
O M
• 4 -
£
oT
2-i
1.5-
1.0-
8'.8-
0.7-
0.6-
0.5-
- •
-1500 j
X
-1000 I
- 800
- TOO
- 600
-500
r 400
- 300 .-
f. 200 E_
c
: o
r 150 3
'o.
-100 «
- 70 •>•"
: 60
-50
r 40
- 30
•
r 20
,
: |5
•
100-
•
•
•
OK _
•
„_
§ 90-
I 1
>? 85-
| 80-
x. :
,
«
1 TOJ
^
I :
o: 60 ^
50-
40-
30-
20-
-10,000
-5,000
4|OOO
- 3,000
- 2,000
-1,000
-5OO
- 300
-200 -
0
-too £
— 50 ^
r40 2
-30 |
-20 x
-10
•
- 2
_(
u P
" " ( 1 - RH/KX))SVPT
F9299- 236°1
SVPT=IOL 273TT]
where
H = Humidity Factor
P = Precipitation, mm/yr
RH = Relative Humidity, Percent
SVPy= Saturated Vapor Pressure at
Given Temperature, mm of Hg
T = Average Temperature, °C
P! S^Q
xT
^s.
RH|
A. Aleti
|HI 1
Midwest Research Institute -
0.4 -1— IU iv
Ti
1 1
8/75
-5-
o
5-
o
e
. 10-
«
h_
3
1 15-
a.
E :
d
*~ 20^
H"
25-
30 -
ns-
-30
r40
r u.
e
-50 f
-60 •
: J-
r70 K-
•_
'
r80
Lgn
FIGURE 111-12 NOMOGRAPH FOR HUMIDITY FACTOR, H
-------
temperature scale to intersect the H scale. The result on the H scale is
194.
Data in Figure 111-10 may be used as a check on current data.
Equations 111-12 and 111-13 may be used to calculate nitrogen content of
soil more precisely if necessary data are available for using these
equations. Again data from State Agricultural Experiment Stations, and SCS
Soil Surveys are much more dependable than the above equations and should be
consulted whenever possible.
The fraction of available nitrogen (f ) or that nitrogen which can be
directly used by plants is usually considered to be the NH4+ + NOs fraction.
The erosion nitrogen load which consists primarily of organic N bound in
detritus does not necessarily correlate well with the NH4 or NOs load
especially in areas with low surface runoff and sediment loss. Forms of
nutrients on the watershed surface will vary with
1. amount and type of plant residue remaining in the field
(Timmons et aj_., 1970)
2. land use (Logan, 1980)
3. application of manures and fertilizers (Reddy, 1980) and
4. type of tillage management (Frere, ejt aj_., 1980)
It is suggested for screening purposes that the user rely on total N load
estimates.
3.2.4 PHOSPHORUS LOADING FUNCTION
Phosphorus occurs naturally in soil from weathering of primary
phosphorus-bearing minerals in the parent material. Additions of plant
residues and fertilizers by man enhances the phosphorus content of the
surface soil layer.
217
-------
ro
1—>
CO
\
PHOSPHORIC ACID
Percent P2®5
0.0-0.04
0.05-0.09
0.10-0.19
0.20-0.30
FIGURE 111-13 PHOSPHORUS CONTENT IN THE TOP 1 FT OF SOIL (PARKER, ET_ AL,, 1946)
-------
Phosphorus in soils occurs either as organic or inorganic phosphorus
The relative proportion of phosphorus in these two categories varies widely.
Organic phosphorus is generally high in surface soils where organic matter
tends to accumulate. Inorganic forms are prevalent in subsoils. Soil
phosphorus is readily immobilized due to its affinity to certain minerals.
In strongly acid soils the formation of iron and aluminum phosphates, and in
alkaline soils, the formation of tricalcium phosphate reduces the
availability of soil phosphorus. Once it enters a stream, the partioning of
phosphorus between the sediment and solution phases becomes significant in
the nutrition of aquatic microorganisms.
Phosphorus transport from a given site to a stream can occur either by
erosion or by leaching. The predominant mode of transport is via soil
erosion. The soil solution usually contains less than 0.1 yg of phosphorus
per milliliter; the leaching losses are thus extremely low even in
well-drained soils. Exceptions are sands and peats which have little
tendency to react with phosphorus.
The loading function for phosphorus is based on the soil erosion
mechanism. The loading function is:
Y(PT) - a-Y(S)E.Cs(PT).rp (111-15)
where
Y(PT) = total phosphorus loading, (kg/year, Ib/year)
a = a dimensional constant (10 metric, 20 English)
Y(S) = sediment loading, (MT/year, tons/year)
CJPT) = total phosphorus concentration in soil, g/100 g
r = phosphorus enrichment ratio
Available phosphorus may be computed as in Equation III-ll:
Y(PA) = Y(PT)-f (111-16)
219
-------
where
Y(PA) = yield of available phosphorus, (kg/year, Ib/year)
f = ratio of available phosphorus to total phosphorus
3.2.4.1 Evaluation of Parameters in the Phosphorus Loading Function
As with the nitrogen erosion loading equation two parameters must be
evaluated - the soil concentration of total P and the phosphorus enrichment
ratio. The sediment load, Y(S) , is known from previous analysis.
Local sources should be consulted in preference to using the estimates
of CS(PT) in Figure 111-13. The Soil Conservation Service or agriculture
extension personnel will be the best sources of this information.
Table 111-13 shows enrichment ratios for phosphorus found in the
literature. On the whole r is slightly less than r having values of about
1.0 to 4.0. Sharpley (1980) and Menzel (1980) have both used equations of
the form
ln[r ] = a + b In [Y(S)r]
P h
(111-17)
where
a = 2.48
b = 0.27
a = 2.0
b = 0.20
(Sharpley, 1980)
and
Menzel (1980)
to predict enrichment ratios for individual storm events.
As with nitrogen the available to total P ratio varies with soil type,
crop type and management practice. Users should use total P in subsequent
analyses whenever possible.
220
-------
TABLE 111-13
ENRICHMENT RATIOS FOR PHOSPHORUS
Soils
Conditions
Slope
Prevailing Practice
Collington
Sandy
Loam
Conservative
Check plot
Cover crop
Manure
Practice
3.5%
Manure & cover crop
Almena silt loam
Fayette silt loam
Dun™ re silt loam
Total P
Available P
Available P
Available P
Available P
Available P
Available P
corn oats 3.0
of a
corn-oats 11.0%
hay-hay
rotation
Wheat) dilute 5-25%
|H2S04
Corn ' sol uble
r
P
1.82
2.03
1.59
1.56
1.47
1.47
3.1
0.99
1.92
2.20
3.74
1.79
3.4
Reference
Stoltenberg &
White (1953)
Knoblauch,
Ko Today &
Brill (1942)
Neal (1941)
Massey, Jackson
& Hays (1953)
Rogers (1941)
Massey & Jackson
1952
221
-------
Schuman, ejt al_. (1970) have reported an empirical relation between
sediment phosphorus (concentration in ppm, C (PT) and soluble phosphorus
O
(concentration in ppm, C (P) ) for Iowa soils. The relation may be stated
as:
CQ(P) = a + b-C$(PT) (111-18)
where a and b are regression coefficients. The reported values of a and b
are 0.018 and 0.047, repectively. Equation 111-18 shows that the ratio of
solution phosphorus to sediment phosphorus is just under 1 to 20.
Taylor (1967) suggested that about 10 percent of the total phosphorus
in eroded soil is ordinarily available for aquatic plant growth. However,
some values as high as 20 to 30 percent have been suggested.
3.2.5 Organic Matter Loading Function
The loading function is:
Y(OM)E = a-Cs(OM).Y(S)E.rOM (111-19)
where
Y(OM) = organic loading, (kg/year, Ib/year)
a = a dimensional constant (10 metric, 20 English)
CJ(OM) = organic matter concentration of soil, g/100 g
Y(SL = sediment loading, (MT/year, tons/year)
rOM = enrichment ratio for organic matter in eroded soil
3.2.5.1 Evaluation of Parameters in the Organic Matter Loading Function
The value of Y(S)E can be obtained from procedures discussed
previously. The value of C$(OM) should be obtained preferably from current
or historical data for a given area, (e.g. from the extension service).
For approximate values, CJOM) may be taken as equal to 20 x CS(NT), where
222
-------
is the total nitrogen concentration in the soil (Buckman and
Brady, 1969).
The enrichment ratio for organic matter rQM has a range of 1 to 5 being
higher on sandy soils and lowest on finely divided highly erodible mineral
soil (McElroy et al_., 1976). Foster (1980) gives a range of rQM of 1.2 to
4.4. Frequently it is assumed that rQM is that of the clay enrichment
ratio. The data of Leonard et §_]_. (1979) show that this was approximately
the case for four watersheds in the Piedmont region of Georgia. Enrichment
rations for organic matter found in the literature are shown in
Table 111-14.
Obviously, the effects of management practices can be pronounced on
r. One would expect higher rgM with practices such as residue cover and
slope reduction. In general, practices which reduce runoff velocities and
rainfall detachment probably increase r.
EXAMPLE II1-3
Example of Loading Computation for
Nitrogen, Phosphorus, Organic Matter
The watershed used in Example III-l (Section 3.2.4) for Parke County in
Indiana will be used to illustrate the methodology presented in this section
for computing pollutant loads. Computation of available nitrogen, available
phosphorus, and organic matter annual loads is required:
The following data, plus soils data, are required:
Soil nitrogen content.
Soil phosphorus content.
The preferred source of data is local records. Jenny's equation
(Equation. 111-12) and Figure 111-13 are alternate sources from which
general values may be estimated.
223
-------
TABLE 111-14
ENRICHMENT RATIOS FOR ORGANIC MATTER IN SURFACE RUNOFF
Soil
CoTlington Sandy
Loam
Almena Silt Loam
Conditions Slope
Check plot 3.5%
Cover crop
Manure
Cover + manure
Prevailing Practice
Conservation Practice
Corn oats of a corn 3%
oats-hay-hay rotation
r
Offi
4.13
4.48
4.23
3.97
1.24
1.38
4.7
4.6
Reference
Knoblauch,
Koloday
and Brill
(1942)
S to. 1 ten berg &
White (1953)
Neal (1944)
Massey,
Jackson &
Hays (1953)
Fayette Silt Loam
Silt Loam Soils
Sandy Loam Soils
Corn oats of a corn 11%
oats-hay-hay rotation
Contoured (PI) 2-5%
Countoured (P2) 2-4%
Terraced w/grassed (P3) 1-2%
Waterway
Terraced w/grassed (P4) 1-2%
Waterway
1.24
1.15
2.6
2.1
2.4
1.9
Massey & Jackson
(1952)
Leonard,
Langdale &
Fleming(1979)
* Values are averages for 4, 1, 3 and 1 surmier storm(s) respectively for PI.
P2, P3, P4 watersheds.
224
-------
Nitrogen Loading
Using the following data, soil nitrogen content is calculated:
Average annual temperature = 10°C
Average annual precipitation = 96.5 cm
Average annual relative humidity = 70 percent
Using the nomograph given in Figure 111-12, the value of the H factor
is determined to be 350. From Figure III-ll, and using H = 350 and
T = 10°C, the value of C_(NT), the soil nitrogen content is estimated to be
0.204 percent or 0.204 g/100 g. Using Equations III-9 and 111-10 and rN =
2.0,
Y(NA)E = 20-Y(S)E 0.2-2.0
= 8-Y(S)E
The values of area! sediment yield as given in Example III-l are shown
below in Table 111-15.
Phosphorus Loading
Assuming C (PT) = 0.15 g/100 g for the area and r = 1.5, equation
111-10 gives
Y(PA)E = 20-Y(S)E-0.15.1.5
- 4.5 Y(S)£
Organic Matter Loading
Using Equation 111-19, data for CS(OM), Y(S)E, rQM are needed.
Assume that the value of CS(OM)/CS(NT) equals 20 and rQM =2.5,
225
-------
TABLE II1-15
CALCULATED SEDIMENT, NITROGEN, PHOSPHORUS AND ORGANIC
MATTER LOADS FOR PARKE CO., INDIANA WATERSHED
Land Use
Sediment
Load (tons/year)
Total Total Organic
Nitrogen Phosphorus Matter
Cropland
Pasture
Woodland
2530
121
430
10.1
0.5
1.7
5.7
0.3
1.0
253
12
43
Total
2797
12.3
7.0
308
226
-------
Y(OM)E = 20-2.5-Y(S)E-20-Cs(NT)
- IOOO-CS(NT)-Y(S)E
Us
ing Cc-(NT) = 0.2 percent.
Y(OM)E =200 Y(S)E
The values for nitrogen, phosphorus organic matter loading are also
presented in Table 111-15.
END OF EXAMPLE III-3
3.2.6 Accuracy of Nutrient and Organic Matter Loadj_n_g_s__fV_qm_Er_qs_i_on
The accuracy of a prediction using loading functions depends upon the
accuracy in predicting sediment loading, watershed concentration (or
availability) of a pollutant, and its enrichment in eroded sediments.
Predicting sediment loss has been and continues to be a problem in Doth
hand calculation methods and computer models. Prediction of sediment loads
is generally better for longer periods; that is, annual predictions are
better than seasonal predictions which are better than for single storm
events. Storm event sediment loads are particularly a problem for hand
calculation methods such as the USLE but the use of MUSLE improves the
accuracy.
Aside from sediment load prediction, the dynamic surface concentration
of nutrient forms is also difficult to estimate unless computer simulation
techniques are used. Frere et aj_. (1980) have indicated that typical values
for soil nitrogen and phosphorus vary from 6 to 10 fold.
227
-------
Since most enrichment ratios observed for organic matter, N and P have
values between 1 and 5, their estimation seems the least critical of the
three parameters of the loading function.
Total nutrient losses for N and P are among the most accurate for
agricultural watersheds. This is primarily because most of the N lost is
organic which is bound to fines or is itself detrital. Similarly, P is
mostly particulate in cultivated watersheds. The prediction of individual
forms of either N or P is less accurate because of the chemical
transformation which must be considered and is impractical using hand
calculator methods.
Table 16 shows the percentage of explained variance for several
parameters by regression on suspended solids from several different parts of
the country. The information is only for regressions which were linear with
respect to both the independent and dependent variables. As expected, TKN
and Total P are among the highest, with explained variance decreasing for
more weakly adsorbed constituents.
3.2.7 Nitrogen and P hos phor u s L pad i_n_g F r om F or es ted _Wa t e r s hed s
Because of the orotection of the soil surface afforded by tree canopy
and tree litter in forested watersheds, the quantity of soil eroded from the
watershed surface is generally small. Contrast the two order of magnitude
differences in sediment loading between cropland and woodland computed in
Example III-l. Hence the mechanism by which nutrients are lost from these
watersheds are guite different from those erosion mechanisms which operate
in agricultural watersheds. For this reason the erosion loading equations
are not recommended for use in forested systems.
An empirical procedure is given below for estimating N outputs for
forested basins based on precipitation input load. The equation is
Y(N)r = A-N -b (111-20)
\ /p pr v /
228
-------
TABLE 111-16
VARIATION IN CONSTITUENT ACCOUNTED FOR BY REGRESSION
ON SUSPENDED SOLIDS (Linear Models Only)
Constituent
Data Base
Explained
Variance
Nutrients
TKN
Total P
NH^-N
NO^-N
Miscellaneous
Watkinsville, GA
Buffalo Bill, 10
Michigan State, MI
Honey Creek, OH
Watkinsville
Honey Creek
Watkinsville
Buffalo Bill
Watkinsville
Buffalo Bill
Michigan State
Buffalo Bill
Michigan State
37.1
11.1
2.1
19.4
18.7
67.9
6.9
5.6
18.5
15.0
5.8
13.8
0.06
BOD
Iron
Buffalo Bill
Honey Creek
32.0
82.8
Source: Zison, 1980
229
-------
where
Y(N)p = the loading of N from a forested watershed (Kg, Ib)
A = the watershed area (ha, acres)
N = the input precipitation nitrogen load (kg/ha, Ib/acre) and
b = an attenuation factor
An alternative form of Equation 111-20 is
Y(N)p = A-C(N)pr-Q(Pr)-b-a (111-21)
where
C(N) = a typical N concentration in precipitation (mg/1)
Q(Pr) = the depth of rainfall over the period (in, cm) and
a = a units conversion factor
= 0.23 English units
= 0.10 metric units
Equation 111-21 should be used if local and reliable N precipitation
concentration data are available. If not Equation 111-20 can be used. The
map in Figure 111-14 or the data in Table 111-17 can be used to estimate N
in this case.
The attenuation factor 'b' can be estimated using Table 111-18. On the
average the attenuation factors for N and P are 0.40 and 0.78 respectively.
The attenuation factors are calculated from the data in Likens, et al.,
1977.
3.2.8 Loading Values for Salinity Loads in Irrigation Return Flow
Perhaps the most useful method of establishing salinity loads is
through loading values determined for particular regions. Lists of such
values are presented in Tables 111-19 through 111-23 for subbasins in the
Colorado River basin, and for irrigated regions in California.
Studies in the Twin Falls area and the Colorado River basin indicate
that the range of values for salt pickup from irrigated lands is roughly 1.3
to 22 MT/ha/year (0.5 to 8 tons/acre/year) (Skogerboe and Law, 1971). These
230
-------
.3kg/ha/yr
.5 kg/ha/yr
2.0 kg/ha/yr
'.0 kg/ha/yr
ro
CO
1.0 kg/ha/yr 1.5 kg/ho/yr 2.0 kg/ha/yr
1.5 kg/ha/yr
1.0 kg/ha/yr
1.0 kg/ha/yr
|.5kg/ha/yr
FIGURE 1 1 1-
NITROGEN (NH^-N AND NO^-N) IN PRECIPITATION, (PERSONAL COMMUNICATION
WITH MRL J.H, CRAVENS, REGIONAL FORESTER, U,S,D,A,-FS EASTERN
REGION, 1974)
-------
TABLE 111-17
ATMOSPHERIC CONTRIBUTIONS OF NITROGEN AND PHOSPHORUS IN RAINFALL
ro
CO
ro
N Contribution in Kq/ha/yr
N03-N+NH4-N Total N
Northeastern U.S.
Southeastern U.S.
Midwestern U.S.
West/Southwestern U.S.
United States
Low.
5.7
1.5
0.2
1.7
-
High, Low High.
12.1 5.7 12.1
12.3
20.9 1.7 20.9
5.7 9.0 14.6
-
P Contribution in Kg/ha/yr
Inorganic P Total P
Low^ High. _Lqw Hj^Jl
_
_
- - -
- - -
0.18 0.18 0.08 0.80
Source: Weiner, et al. (1976)
-------
TABLE 111-18
NUTRIENT BUDGETS FOR VARIOUS TERRESTRIAL ECOSYSTEMS OF THE WORLD (kg/ha-yr)
ro
CO
co
la&iM Mill, .^lo.pem
Teryei d Uf_ fcostly coniferous and
evergreen forest
Temperate bcnj vegetation
TropJcJl ingioipprm, mostly
evergreen forest
Tempe ra te ncstly angiospern
ana deciduous forest
TCTiueraU mostly coniferous and
evergreen forest
Temperate bog vegetation
Tropical dngtQspenn mostly
MA| ireta.iorpnic (* refers to subsc
JTo water table
Calculated frun Mg inputs and Na
lNVN *"VN
j
"Wrije for $ yr l'J/1-1975
"Calculated fran B9 cm of precipit
"•Trace •
L OL j 1 1 wn
Hubbard Bruoi, U S
Pago Cdtcnront, Australia
Silverstrean. Nc« Zealand
Tiugrannoa Creek. NT. U S
Walter Srdncn. TE . U S
Blue Rjnge Caurn.*nt, Australia
Boundary haters Canje Area, MlN,
U S (24)
Carnation Creek. Vancouver Island,
Canada
Cedar River. WA. U S
Clear Lake, Ontario, Canada (25)
Finland
Western Cascades Range, OR. U S
Rio Negro, Brazil
Kuokket area. Sweden
Cosnocton. OH, U.S. (2!)
Huboard Brook, U.S
S E. U S.
Stlverstream. New Zealand
Ta-ghannocK Creek, NY, U.S.
Walker Branch, TE , U S
Birkenes Watershed, Norway
Canada
Cedar River, HA, L! S.
ELA, Ontario, Canada
Finland
Storsjon, Sweden
Velen. Sweden
(22) 9
Rough Sike Catchment. England
Rio Negro, Brazil
ation times concentration (0 02 mg P/l , lay)
Precipitation
Input
0 036
0 33
0 I
0 07
0 54
0 39
0 14
0 11
0 35
a i
0 14
0 29P
0 2
0 055*
20
2 2.
9 71
a 7
H.SJ
1 1
6 4
6
8 2
5.6
1 15
or et «1 , 1971)
samples
s<~-
PMsphorus
0 019
0 26
0 03
0 ?0
0 0^
0 42
0 (,15
0 05
0 0?
0 09
0 3
0 Q?
0 51P
0 l'
0 041*
nitrogen
4 O1
1 B.
5 61
1 8
0 9
LV
30f
4 79
097
»-,•;-
• 0 OP
«0 07
• 0 2
-0 13
*0 52
-0 03
tO 13
*0 06
-0 2
*Q 12
-0 22
+0 1
+0 014
ti7 5
*0 4
+ 4 1
*6 9
+ 12 3
»0 5
+5 5
+4
+ 7 7
+5 5
+5 2
+0 9
AM.ual
precipitation,
cat
89
130
150
135
96
155
132
70
315
136
90
57
87
200
213
75
89
130
127
135
96
155
134
136
83
5?
67
72
213
200
, Trp, tropical
DOfftfUAt
vegetation
Ac. F, 6
E
N
Ac, Ti. Ts
g. c
Pin
Pin, Pic, &
At), Ts, Th
Ps
Ac, F, Q
Pin, Pic. B
Pin, Pic
Trp
Ca. En, B
L, Pin
Ac, F. B
Q. Pin
N
Ac. Ti. Ts
0, C
B, Pi Pic
's
-in! c, B
'in, c
1 1 11 , C
Bog
Trp
species
amis tone,
Source Likens
G.OIO,/ Aiteao.l,
Jshr^cr Ss •»
n9;; i" .79
S5hrS ' M M 15
S5' 5hr ir s" l.K
Sjhr c 04
I H 93
'gr S!h "
"9r Hsr "c -'5
Is H .it
Igr M9 16
Iqr M' 3ioo
I9r 9 .14
'cr si sh
S M H .SO
H 75
Grand Avg ,
!"»-«V ^ :!?
$?£«„,„ •»
ssr jhr 9r sh 58
S^ c 21
I 15
S 54
I M 14
I9r H9 .33
,gr 9 23
M9 I 07
,gr 9 6S
't
Scr 5si 5sh X
SShr "«r », «
Grand Avg
ejil (1977)
on Fattur |t<}
Avg 77
- 0 78
Avg - M
= 40
-------
IN3
CO
-p.
TABLE 111-19
SALT YIELDS FROM IRRIGATION • IN GREEN RIVER SUBBASIN (EPA, 1971)
Average salt yiel
Area
Green River above New Fork River
Big Sandy Creek
Blacks Fork in Lyman area
Hams Fork
Henry's Fork
Yampa River above Steamboat Springs
Yampa River, Steamboat Springs to Craig
Mill Creek
Williams Fork River
Little Sanke above Dixon
Little Sanke, Dixon to Baggs
Ashley Creek
Duchesne River
White River below Meeker
Price River
San Rafael River
(tons/acre/yr)
0.1
5.6
2.4
0.3
4.9
0.2
0.4
0.1
0.3
0.3
0.5
4.2
3.0
2.0
8.5
2.9
(kq/ha/day)
0.6
34.3
14.7
1 .8
30.1
1.2
2.5
6.1
1.8
1.8
3.1
25.8
18.4
12.3
52.2
17.8
d
(Ib/acre/day)
0.5
30.7
13.2
1.6
26.9
1.1
2.2
5.4
1.6
1.6
2.7
23.0
16.4
11.0
46.6
15.9
-------
TABLE 111-20
SALT YIELDS FROM IRRIGATION IN UPPER COLORADO MAIN STREAM SUBBASIN (EPA, 1971)
GO
en
Area
Main stern above Hot Sulphur Springs
Main stem, Hot Sulphur Springs to
Kremml i ng
Muddy Creek Drainage Area
Brush Creek
Roaring Fork River
Colorado River Valley, Glenwood Springs
to Silt
Colorado River, Silt to Cameo
Grand Valley
Plateau Creek
Gunnison River above Gunnison
Tomichi Creek above Par! in
Tomichi Creek, Parlin to mouth
Uncompahgre above Dallas Creek
Lower Gunnison
Naturita Creek near Norwood
(tons/acre/yr)
0.3
0.9
2.4
0.7
3.5
2.3
3.5
8.0
0.9
0.3
0.3
0.3
4.5
6.7
2.8
Average salt yield
(kg/ha/day)
1.8
5.5
14.7
4.3
21.5
14.1
21.5
49.1
5.5
1 .8
1.8
1.8
27.6
41.1
17.2
(Ib/acre/day)
1.6
4.9
13.2
3.8
19.2
12.6
19.2
43.8
4.9
1.6
1.6
1.6
24.7
36.7
15.3
-------
PO
00
en
TABLE 111-21
SALT YIELDS FROM IRRIGATION IN SAN JUAN RIVER SUBBASIN (EPA, 1971)
Average salt yield
Fremont River above Torrey, Utah
Fremont River, Torrey to
Hanksville, Utah
Muddy Creek above Hanksville, Utah
San Juan above Carracas
Florida, Los Pinos, Animas drainage
Lower Animas Basin
LaPlata River in Colorado
LaPlate River in New Mexico
(tons/acre/yr)
0.4
5.8
3.1
2.7
0.2
3.5
1.4
0.3
(kg/ha/day)
2.5
35.6
19.0
16.6
1.2
21.5
8.6
1.8
(Ib/acre/day)
2.2
31.8
17.0
14.8
1.1
19.2
7.7
1.6
-------
TABLE 111-22
SALT YIELDS FROM IRRIGATION IN LOWER COLORADO RIVER BASIN (EPA, 1971)
Average salt yield
Area
Virginia River
Colorado River Indian Reservation
Palo Verde Irrigation District
Below Imperial Dam
(Gila and Yuma projects)
(tons/acre/yr) (kg/ha/day)
2.3 14.1
0.5 3.1
2.1 12.9
variable
(Ib/acre/day)
12.6
2.7
11.5
-
ro
CO
TABLE 111-23
SALT YIELDS FROM IRRIGATION FOR SELECTED AREAS IN CALIFORNIA
(WATER RESOURCES COUNCIL, 1971)
Average salt yield
Area
North coastal
Central coastal
Sacramento
Delta-Central Sierra
San Joaquin
Tulare
Colorado Desert
(tons/acre/yr) (kg/ha/day) (Ib/acre/day)
0.353
0.808
0.707
0.974
0.827
0.768
10.9
2.2
5.0
4.3
6.0
5.1
4.7
67
1 .9
4.4
3.9
5.3
4.5
4.2
60
-------
values are site specific. An average salt pickup rate might be 5 MT/ha/year
(2 tons/acre/year). On a per day basis, the range becomes 3 to 50 kg/ha/day
(3 to 44 Ib/acre/day), and the average becomes 12 kg/ha/day (11 ]b/acre/day).
The most common means of expressing the salinity of water is its
electrical conductivity. Conductivity is the inverse of resistance (ohms)
and is expressed in units of "mhos".
The salt pick up in an irrigation water as it is diverted from a river
through a field and returns to the river can be measured by the change in
the conductivity of the water. The conductivity of the drainage water can
be determined from the conductivity of the irrigation water and the leaching
fraction (USDA, 1954) by
ECdw= ECiw/(LF) (111-22)
where EC, = drainage water conductivity (umho)
dw
EC = irrigation water conductivity (umho) and
iw
LF = leaching fraction
The usual range of the leaching fraction is about 0.1 to 0.3.
The conductivity of the irrigation water is related almost linearly to
the concentration of the salt. Thus the concentration of salt can be
determined from the conductivity and vice versa. The general model is
S = aK (111-23)
where
S is the dissolved solids concentration (mg/1)
K is the conductance (umho) and
a is a regression coefficient
The value of a is usually between 0.55 and 0.75 with higher values generally
occurring in waters having higher sulfate concentrations (Hem, 1970). An
average value is about 0.63.
238
-------
Another important parameter affecting the quality of water for
irrigation is the sodium adsorption ratio (SAR). This is defined as
[Na]
SAR = ^__=— (111-24)
/[Ca + Mg]
where
SAR is the sodium adsorption ratio (meq/1)
[Ca+Mg] is the concentration of calcium plus magnesium in the
irrigation water (meq/1) and
[Na] is the concentration of sodium in the irrigation water
(meq/1)
The SAR is a measure of the sodium hazard of irrigation water to
plants. Like the EC, the SAR of irrigation drainage water can be calculated
knowing the SAR of the irrigation water and the leaching fraction (Bower,
et aj[., 1968) by
SAR.
SARdw = ~~ (111-25)
where
dw denotes drainage water and
iw denotes irrigation water.
This equation assumes that no precipitation or dissolution of salts occurs
in the soil during irrigation.
3.2.9 Toxic Chemicals in Agricultural Environment
The primary group of toxic chemicals that will be of concern in
agricultural or forested settings will be the pesticides including
herbicides, insecticides, etc. Metals and other organic materials may be a
problem if municipal sludges or landfills are of concern. These problems
will not be dealt with in this text however. In addition the discussion
239
-------
will be limited to those insecticides occurring on the EPA 129 priority
pollutant list.
While pesticides may be lost during application, enter the air and be
diffusely redeposited and washed off the land surface, this mechanism will
not be as great a pathway for loss as washoff from the field surface to
which the pesticide is applied. It will be assumed that the only sources of
pesticides will be slug applications to the surface in a short time period
following which the pesticide will disappear at some first order rate. It
is important that the amount of pesticide remaining on the watershed surface
be known at the time of the storm event. Another important quantity to know
is the amount of pesticide that is adsorbed onto soil materials versus that
quantity which is dissolved in the soil solution. This ratio greatly
affects the amount of pesticide lost in the runoff. Thus for toxic
pollutants three key processes must be described in order to simulate their
magnitude in runoff losses:
• the rate at which they accumulate or are deposited at the
watershed surface
• the rate at which they disappear from the watershed surface
and
• the ratio with which they partition themselves between the
dissolved and sorbed phases.
3.2.9.1 Disappearance of Pesticides from the Watershed Surface
Pesticides are lost from the watershed surface after application
through several important processes:
1. leaching
2. runoff
3. volatilization
4. degradation and
5. plant uptake
240
-------
Pesticides which have low water solubilities and are strongly adsorbed
to soil materials are most resistant to leaching. Chlorinated hydrocarbons
(e.g. aldrin, chlordane, DDT, endrin, dieldrin, endosulfan, heptachlor,
toxaphene) fall into these categories. They tend to be adsorbed
hydrophobically and to a greater extent in soils with high organic carbon
contents. Nonbiological degradation is thought to be a minor mode of loss
for chlorinated hydrocarbons (El Beit, jet aK, 1981) as well as chemical
degradation. They are degraded biologically however, especially under
anaerobic conditions (Kaufman, 1974).
Volatilization is apparently one of the major pathways of loss of
organochlorine pesticides from the soil. Many factors affect volatilization
rates including sorption, concentration, soil water/air flow rate,
temperature, diffusion and physical and chemical properties of the
pesticides.
The two factors that most affect plant uptake are polarity and water
solubility (Nash, 1974). Although polar, the chlorinated hydrocarbons are
relatively insoluble in water and plant uptake is probably not a great
mechanism of loss.
Usually runoff losses are greatest for those pesticides with long half
lives and those which are strongly adsorbed to soil materials. These
pesticides tend to remain on the surface for long periods of time where they
continue to be lost in eroded sediments.
Except for the processes of leaching and runoff losses which will be
described mathematically, this methodology for calculating pesticide runoff
losses will assume that the disappearance of pesticide from the soil surface
is first order. The rate constant k , for this disappearance is the sum of
the rate constants for all the individual processes of volatilization,
degradation, etc.
The equation which gives the quantity of pesticide in the soil surface
layer (1 cm) is (Haith, 1980):
241
-------
Pt = PTexp [-ks (t-i)] (111-26)
where
Pf is the pesticide remaining on day t
*
P is the amount of pesticide in the soil on the day of the last
T
rainfall event or pesticide application event
ks is the disappearance rate constant
t is the number of the day of the runoff event and
T is the number of the day of the last runoff event or
application event.
If the event on day t was a rainfall event then
P* - PT - PXT- (1-R/RT)DT (111-27)
or if the event was a pesticide application
P* = P + AP (111-28)
T T T
The parameters PX , 6, R and D are discussed in section 3.2.9.3.
Degradation rate coefficients (ks) can be estimated for most of the
chlorinated hydrocarbons on the priority pollutant list by choosing a
representative value from either Table 111-24 or Table 111-25.
3.2.9.2 Partitioning of Pesticides between Apjj__anjj _Water
Sorption refers to the removal of pesticide from solution by soil
materials. Adsorption mechanisms on clay minerals include cationic and
anionic exchange, hydrogen bonding and Van der Waals attraction. Adsorption
onto organic matter is a result of cation or anion exchange, hydrogen
bonding, and/or hydrophobic bonding.
242
-------
TABLE 111-24
VALUES OF ks FOR DISSIPATION OF PESTICIDES FROM SOIL SURFACES
Pesticide
Aldrin
Aldrin
(+dieldrin)
(granules)
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
Aldrin
(Dieldrin)
Aldrin
(Dieldrin)
Aldrin
(Dieldrin)
Soil
Type pH
Coachella fs
Carrington sil
Ulysses sil
Knox sil
Celeryville muck
Marietta si
Fox fsl
Miami sil
Muck
Carrington sil Nondisked
Carrington sil Disked
Udaipur cl 7.8
Jobner 8.6
Muck
Miami sil
Composite
Carrington sil Nondisked
Carrington sil Disked
Carrington sil Disked
OM conditions
(%)
Fallow
Fallow
Fallow
Fallow
Fallow
Fallow
Fa 1 low
Fallow
Fallow
1.6 Various
.26 Various
Granules
Application
rate
(kg/ha)
20.2
5.6
2.24
2.24
2.24
2.24
2.24
2.24
2.24
4.5
4.5
3.0
3.0
22.4
22.4
22.4
4.5
4.5
5.6
ks
'
0.2406
.0045
< .0032
.0264
.0259
.0014
.0136
.0256
.0258
.0066
.0101
.0136
.0149
x of 19
x of 19
.0165
x of 19
.0061
.0096
.0038
.0006
.0008
.0012
(continued)
-------
TABLE 111-24 (continued)
ro
Pesticide
Aldrin
(Dieldrin)
BHC
BHC
BHC
BHC
BHC alpha
BHC beta
BHC gamma
BHC delta
Chlordane
Chlordane
Chlordane
Chlordane
DDT
DDT
DDT
DDT
DDT
DDT
DDT
DDT
DDT
DDT
DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
Type
Carrington sil
Udaipur
Jobner si
Berwick si
Berwick si
Berwick si
Berwick si
Berwick si
Composite
Gullatin Valley
Gullatin Valley
si
Coachella fs
Houston c
Pima sic
Pinal gl
Blackwater River
Pollard Mountain
Mosquito Bnk Pod
Route 11
West Oxbow
Beach Mountain
Carrington sil
Carrington sil
Miami sil
Soil
PH
7.8
8.6
>7
>7
>7
<7
<7
<7
<7
<7
<7
Nondisked
Disked
OM conditions
(%)
Spray
1.6 Various
.26 Various
Vegetables
Vegetables
Vegetables
Vegetables
Alfalfa
Alfalfa
Forest
Forest
Forest
Forest
Forest
Forest
Fallow
Application
rate
(kg/ha)
5.0
5.0
7.4 BHC
7.4 BHC
7.4 BHC
7.4 BHC
2.0
2.0
>2
83
22.4
1.12
1.12
1.12
1.12
1.12
1.12
4.5
4.5
11.6
ks
.0017
.0021
.0140
x of 19
.0098
x of 19
.0006
.00015
.00042
.00036
.00072
.0020
.0101
.007
.004
.053
.0060
.0049
.0060
.00015
.000023
.00040
.00014
.00024
.00044
.0024
.0048
.0003
(continued)
-------
TABLE 111-24 (continued)
ro
-F>
en
Pesticide
p,p'-DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
o,p'-DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
p,p'-DDT
Dieldrin
Dieldrin
Dieldrin
Dieldrin
Dieldrin
Dieldrin
Endosulfan
Endrin
Endrin
Heptachlor
Heptachlor
Heptachlor
Lindane
Lindane
Lindane
Type
Carrington sil
Muck
Miami sil
Berwick si
Composite
Berwick si
Ulysses sil
Knox sil
Celeryville muck
Marietta si
Fox fsl
Miami sil
Muck
Commerce sil
Carrington sil
Carrington sil
Imperial sc
Holtville fsl
Composite
Various
Mhoon sicl
Coachella fs
Composite
Composite
Composite
Imperial sc
Holtville fsl
Composite
Soil
pH
Disked/non-
disked
6.9
6.8
4.9
6.0
7.2
7.1
6.8
Nondisked
Disked
7.8
7.8
6.0
7.8
7.8
OM
(%)
1.8
.8
74.5
2.0
.8
3.6
40.0
1.0
.5
1.2
1.0
.5
Crop or
conditions
Fallow
Fallow
Fallow
Fallow
Fallow
Fallow
Fallow
Fallow
Fallow
Fallow
Sugarcane
Application
rate
(kg/ha)
4.5
11.2
11.2
37 DDT
9.4
9.4
9.4
9.4
9.4
9.4
1.0
24.4
4.5
4.5
20.0
20.0
1.3
5.4
20
20
k
.0002
.0011
.0029
.00016
.0007
.00029
.0008
.0005
.0021
.0014
.0009
.0004
.0009
.0037
.0142
.0187
.0003
.0002
.0001
.0008
.0162
.0110
.2436
.0021
.0025
.0028
.0022
.0026
.0017
(continued)
-------
TABLE 111-24 (continued)
en
Pesticide
Lindane
Lindane
Lindane
Lindane
Lindane
Lindane
Lindane
Lindane
Lindane
Lindane
Lindane
Toxaphene
Soil
Type pH
Gila sil 7.7
Miami sil
Muck
Miami sil
Ulysses sil
Knox sil
Celeryville muck
Marietta si
Fox fsl
Miami sil
Muck
Galestown si 6.7
OM conditions
(%)
.6 None
Fallow
None
None
None
None
None
None
5.2 Cotton
Appl ication
rate
(kg/ ha)
11.6
11.2
11.2
1.12
11.2
11.2
11.2
11.2
11.2
11.2
(6x2.7)
ks
2/.0046
.0011
.0014
.0048
.0147
.0264
.0074
.0263
.0264
.0139
3/.0059
.0046
I/ si = silt; s = sand; sh = shale; c = clay; 1 = loam; f = fine; g = gravelly.
Source: Nash, 1980.
-------
TABLE 111-25
DEGRADATION RATES COEFFICIENTS FOR SELECTED PESTICIDES
Pesticide
Aldrin
Chlordane
DDT
Dieldrin
Endrin
Heptachlor
Lindane
(Y-BHC)
Conditions
Lab
Field
Field
Lab
Lab (anaerobic)
Lab
Field
Lab (anaerobic)
Field
Field (anaerobic)
Lab
Field
Lab
Lab (anaerobic)
k
s
0.013
0.0023
0.0024
0.00013
0.0035
0.013
0.0023
0.03
0.0015
0.0053
0.011
0.0046
0.0026
0.0046
%cv
100.
104.2
130.8
82.7
100.
53.3
119.6
Half
Life
53
1237
1214
1657
692
53
1237
31
460
130
63
426
266
151
Source: Rao and Davidson (1980)
247
-------
Pesticides can be broken down into two categories
1. ionic and
2. nonionic
If pesticides are ionic they are in general cationic, basic or acidic.
Their primary mechanism of sorption is through ion exchange when in their
ionic form. Chlorinated hydrocarbons are nonionic and their major mechanism
of sorption appears to be hydrophobic bonding. Therefore their sorption is
highly related to the amount of organic carbon in the soil (Woodard and
Weber, 1974).
For hydrophobic sorption the partition coefficient Kj can be most
successfully estimated using the octanol-water partition coefficient and the
following equations (Rao and Davidson, 1980):
log KQC = 1.03 log KQW - 0.18 (111-29)
and
K • %OC
K, = -flC (111-30)
d 100
where
K is the octanol - water partition coefficient,
ow
K is the organic carbon partition coefficient
\J \f
%OC is the percent soil organic carbon and
Kd is the adsorption partition coefficient (cm3/g)
Values of K , K and «d are given in Table 111-26 for some of the
chlorinated hydrocarbons on the priority pollutant list.
248
-------
TABLE 111-26
OCTANOL-WATER PARTITION COEFFICIENTS
FOR SELECTED PESTICIDES1
Pesticide
Aldrin
Chlordane
ODD
DDE
DDE n n
P>P
DDT
DDT P.P
Dieldrin
Endrin
Heptachlor
Lindane (y-BHC)
Toxaphene
Kow
2.1E3
1.2E5
7.3E4
4.9E5
3.7E5
1.5E6
4.9E3
1.6E3
7.4E3
6.4E2
1.7E3
Koc Kd
2.8
2.4E5 4.2E4
6.3E2
1.1E3 20.1
'•Kow and Koc from Rao and Davidson, 1980.
Kd from Pionke and DeAngelis, 1980.
Kd Lindane from Rao and Davidson, 1980.
249
-------
3.2.9.3 Determination of Pesticide Runoff Loss
To determine the magnitude of runoff loss the quantity of adsorbed and
dissolved pesticide is first determined. Total pesticide is the sum of the
adsorbed and dissolved fractions
Pt = \ + Dt (111-31)
The adsorbed quantity A is given by
At = [I/O + 9/K.p)] P. (111-32)
L d t
while the dissolved fraction, D. , is
+ d/e)J P
u
The loss of adsorbed pesticide is
Kdp/e P (111-33)
t r, (111-34)
N '^V\A Aj^*
and the dissolved pesticide loss is
PQt = CQ/I\.] Dt (111-35)
In the above equations
A = sorbed pesticide loss (Kg/ha, Ib/ac)
6 = available water capacity of the top cm of soil (difference
between wilting point and field capacity) (dimensionless)
p = soil bulk density (g/cm3)
P. = total pesticide concentration (kg/ha, Ib/acre)
250
-------
D = dissolved pesticide concentration (kg/ha, Ib/acre)
PX = sorbed pesticide loss (kg/ha, Ib/acre)
PQ = dissolved pesticide loss (kg/ha, Ib/acre)
Q = total storm runoff depth (in, cm)
R = total storm rainfall depth (in, cm) and
K . = sorption partition coefficient (cm3/g)
An example problem for determining the runoff losses of a chlorinated
hydrocarbon follows.
EXAMPLE III-4
Estimation of Lindane Loss
From an Agricultural Watershed
On 5 June an application of 5.0 kg/ha of lindane (Y-BHC) was made to
the corn crop on the watershed in Example III-l. There was lindane residue
of 1.0 Kg/ha already in the soil. On June 8 the rainfall event evaluated in
Example 111-2 occurred. Runoff from this storm was 0.20 cm and the total
sedirrent loss was 45.5 metric tons. The soil is a Fayette Silt Loam (si!)
with 9 = 0.3 and 10 percent organic carbon. The bulk density is 1.2 g/cm3.
Evaluate the lindane loss in sediment and water.
Solut ion:
First the amount of lindane in the surface layer on 8 June, three days
after application, must be estimated. Using Equation 111-28
Pt = 1.0 + 5.0 = 6.0 kg/r
na
251
-------
of lindane on June 5; Looking over the k values for lindane in silty loam
soils, values of .0011 to .0164 per day are found. To simulate worst case,
the smallest disappearance rate will be used.
From Equation 111-26
Pt = 6.0 exp[-0.0011(3)]
=5.98 Kg/ha
The dissolved and adsorbed fractions of lindane must be known. The K
ow
according to Rao and Davidson (Table 111-26) is 643. Using Equation 111-29
log Koc = 1.03 [log (643.)] - 0.18
= 2.71
KQC = 516.
Using 10 percent soil organic carbon and Equation 111-30,
Y - 516(10) _ ,. , cm3
Kd TSO~~ " 51'6 ~~g~
The dissolved and adsorbed portions are
+ (51.6 • 1.2)/.3)] 5.98
6.03 kg/ha
+ .3/(51.6 • 1.2))] 5.98
= 5.95 Kg/ha
The lindane lost due to erosion is (Equation 111-34)
PXt = [45. 5/(72. 8-100-1. 2)]5. 95
= 0.03 kg/ha
The lindane lost due to runoff is (Equation 111-35)
PQt = [0.2/4.5] 0.03
= .001 kg/ha
252
-------
The total loss of lindane is the sum of PQt and PXt times the total area or
Ptotal = (0-03 + 0-001)(72.8)
= 2.26 kg
It is evident that lindane travels primarily with sediment and that erosion
control practices would be effective in reducing lindane losses. This will
not be the case for all pesticides but will be true for most chlorinated
hydrocarbons.
END OF EXAMPLE III-4
3.3 URBAN NONPOINT SOURCE LOADS
From established urban areas, stormwater may pick up various wastes
ranging from settled dust and ash to debris coming directly from man
himself. The quantities of solids from urban nonpoint sources are quite
significant in quantity. Fly ash and dust from industrial processes such as
steel mills, cement manufacturing, and certain chemical processes are known
to be profuse. Dusts from the burning of organic fuels are a significant
factor, and solids in sizable quantities also result from off-street mud,
automotive exhaust, organic debris from tree leaves and grass trimmings, and
discarded litter.
In this report, the nonpoint source loading calculation for
conventional pollutants follows the procedures contained in a recent EPA
study (Heaney, e_t aJL, 1976). This procedure is used for annual loadings.
For storm event loadings, excerpts from another EPA study (Amy, et aj_.,
1974) are utilized. The original procedure has been modified to make it
more realistic in terms of pollutant accumulation.
253
-------
3.3.1 Annual Urban Loads
The predictive equation in this procedure allows the user to make a
determination of average annual loads of BOD5, suspended solids, volatile
solids, P0i» and Total N as a function of land use, type of sewer system,
precipitation, population density and street sweeping frequency. The
procedure requires little external data. The loading equation is
*j • ••';
In this equation
P = annual precipitation (in) -fern)
e- = fraction of urban area that is made up of the
following land uses:
1) Residential
2) Commercial
3) Industrial
4) Other developed (parks, cemeteries, schools, etc.)
ex.. = pollutant loading factor (Ib/acre-in,-teg^hjrew) for
land use 'i'.
f2.(PD.) = population density functions for land use 'i1 and
Y; = the street sweeping frequency
a = a units conversion
1.0 English units
0.442 metric units and
h '
fv \ \ n
_ ToU<^
M. = the^annual area weighted load of pollutant 'j1
J 0»>>AMxa£L
(lb/acre-yr, kg/ha-yr)
254
-------
Evaluation of Factors In the LoadijTg_£ujictio£
The value of e. is determined from areal photographs or may come from
talking to local sources such as city planners or engineers. When better
data are lacking, Table 111-27 may provide some general guidance.
Values fora— are found in Table 111-28. Notice that a different set
of a., are used for separate and combined sewer systems.
' J
The population density function is given by the following set of
equations
Residential (i = 1)
f2(PD.) = 0.142 + 0.218 (PDd)°'54 (111-37)
Commercial - Industrial (i = 2,3)
f2(PDd) = 1.0 (111-38)
Other developed (i = 4)
f2(PDd) = 0.142 (111-39)
The population density PD^ must be obtained locally.
The street sweeping factor Y is evaluated by the following equations
Y = N£/20 (0<_N <_ 20 days) (111-40)
Y = 1.0 (Ns > 20 days) (111-41)
Where
N is the street sweeping interval (days)
255
-------
TABLE I II- 27
GENERAL LAND CONSUMPTION RATES
FOR VARIOUS LAND USES
(AMERICAN PUBLIC WORKS ASSOCIATION, 1974)
Land Use
Residential
Commercial
Industrial
Park
Land consumption (acres/capita)
<100,000
Population
0.1049
0.0101
0.0177
0.0146
>100,000
Population
0.0714
0.0084
0.0083
0.0093
>250,000
Population
0.0585
0.0073
0.0077
0.0078
TABLE III-28
POLLUTANT LOADING FACTORS (a. .)
i»J
Land Use, i
f\ Separate
Areas,
Combined
Areas,
1.
2.
3.
4.
1.
2.
3.
4.
Residential
Commercial
Industrial
Other
Residential
Commercial
Industrial
Other
1.
0.
3.
1.
0.
3.
13.
5.
0.
BOD5
799
20
21
113
29
2
00
467
2. SS
16.
22.
29.
2.
67.
91.
120.
11.
3
2
1
70
2
8
0
1
3.
9.
14.
14.
2.
38.
57.
59.
10.
VS
45
0
3
6
9
9
2
8
4.
0.
0.
0.
0.
0.
0.
0.
0.
P04
0336
0757
0705
00994
139
312
291
0411
5. N
0.131
0.296
0.277
0.0605
0.540
1.22
1.14
0.150
256
-------
It is evident from Table 111-20 that higher loads for all pollutants
will be generated if combined sewers are specified. The data of Lager and
Smith (1974) (Tables 111-29, 30) suggest that the ratio of 4.12 for loads
from combined areas to separate areas is good for BOD,-, N, and P (assumed
in the above methodology). Suspended solids loads on the average appear to
be less from combined areas than separate areas. Total coliform loadings on
the other hand appear to be over an order of magnitude greater from combined
than from separately sewered areas. The above trends are also verified by
the data in Tables 111-29, 30, 31 with the exception of total N.
EXAMPLE III-5
Estimation of Annual Urban Pollutant Loads
Consider a city of 10,000 acres of which 20 percent is commercial, 10
percent industrial, 65 percent residential and 5 percent is in other
developed areas. The residential population density is 10 persons/acre.
Most of the city has separate sewers but approximately 30 percent of the
residential area still has combined sewers. The streets are swept every
five days in the commercial and industrial areas and are not swept in the
residential areas. The mean annual precipitation is 42 inches. Determine
the average annual loadings of total nitrogen and phosphate.
So 1 u t i on :
The population density function value for the residential area is
(Equation 111-37)
fa (Pp)d) = 0.142 + 0.218(10)°-54
= .90
The street sweeping parameter is (Equation 111-40)
Y = 5/20 = .25
257
-------
TABLE II1-29
COMPARISON OF QUALITY OF STORM SEWER DISCHARGES FOR VARIOUS CITIES'
(LAGER AND SMITH, 1974)
Type of wastewater,
location, year,
Ref. No.
Typical untreated
municipal
Typical treated
•unlclpal
Primary effluent
Secondary effluent
Storm sewer
discharges
Ann Arbor. Mich.,
1965 (2)
Castro Valley,
Calif., 1971-72 (14)
-------
TABLE 111-30
COMPARISON OF QUALITY OF COMBINED SEWERS FOR VARIOUS CITIES3,
(LAGER AND SMITH, 1974)
Type of wastewater,
location, year,
Ref. No.
Typical untreated
municipal
Typical treated
municipal
Primary effluent
Secondary effluent
Selected combined
Atlanta, Ga.,
1969 (31)
Berkeley, Calif.
1968-69 (34)c
Brooklyn, N.Y.,
1972 (8)
Bucyrus, Ohio
1968-69 (35)
Cincinnati, Ohio,
1970 (36)
Des Moines, Iowa,
1968-69 (6)
Detroit, Mich.,
1965 (2)
Kenosha, Wis. ,
1970 (18)
Milwaukee, Wis. ,
1969 (7)
Northampton, U.K. ,
1960-62 (22)
Racine, Wis.,
1971 (18)
Roanoke, Va.,
1969 (12)
Sacramento, Calif.,
1968-69 (37)
BODj, COD DO,
mg/1 mg/1 mg/1
Avg
200
136
25
100
60
180
120
200
115
153
129
55
150
119
115
165
San Francisco, Calif.,
1969-70 (3) 49
Washington, D.C.,
1969 (5)
71
Range Avg Range Avg
100-300 500 250-750 —
70-200 330 165-500 —
15-45 55 25-80
48-540 - — 8.5
18-300 200 20-600 —
86-428 --
11-560 400 13-920 —
80-380 250 190-410 —
29-158 -
74-685 115
464
26-182 177 118-765 —
80-350
—
_.
70-328 238 59-513 —
1.5-202 155 17-626
10-470 362 80-1,760 --
^ Total Total Total
conforms, nitrogen, phosphorus,
mg/1 MPN/100 ml mg/1 as N mg/1 as P
Avg
200
80
15
—
100
1,051
470
1,100
295
274
458
244
400
439
78
125
68
622
Range Avg Range
100-350 5xl07 Ixl07-lxl09
40-120 2xl06 5xl06-5xl08
10-30 IxlO3 Ixl02-lxl04
IxlO7
40-150
132-8,759
20-2,440 IxlO7 2xl05-5xl07
500-1 ,800
155-1,166
120-804
2xl06
113-848 — 2xl05-3xl07
200-800
—
7xl07
56-502 5xl06 7xl05-9xl07
4-426 3xl06 2xl04-2xl07
35-2,000 3xl06 4xl05-6xl06
Avg Avg
40 10
35 7.5
30 5.0
1.2b
-
1.2b
13 3.5
-
12.7 11.6
16.3d 4.3
10. 4d 5.9
3-24 0.8b
10C
„
--
__
„
3.5 1.0
a. Data presented here are for general comparisons only. Since different sampling methods, number of samples, and other
procedures were used, the reader should consult the references before using the data for specific planning purposes.
b. Only orthophosphate.
c. Infiltrated sanitary sewer overflow.
d. Only ammonia plus organic nitrogen (total) Kjeldahl).
e. Only ammonia.
f. Only fecal.
259
-------
TABLE III-31
SUMMARY OF STORMWATER POLLUTANT CONCENTRATIONS
(KAISER ENGINEERS, 1969)
Pollutant^
BOD5
COD
S.S.
Total
Total
(as
Total
(as
Coliforms^
Nitrogen
N)
Phosphorus
P)
Stormwater Overflow
Separate
Mean
27
205
608
3xl05
2.3
0.5
Drainage Areas ^
Standard
Deviation
25
118
616
-
1.4
0.4
Concentrations
Combined
Mean
108
284
372
6x1 06
9
2.8
Areas(b)
Standard
Deviation
36
no
275
-
6
2.9
(a) Summary of 20 cities, storm sewers and unsewered areas
(b) Summary of 25 cities, combined sewer areas
(c) All units mg/1 except coliforms, MPN/100 ml
(d) Geometric mean
260
-------
Using Equation 111-36,
Residential
%otal = 42 [o.65 (0.131 - 0.70 + 0.54- 0.30) (.9) (1.0)
Commercial
+ 0.20 (0.296)(1.0)(0.25)
Industrial
+ 0.10 (0.277)(1.0)(0.25)
Other
+ 0.05 (0.06059(0.142)(1.0)1
= 7.2 Ib /acre-yr
Residential
~ii = 42 [o.65 (0.0336 • 0.70 + 0.139- 0.30) (.9) (1.0)
Commercial
+ 0.20 (.0757)(1.0)(0.25)
Industrial
+ 0.10 (0.0705)(1.0)(0.25)
Other
+ 0.05 (Q.0099)(0.142)(1.0)l
1.8 Ib-PO^/acre-yr.
Assuming the urban nonpoint source flow is untreated, the annual N and
PO^ loads to the stream are 36 tons - N/year and 9 tons-POt/year.
END OF EXAMPLE III-5
261
-------
3.3.2 Estimation of Single Event Pollutant Loads
Two factors are of primary importance in determining pollutant loads
from urban stormwater events. First, the accumulation (supply) of materials
on the watershed must be accurately known and second, the capability of the
system to move pollutants from the street surface to the stream must be
estimated. The system here is a combination of rainfall event
characteristics and watershed characteristics.
The accumulation of material on the watershed surface is addressed
first.
3.3.2.1 Accumulation of Pollutants on Street Surfaces
The amount of material available to be washed off can be described as a
function of the time since the latest of these events occurred. This time
is called the equivalent days of accumulation (EDA) and is computed by
EDA = (DR - Ds) (1 -es) + D$ (111-42)
where
D = days since last "significant" storm event
R
D = days since last street sweeping event, and
O
e = the street sweeping efficiency
According to Amy, et al_., 1974, a "significant" storm event is one in
which 0.5 inches falls within a period of one to five hours. A storm of
this size is considered to remove 90 percent of the surface particulates.
Typical values for the street sweeping efficiency term, e_, are given
in Table 111-32. Notice that the efficiency is reduced for smaller particle
sizes. Table 111-33 shows the percent of solids, BOD5, heavy metals and
pesticides associated with different size fractions of debris, dirt and dust
262
-------
ro
cr>
CO
TABLE 111-32
SUMMARY OF STREET CLEANING METHODS
Type
Hand Cleaning
3 Wheel Mechanical
4 Wheel I'cchanlcil
Vacuum
Air
Flusher
Sweeping Action
Pushbroan
Gutter and
Main Brooms
Cutter and
Main Brooms
Gutter Broom,
Vacuum Pickup Head
Gutter Broom, Air
Pump System
Water Under
Pressure
Effectiveness
Debris
Percent
95-100
95-100
95-100
95-100
95-100
Small
Dirt
Percent
Est. 70
50-65
50-65
60-80
Est. 50-
70
30
Dust
Percent
Est. 45
15-20
15-20
40-70
Est. 30-
60
Overall
Travel
Speed,
mph
Trucks
Needed
15-20
55
55
55
55
Sweeping
Speed,
n.ph
Slow
4-8
4-8
4-10
4-8
12
Total
Cost
High
Medium
Medium
Low
Low
Low
Special
Li nil -
tatlons
Heavy
Traffic
Wet,
Snowy
Streets
Wet,
Snowy
Streets
Less Man-
euvera-
bility
Wet
Streets
Very flat
grades
Potential Problems
Parl'ed
r * i-c
Cars
Minor
Major
Major
Major
Major
No
Pavement
Cond 1 1 1 on
No
Major
Major
Minor
Minor
•
No
Unimproved
c t r Af> f t
j irec u
lio
Not Useablc
Not Usrjble
Not Effec-
tive
Not Effec-
tive
Can't use
If no
drainage
Special
n(J V Ait-*
tagcs
Adapt. to
Special
Heeds
Maneu-
vers
well
Wand cm
Clean
Catch
Basins
Source: Tetra Tech, Inc., 1978
Final Report. Surface Sanitation Program for
Newport Bay, California
-------
ro
TABLE 111-33
REMOVAL RATES FOR SELECTED CONTAMINANTS BY SIZE
Particle Size
(u)
^2.000
. 840-2,000
240-840
104-240
43-104
<43
Total Removal
Sweeper
Efficiency
(x)
79
64
60
43
20
15
' Total
Size Oistr
(X)
24.4
7.6
24.6
27.6
9.7
3.9
Solids
. Removal
U)
19.3
4.9
14.9
13.3
1.9
0.9
55.2
BOD5
Size Oistr.
(X)
7.4
20.1
15.7
15.2
17.3
24.3
Heavy Metals
Removal
(X)
5.8
13.3
9.4
7.3
3.8
3.4
43.0
Size Oistri.
(X)
14.8
17.8
14.9
33.8
27.8
- .
Removal
(X)
13.9
11.4
9.9
14.5
5.6
- -
55.3
Pesticides
Size Distri. Removal
(X)
0
16
25.8
28.8
31.7
- .
(X)
0
10.2
15.5
12.4
6.3
- -
44.4
Source: Sartor, J.D. and G.B. Boyd, 1972.
-------
and their corresponding removal rates assuming the typical efficiencies in
column 2 of the table.
Multiplication of EDA by the daily loading rate (Ib day ) gives the
total load of material on the street surfaces. (Residual loads remaining
through pervious storm or sweeping events are not accounted for in this
approach.) Notice that the removal of material will only be from street
surfaces by this method. Material loads from previous areas in the
watershed or nonconnected impervious areas will not be a calculated part of
the load. Daily loading rates are calculated from the following procedure.
3.3.2.2 Street Surface Pollutant Loadings
Data developed in Amy, et a]_. (1974) include nationwide means of
solids loading rates and pollutant composition of street solids, as well as
a more detailed breakdown of data into major source categories. Table
111-34 shows data from the URS report which are divided into 13 subsets
among three major source categories including climate, land use, and average
daily traffic. These data may be different from the means which are given
in the last column of the table, at the 80 percent confidence level.
Whenever the mean of any parameter (solid loading rates or composition) in
any subset differs significantly from the mean of the set of all data, that
number may be substituted for the mean of the set of all data. Table 111-34
also gives the percent standard error of the mean which indicates the degree
of confidence that may be placed on the mean.
3.3.2.2.1 Loading Functions for Solids
Y(S)U = L(S)u-Lst (111-43)
where
Y(S) = daily total solids loading, (kg/day, Ib/day)
L(S) = daily solids loading rates, (kg/curb-km-day, Ib/curb-
mile-day)
265
-------
TABLE II1-34
ro
CTl
cr>
SOLID LOADING RATES AND COMPOSITION — NATIONWIDE MEANS AND
SUBSTITUTIONS OF THE NATIONWIDE MEANS AT 80% CONFIDENCE LEVEL* (AMY, _EJ AL^ , 1974)
Climate
Land Use
Average Daily
Traffic
No. /day
Category
Northeast
Southeast
Southwest
Northwest
Openspace
Residential
Commercial
Light Industry
Heavy Industry
< 500
500-5,000
5,000-15,000
< 15,000
All data**
Ibs/curb Concentrations in micrograms per gram of dry solid No /gram
Loading BOD, COD OP04 N03 OrgN Cd Cr Cu Fe Pb Mn III Sr Zn TCOLI+ FCOL 1+
291c 5,970c 2.6b 139b 17,700b 870c 363fl 21c 27b 260b 4.4r5c
103fa 29,100b 2,240a 1 ,970a 137b 1 ,370b 21b 28b 7.0i4rf
5Qc 470b 241a 78a 2,520b 57, 15, 5.7Z6,
30 246 34,500. 2,600, 10, 480, 6.8z5f 1.114.
C a D D C a T T
14,000b 82,000b 850b 550c 1 ,800a 93fl 1 ,430b 28b
74c 58,700c 269,000C 2,250C 1 ,580c 6.430a 133b 3,440b 48b 520fa
278b 28,600b 1,160C 570b 8.2Z5g
l,210d 252b 6.9Z4f
9,500c 83,000c 741 d 419fa 18,900a 1 ,060C 17rf 34C 3.4Z5d
18a
82d 357a 3.8Z5a
156 19,900. 140,000, 1,280. 804, 2,950, 3.4. 211 104, 22,000a 1,810 418 35 21 370 2.5S6 1.7E5,
u u u uu D U d a a d d d 3 d CD
- ..
Freedom >_ 10). Total number of permitted substitutions = 103. Percent Standard Error of the Mean Subscripting Code- a=0-9 b=10-T9 c=20-29 d=30-39
e=40-49, f=50-62.
+Co1iform counts are expressed in computer notation, i.e. /,5=10^.
** Average TP04 is 2,930C and NH4 is 2,640C
-------
L = street curb-length (approximately 2.0 x street length),
O v
(curb/km, curb-miles).
3.3.2.2.2 Loading_Functions for Other Pollutants
Y(i)u = a-Y(S)u-C(i)u (111-44)
where
Y(i) = daily total loading of pollutant i, (kg/day, Ib/day)
MPN per day for total col i form and fecal col i form
a = conversion factor, 10 (metric and English).
Y(S) = daily total loading of solids, (kg/day, Ib/day),
u
calculated in Equation 11
C(i) = concentration of pollutant i in solids, (ug/g); MPN/g for
u
total coliform and fecal coliform.
Equations 111-43 and 111-44, along with solid loading values and
compositions in Tables 111-21 and 111-22, provide the means to assess daily
average pollutant loadings to urban street surfaces.
It is important to note that pollutant loadings so calculated are
street surface loadings rather than loadings at outfalls to the receiving
waters. The transport of storm runoff in sewers and removal of pollutants
in some treatment systems would reduce pollutant loads to some extent. Such
effects are not included in loading factors suggested in Table 111-34. The
use of catchment basins to remove solids and organic matter, will reduce
pollutant loads from streets to receiving waters.
267
-------
3.3.2.2.3 Procedure for Loading Calculations
Data in Table 111-34 represent two options as well as two levels of
accuracy for a user to assess pollutant loadings from a given urban area.
Application of the "subset" data may result in higher accuracy, but require
more data and more computation effort, than if "nationwide means" are used.
Option I - In this option the user will use nationwide means presented
in Table 111-21. Proceed as follows:
1. Determine solids loading rate and solids composition from
tables.
2. Determine street length (include that of primary and secondary
streets but not driveways, alleys, or parking lots).
3. Calculate daily solids loading using (Equation 111-43).
4. Calculate daily loading of other pollutants using
(Equation 111-44).
Option II - In this option the user will make use of data presented for
source categories in Table 111-34. Steps needed for loading calculations
are:
1. Characterize the study urban area. When applicable, the
entire area should be divided into individual homogeneous
sections with unique characteristics. Each individual section
is then defined as a subarea (e.g., residential area).
1. Determine street length in each subarea.
3. Enter the Table 111-34 at the line labeled "All Data."
4. Select a category of climate, land use, or average daily
traffic, which best applies to an area and move upward to the
line of data to the right of the category heading.
268
-------
5. Substitute those values available in the row selected for the
corresponding values in the row labeled "All Data." In
choosing the substitute loading factors, the following
priority sequence of source categories is suggested: (a)
climate; (b) land use; (c) average daily traffic. The
climatic zones of the U.S. delineated by the URS are shown in
Figure 111-15. Caution: it is not permissible to use more
than one row of substitutions at a time, i.e., to use a BOD
value for land use and COD for climate in order to form a new
row of loading rate and composition data. It is both proper
and useful, however, to repeat the above process to obtain
several new rows of data to present a range of composition and
loading rates.
6. Repeat Steps 4 and 5 for all subareas.
7. Use equation 111-43 to calculate total solid loading in a
subarea.
8. Use solid loading (Step 7), Equation 111-44 and selected
composition data to calculate total loading of other
pollutants in a subarea.
9. Sum up loadings of subareas to obtain the loading of entire
study area.
Option III - In this option, the user will make use of site specific
data.
The recent URS study has assembled all presently available data on the
rates of accumulation of solids and on the concentrations of various
pollutant constituents in those solids that collect on street surfaces.
These data are probably adequate for most urban planning operations. The
user, however, may alternatively replace these loading factors by site
specific data to obtain better prediction.
269
-------
Gr&d
(INSUFFICIENT DATA) i l«Wwauk"
-------
If site specific data are lacking, users are encouraged to conduct
sampling and analytical programs of their own. The data from site specific
tests, if handled properly, may be used in analyzing the area's runoff
problems instead of using values given in this report. This would be
desirable in most instances, especially in areas or under specific
conditions that were not documented in the URS study.
Recommended procedures for conducting site specific tests are given in
Appendix B of the URS Report (Amy, et al_., 1974).
With the lack of site specific data, the user may wish to examine the
available published data for source and reliability. The user is referred
to Appendix A of the URS Report for description of available data sources,
as well as procedures for processing these data.
3.3.2.2.4 Street Length and Land Use Data for Urban Areas
Street length data are available from local public works departments or
street departments. They can also be obtained by measurement of aerial
photographs.
Survey statistics for the U.S. indicate that street surfaces occupy on
the average about one-sixth of the urban area (Manuel, et aj_., 1968). The
American Public Works Association (1974) recently developed a regression
relationship between curb length of urban area versus population density.
Data from many cities across the country were used. The resulting
regression equation is:
CL = 413.11 - (352.66)(0.839)PD (111-45)
where
CL = curb length density, ft/acre
PD = population density, number/acre
271
-------
The correlation coefficient for the equation is 0.72. The regression
! is shown in Figure 111-16.
curve is shown in Figure 111-16
3.3.2.3 Washoff of Pollutants to Receiving Waters
Once the equivalent load of pollutant on the watershed is known the
next step is to calculate the runoff and solids loading from the storm.
The storm event runoff is found by
R = CR • P - DS (111-46)
where
P = storm precipitation (inches, cm)
R = total storm runoff (inches, cm)
CR = runoff coefficient and
OS = depression storage
The runoff coefficient will be dealt with first.
3.3.2.3.1 Runoff Coefficient
In order to estimate the runoff coefficient the percent of impervious
watershed area must be known. If this information is not readily available
the following equation can be used for estimation:
I = 9.6 PD {0-573 - O^L°910 PDd) (111-47)
where
I = percent impervious area and
PD = population density (r persons/acre)
272
-------
GROSS POPULATION DENSITY, POP/HECTARE
600
550
500
450
UJ
^ 400
£>
UJ
•^ 350
t—
§ 300
UJ
£ 250
O
Z
^J 200
03
ID
U 150
) 20 40 60 80 ICC 120 140 160 180 200 220 240
11111 i i • • i i i
-
.
,
'*
•
• A , • ~
S*^^ • •
s -^
r *
1
-•/ .
J
• /••• •
-/
/
"/ •
>/*
I
1
100f-~
50
0
—
i i i i ii i i i
797
750
700
650
600
550 £
i—
500 m
3^
450 g
| ,---
r m
LU
400 §
^«
i —
350 ^
Z
LU
300 Q
x
250 o
Z
LU
200 -1
co
&£.
150 3
100
50
n
0 10 20 30 40 50 60 70 80 90 100"
GROSS POPULATION DENSITY, POP/ACRE
FIGURE 111-16 CORRELATION BETWEEN POPULATION DENSITY AND CURB
LENGTH DENSITY, (AMERICAN PUBLIC WORKS ASSOCIA-
TION, 1975)
273
-------
The STORM algorithm for computing the runoff coefficient is:
CR = 0.15
100
\
where
(m)
CR = runoff coefficient
I = percent impervious area and
K = impervious area runoff coefficient based on slope information
(See Table 35).
TABLE 111-35
VALUES OF RUNOFF COEFFICIENT, k
Impervious Surfaces Approximate k
Flat (<2% slope) 0.80
Moderate (2 to 7% slope) 0.85
Steep (>7% slope) 0.90
3.3.2.3.2 Depression Storage
Depression storage on the watershed is computed as
DS = 0.25 - 0.1875 i~] English units or (111-49)
DS = 0.63 - 0.48 UQQ-! metric units (111-50)
where
DS = depression storage (inches, cm)
274
-------
3.3.2.3.3 Solids Removal
Once the total storm runoff has been computed the percentage of solids
removed from the street surfaces can be estimated by referring to the graph
in Figure 111-17. The solids load for the runoff event is
Y(S)w = EDA • Y(S)u • (PC) (Ill-Si)
when
Y(S) is the load of solids removed to the receiving water body
(Ibs, kg)
EDA is the equivalent number of days of accumulation
Y(S) is the street surface loading rate (Ib/day, kg/day) and
(PC) is the percentage (expressed as a decimal fraction) of
solids removed during the storm event
The user should realize that the runoff coefficient and hence the
stream loading rate does not incorporate the effects of losses in sewers or
stormwater detention basins. For more accuracy in these complex situations
the user is directed toward a more sophisticated urban stormwater model
which would require a hydraulic description of these structures.
3.3.2.3.4 Other Pollutants
As in the street surface loading equation the quantity of other
pollutants is determined by a product of a concentration factor and the
loading of solids. Factors for several conventional pollutants and some
metals can be found in Table 111-34.
Smolenyak (1979) developed loading equations for other conventional
pollutant forms given the suspended solids load. His coefficients for
linear and power function models are given in Table 111-36. These
275
-------
GO
o
e
-s
o
ft)
IfD
CT>
CD
c
JO
m
3> CO
CO H
33
> m
m
o
—I
^-i
o
oo
O
m
o
~n c >
o
C? H
O 2
Tl •-•
-n z
>
O
>
ID i— >
n> 3
3 O
3 to
n>
o>
-s
CO
o
-h
70
C
o
O O
ro
en
o o o
ro . «—
T3 en no
co
01
(-• O
o ->"
i_i o
'ivi en"
Percent of Contaminat Removal
•vj
O
8
\
\
-------
TABLE 111-36
RELATIONSHIPS BETWEEN TOTAL SUSPENDED SOLIDS (TSS) AND OTHER POLLUTANTS
Dependent
Variable
BOD
COD
NHj
I « W ""I
3
N02 + NO^
Organic N
Total N
Dissolved PO.-P
Total P04
Total P-P
No. of
Events
76
109
20
20
100
38
14
31
113
28
Linear
R2
.18
.52
.41
.43
.14
.67
.83
.22
.06
.25
(Load = a+b
Req.
a
24
3.4
.065
.007
.011
.072
.0083
.0081
.003
.084
• TSS
Coef.
b
1.7
.19
.0027
.0012
.00077
.0029
.0027
.00014
.00017
.028
Loq-Loq
R2
.42
.73
.18
.54
.57
.83
.75
.65
.65
.68
(Load = aTSSb)
Req. Coef.
a
.39
.92
.022
.0046
.0071
.011
.0075
.0010
.0012
.0041
b
.99
.72
.48
.60
.55
.80
.78
.67
.68
1.2
-------
regressions were performed on combined data from 23 different urban
catchments in various U.S. cities.
3.3.2.4 Toxic Pollutant Accumulation
Toxic pollutants in urban watersheds are introduced by a variety of
activities. Atmospheric deposition from industrial emissions and vehicular
emissions are primary sources. Other sources include spills onto highways
and loading docks. Such spills may represent significant loads of toxicants
because they generally occur onto impervious surfaces.
The discussion of deposition of toxic pollutants to urban areas will be
divided into two categories
1. metals and
2. organic pollutants
3.3.2.4.1 Deposition of Metals
The loading of metals to street surfaces is estimated in the same
manner as for conventional pollutants. The product of equivalent days of
accumulation, solids loading rates and concentration factor for the metals
in solids (Table 111-34) is used.
EXAMPLE III-6
Estimation of Lead Washoff
Load for a Single Storm Event
Estimate the amount of lead removed from the street surface of Laurel,
MD, during a single storm event. The rainfall depth of the storm is 1.21
inches. The population density for Laurel is 13.2 persons acre~ and there
are 70 curb-miles in the town. When the storm occurred, it had been eight
days since the last significant rainfall and four days since the streets had
278
-------
been swept. (Sweeping efficiency = 0.750) Use a solids supply rate of
103-lb/day curb-mile.
Solution
First determine the percent impervious area using Equation 111-47
I = 9.6 (PDd) (0'573 ' °'0392 logi° Pdd)
PDd = 13.2 and
I = 37.6 percent
Next, compute the runoff coefficient for the area using Equation 111-48
CR = 0.15 (1 - -L) + 0.85 (^J
I = 37.6 and
CR = 0.41
The depression storage is computed as (Equation 111-49)
DS = 0.25 - 0.1875 (1/100)
I = 37.6
DS = 0.18
The storm runoff can now be computed by Equation 111-46 as
Q = CR - P - DS
= 0.41 (1.21) - 0.18
- 0.32 inches
Now, the supply of solids which is present on the watershed must be
determined. Compute the equivalent days of accumulation (EDA) as (Equation
111-42)
EDA = (Dr - Ds) (1 - es) + Ds
= (8 - 4) (1 - 0.75) + 4
= 5 days
279
-------
Using a solids loading rate of 103 Ib/curb-mile - day, Equation 111-43 gives
Y(S) = 103 • (70) = 7210 Ib/day
Equation 111-44 gives the amount of lead deposited per day as a function of
solids
Y(i)u = 10"6 (7210)(1370)
= 9.9 Ib - Pb/day
Since the washoff of lead as opposed to solids is of concern, the Y(S)
becomes Y(i) in Equation 111-51 and
w
Y(1)w = 5 (9.9)(.75)
= 37.1 Ib - Pb
washed off in the event.
END OF EXAMPLE 111-6
3.3.2.4.2 Deposition of Organic Pollutants
Many pollutants are discharged into the atmosphere and eventually
settle out directly onto water surfaces, or onto the watershed surface where
they become available for transport. Pollutants occur in the atmosphere as
1) particulates; 2) gases; or 3) dissolved in water vapor. Eisenreich
ert al. (1981) have suggested that compounds having saturation vapor
_Q
pressures (P ) of 10 will primarily be particulate whereas those
s .4
compounds with P >10 will be primarily in the vapor phase. Cautreels and
Van Cauwenberghe (1978) give distribution coefficients between the gas and
particulate phases for 55 aliphatic hydrocarbons, PAH's, phthalic acid
esters, fatty acid esters, aromatic acids and basic compounds.
Both particulates and gases may settle out onto receptor surfaces. For
particles <0.3 ym in diameter, the major process is Brownian diffusion for
diameters 0.5 to 5 ym inertial impaction-interception governs and for
diameters >5 ym, gravitational settling is dc~n'nent. For gravitational
settling, Stokes1 Law may be used to predict the settling velocity. Since
Stokes1 Law is applicable only to quiescent redia, it should give an upper
230
-------
bound for V. (the deposition velocity). It is stated as
Vd aaifg£-(p - Pa} (IH-52)
where
V = settling velocity (cm/sec)
d _4
a = conversion factor (10 )
g = acceleration of gravity, 981.46 (cm/sec2)
y = viscosity of air, 0.000177 (g/cm-sec) at 50^ (10°C)
P = particle density, % 2 (g/cm3)
Pa = density of air, 0.001243 (g/cm3) at 50°F (10°C)
d = particle diameter (microns)
For particles <5 ym in diameter Stokes Law is not applicable and
experimental values for the deposition velocity should be used. Eisenreich
e_t aj_. (1981) suggest values of V , = 0.1 to 0.5 cm/sec for trace organics.
Some experimental values are shown in Table 111-37.
Once the settling velocity is known, the following procedure can be
used to predict the dry deposition loadings:
L = Vd • Cp • A • f (111-53)
where
L is the load of the pollutant delivered to the receptor surface
as dry deposition (appropriate mass units/sec)
V is the particle settling (deposition) velocity (m/sec)
C is the concentration of atmospheric particulates (mass/m3)
A is the projected receptor ares (m2)
f is the fraction (by weight) of the pollutant in the
particulates
Normally, smaller size particles are more chemically and physically
reactive than larger particulates, and therefore pollutants will be
associated with these smaller particles. Obviously the particle size to
which pollutants are adsorbed affects their atmospheric residence time and,
hence, loadings. According to Neff (1979), most PAH are associated with
281
-------
TABLE III-37
FIELD-MEASURED DRY DEPOSITION VELOCITIES
Compound
V .
(cm/s)
Collection
Surface
PCB
(Aroclor
1242, 1254)
PCB
PCB, DDT
(gas phase)
PCB, DDT
PCB
(total)
PCB
(Aroclor
1016)
PCB
0.5
0.3-3
0.19
1.0
0.14
0.04
Mineral-oil-coated
plates
Estimated
Estimated
Glycerol-coated
plates
Glycerin-water,
Al pans
0.43
Source: Eisenreich et al., 1981
282
-------
participates in the 1 to 2 micron range. Van Vaeck and Cauwenberghe (1978)
have shown that aerosol PAH are associated with particles of median diameter
from 0.7 to 1.4 urn. In addition, they give the concentrations of 50 trace
organic compounds associated with different size particles. Higher weight
PAH, alkanes, and carboxylic acids had significant mass fractions associated
with >1 ym diameter particles.
EXAMPLE III-7
Dry Atmospheric Deposition of Pollutants
Adsorbed to Participates
Estimate the maximum daily loading of pyrene to a watershed having an
area of 106m2 overlain by an air mass having a mean daily particulate
concentration of 50 yg/m3. The average pyrene content of the participates
is 1.0 x 10 yg-pyrene/yg. Assume a deposition velocity of 0.1 cm/sec.
Solution:
Compute the daily dry deposited load of pyrene.
L = Vt • Cp • A . f
= 0.001 -OL. • ^^ • 106m2
sec m3
86400 sec
• 1 0 x 10" " -
yg day
= 4.32 x JLOJ_ ug/day
Ans.
•END OF EXAMPLE 111-7
Gas phase pollutants may also be deposited directly to the v.'atershed
surface. In this case the loading equation is
283
-------
L = V
(111-54)
where
is the dry deposited load (mass/sec)
, . is the gas deposition velocity (m/sec) and
L
V
a
A is the receptor area (m2)
C is the ambient concentration of the gas phase pollutant.
EXAMPLE 111-8
Dry Atmospheric Deposition of Gaseous Pollutants
of Gaseous Pollutants
Estimate the annual deposition of toxaphene to a 1 Km area at
Stoneville, MS during 1974. The mean monthly atmospheric concentrations are
shown in Table 111-42. Assume an average deposition velocity of 0.2 cm/sec
for the entire year.
Solution:
1
2
3
4
5
6
7
8
9
10
11
12
.002
.002
.002
.002
.002
.002
.002
.002
.002
.002
.002
.002
12
A • At
Vd
(m/sec)
C
n
(ng/M3)
A t
n
(M2)
L
10.9
9.7
19.1
27.7
44.
38.
,3
.6
175.0
.6
,6
903.
524.
114.8
32.9
12.6
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
31 x
28 x
31 x
30 x
31 x
30 x
31 x
31 x
30 x
31 x
30 x
31 x
86400
86400
86400
86400
86400
86400
86400
86400
86400
86400
86400
86400
5
4
1
1
2
2
9
4
2
6
1
6
.84
.69
.02
.43
.37
.00
.37
.84
.72
.15
.71
.75
x
x
X
X
X
X
X
X
X
X
X
X
108
108
109
109
109
109
109
1010
1010
109
109
108
1.01 x 1011
ng/year
or 101.4 q/year Ans.
284
-------
In actual situations the deposition velocity changes with
meteorological conditions especially wind speed. In general, deposition
velocities diminish as wind speeds decrease. Further reading on gaseous
pollutant deposition can be found in Murphy et. a_l_. (1980).
•END OF EXAMPLE III-8
Precipitation falling through the atmosphere tends to scavenge
particulates and absorb gases so that it contains a variety of substances.
Because of the volume of precipitation which generally occurs, it may
constitute a significant source of pollutant loading to the watershed
surface. Load calculation for wet deposition is shown below:
where
L = 10 • C • P • A
(111-55)
L is the load of the pollutant delivered to the receptor as wet
deposition (appropriate mass units/sec)
C is the concentration of the pollutant in precipitation
(mass/liter)
P is the precipitation rate (cm/sec)
A is the projected receptor area (m2)
If concentrations of constituents in rainwater are unknown, they can be
estimated using the ambient vapor phase concentrations of the pollutant and
the Henry's Law constant for the pollutant. By assuming equilibrium between
the gas and water phases, one can write
C1V1 + CgV9 = CtVt
or
111-56)
285
-------
.hft")
C, = x^ ' (IH-57)
1
where
Ci is the concentration of the pollutant in water yg/m3
Vi is the volume of rainfall m3
C is the concentration in the gas phase (yg/m3)
V is the gas phase volume (m3)
C. is the total pollutant concentration yg/m3
V is the total volume of the atmosphere (m3) and
H is the Henry's Law constant for a particular pollutant.
Because HV »VX for almost all of the priority pollutants and V^ V
the equation reduces to
C
' '\j V
g - - . g t
at
1
H (111-58)
Thus for toluene, for instance, if Ct is 1 yg/m3 air
r _ 1.0
Ll .27
=3.7 yg/m3 water
For Henry's Law constants on the order of 10" the equation for Ci can be
rewritten as
/W DI
r \ 1 q /
cl = w~pT (ni-59)
H . ^.-1 + 1
l pg
where
p is the gas density
P! is the water vapor density
286
-------
Wi is the weight of water vapor in the atmosphere and
W is the weight of dry air.
W , Wi, p , and PI can be determined for any temperature and relative
y «?
humidity from a standard psychrometric chart.
Another similar approach to determining wet deposition is the use of
washout rations (Eisenreich et al., 1981). The wet flux is calculated by
F = W. • I . C (111-60)
i a
where
F is the wet pollutant flux
W. is the washout ratio for pollutant i
I is the precipitation rate (LT~ ) and
_3
C-, is the ambient pollutant concentration (ML )
Calculated and observed valued of W for some contaminants are shown in
i
Table II1-38. Note that field measured washout ratios contain the effect of
particulate scavenging by rainfall whereas calculated values only consider
constituents absorbed as gases from the atmosphere.
In order to use the methods for determining dry and wet deposition and
gaseous absorption, the atmospheric concentration of the pollutant must be
known. For the 129 priority substances, very little monitoring data is
available. Some scattered data are available such as that in Tables 111-39
through 42. For a particular locale, some monitoring will need to be done
at this stage.
3.3.2.5 Washoff of Toxic Pollutants
The amount of organics or metals lost in solids or dissolved phase is
estimated using the adsorption partition coefficient. The fractions are
given by
287
-------
TABLE 111-38
WASHOUT RATIOS FOR SELECTED TRACE ORGANICS
Compound RT/H = W. Wfield
Dieldrin 1.1 x 105 2-9 x 103
Aldrin 1.6 x 103
DDT 5.9 x 102 2-8 x 10"
HCH 5 x 10" 1-5 x 10"
Aroclor 4 x 101 1 x 10"
1242
Aroclor 6.3 9 x 10"
1248
Aroclor 8.3 2-9 x 10"
1254
8-35 x 10"
Aroclor 2.9
1260
Hexachlorobenzene '3.7 x 102 1.5 x 103
Chloroterpenes 3.5 x 10" 7 x 103-
3 x 10s
Di-2(ethylhexyl) — 1-9 x 10"
phthalate
Source: Eisenreich eit aj_., 1981
Note: H (Henry's law constant) in this table has units of
1iter-atm/mole. Essentially W. = - as defined previously.
288
-------
TABLE II1-39
PCB's, DOT'S, AND PHTHALATE ESTERS IN THE GULF OF MEXICO ATMOSPHERE*
Sample No. Date
1 3/26-27
2 3/28-29
3 3/29-30
4 3/30-31
5 3/31-4/1
6 4/1-2
7 4/2-3
8 4/3-4
9 4/4-5
10 4/6-7
Average
PCB
0.66
0.37
0.17
0.18
0.13b
0.17
0.23
0.71
0.79
0.35
0.35
Concentration (ng nT
p,p'-ODT p,p'-DDE
0.042
0.078
0.047 (90)
0.020
0.016b
0.022
0.041
0.021
0.044
0.010
0.034
0.116
0.065
0.026
0.018
0.010b
0.009
0.018
0.017
0.180
0.031
0.049
)
DEHP
1.92 (51)
0.72 (67)
1.45 (51)
1.75 (78)
0.53b(-)
1.72 (71)
1.34 (50)
1.80 (25)
1.34 (69)
0.83 (58)
1.16 (57)
DBP
3.71 (68)
0.84 (94)
3.34 (20)a
0.65 (88)
0.75b(-)
1.03 (J5)
0.16C(-)
1.30 (46)
0.80 (66)
0.38C(-)
1.30 (63)
Questionable value, not included in average.
Vapor phase only
cParticulate only.
Concentrations (in ng m" ) are for the total sample. The numbers in parentheses are the percent
of the compound measured in the vapor phase only. Unless otherwise indicated all PCB's, DDT,
and DDE vapor concentrations are>98S.
Source: Giam et al., 1980.
289
-------
TABLE 111-40
1975 MONTHLY AVERAGE CONCENTRATIONS OF THREE ORGANIC COMPOUNDS AT THREE NEW YORK CITY LOCATIONS
Month
January
February
March
April
May
June
July
August
September
October
November
December
Average
Urban
Suburban
(Sterl ing
Forest, N
Mi
23
85.43
NDA
47.39
53.42
NDA
NDA
NDA
NDA
57.19
98.78
69.53
NDA
•Y.)
TSPM*
crograms/m3
Sector
27 34
59.52
61.21
52.71
70.31
89.38
67.28
74.41
63.51
54.74
73.65
65.69
57.96
65.86
63.71
NDA
56.63
65.72
68.58
67.25
56.97
55.34
51.67
73.64
79.74
NDA
23
4.90
NDA
2.93
3.00
NDA
NDA
NDA
NDA
4.40
2.98
4.15
NDA
DBP*
Sector
27
4.93
8.49
5.03
4.84
3.17
1.99
3.48
5.00
4.79
10.28
5.31
10.99
4.41
1.41
Nanograms/m3
34 23
3.87
NDA
4.33
2.72
2.22
1.80
0.14
1.80
7.38
2.81
5.69
NDA
10.89
NDA
10.91
4.93
NDA
NDA
NDA
NDA
8.81
12.15
13.52
NDA
DEHP*
Sector
27
11.96
15.41
12.30
12.48
14.72
10.06
14.05
15.48
17.93
28.60
23.24
25.30
14.45
2.51
34
18.58
NDA
13.06
13.50
12.23
11.02
8.96
13.58
12.15
22.15
16.76
NDA
BEO*
Mi crograms/m3
Sector
23 27 34
3.87
NDA
1.76
1.88
NDA
NDA
NDA
NDA
1.43
3.98
3.79
NDA
3.08
4.13
2.56
2.96
3.89
2.39
3.84
3.45
3.04
5.24
5.01
5.66
3.21
4.15
NDA
2.54
2.54
2.67
2.06
1.63
1.87
3.26
3.53
3.56
NDA
TSPM, Total Suspended Particulate Matter; DBP and DEHP, di-Butyl and di-(2-Ethylhexyl) phthalate; and
BEO, Benzene Extractable Orgam'cs.
Source: Bave et al, 1978.
-------
TABLE 111-41
SEASONAL FLUCTUATIONS IN THE GEOMETRIC MEAN PAH CONCENTRATIONS
IN AIR SAMPLES COLLECTED AT 13 STATIONS
IN THE LOS ANGELES, CALIFORNIA, AREA
Compound
Pyrene
Fl uoranthene
Benz[a]anthracene
Chrysene
Benzo[e]pyrene
Benzo[a]pyrene
Benzo[b]fl uoranthene
Benzo [i ]f 1 uoranthene
Benzo [k]fl uoranthene
Perylene
Anthanthrene
Benzo [ghi ]peryl ene
Indeno[l ,2,3-cd]pyrene
Coronene
Total PAH measured
1
0-58
0-38
0-30
0-70
1-30
0-77
0-26
0-27
0-22
0-33
3-80
1-79
2-49
13-19
Quarter concentration (ng/m3
2 3
0-23
0-15
0-06
0-26
0-42
0-17
0-24
0-06
0-07
0-08
1-35
0-68
1-13
4-90
0-25
0-24
0-10
0-44
0-62
0-26
0-33
0-12
0-15
0-06
0-14
2-71
1-00
1-66
8-08
)
4
1-24
0-68
0-59
1-57
1-96
1-27
1-30
0-43
0-52
0-22
0-79
8-25
2-64
4.44
25-90
Source: Neff, 1979
291
-------
TABLE 111-42
AVERAGE MONTHLY ATMOSPHERIC LEVELS OF
FOUR PESTICIDES AT STONEVILLE, MISSISSIPPI
Endrin (nqm"
January
February
March
April
May
June
July
August
September
October
November
December
Average
January
February
March
April
May
June
July
August
September
October
November
December
Average
1972
1.1
1.1
2.1
3.1
1.0
0.9
5.2
10.1
8.8
4.0
0.5
0.0
3.2
Methvl
0.0
0.0
0.0
0.0
0.0
1.6
61.4
216.9
111.7
1.4
0.0
0.0
32.8
1973
0.1
0.1
0.7
0.7
1.2
3.8
0.7
5.0
8.4
5.0
1.1
0.2
2.3
Parathion
0.0
0.0
0.0
0.0
0.0
22.8
4.5
129.3
791.1
17.1
0.0
0.1
80.4
3
' )
1974
0.2
9.2
0.6
0.5
0.7
0.7
9.3
27.2
18.8
4.3
1.0
0.5
5.3
3
(nonf )
1.0
0.3
0.3
0.6
0.6
0.9
40.9
341.1
167.9
2.0
0.0
0.0
46.3
1972
0.0
13.0
68.0
67.4
32.4
44.2
400.7
1540.0
827.9
97.9
9.3
0.0
258.4
10.8
12.6
32.6
34.1
17.2
16.2
117.3
515.3
378.8
37.6
14.8
6.3
99.5
3
Toxaohene (nqm" )
1973
0.0
0.0
16.8
10.8
46.8
109.9
41.1
268.8
322.6
161.1
0.0
9.9
82.3
3
Total DDT (nqm" )
3.9
4.8
11.1
11.4
18.6
49.5
9.6
25.6
24.6
18.9
11.9
2.4
16.0
1974
10.9
9.7
19.1
27.7
44.3
38.6
175.0
903.6
524.6
114.8
32.9
12.6
159.5
3.0
3.6
7.6
7.7
15.6
12.8
24.3
37.9
19.4
5.1
3.3
2.1
11.9
Source: Arthur et al, 1976
292
-------
-]' (111-62)
s c!
where
F = fraction of dissolved pollutant
p - sediment concentration (g/cm3)
K, = adsorption partition coefficient (cm3/g) and
d
F = the fraction of sorbed pollutant
The average sediment concentration p is determined by dividing the
sediment loss by the total storm runoff
Ps = a • Y(S)w / (Aw • Q) (111-63)
AW = watershed area (ha, ac)
when
p is the suspended sediment concentration (g/cm3, lb/ft3)
Y(S) is the sediment load (kg, Ib)
Q is the total storm runoff depth (in, cm)
a is a conversion constant
= 10 metric
= 2.75 x 10"4 English
Y(S) and Q are determined by urban runoff single event procedures in
section 3.3.2.3.
Tables 111-43 and 111-44 show the relative magnitudes of concentrations
of metals and some organic pollutants in wet and dry weather flows. Wet/dry
ratios show that the concentration of metals in nonpoint source runoff is
from 1 to 10 times greater than dry weather flows in which the contributions
are primarily from point sources. Concentrations of PAH are 100 times
greater from nonpoint than point sources.
293
-------
TABLE 111-43
FLOW WEIGHTED MEAN CONCENTRATIONS OF TRACE METALS AND CHLORINATED
HYDROCARBONS IN THE LOS ANGELES RIVER ( yg/1)
1971-1972
Constituent
Ag
Cd
Cr
Cu
Hg
Ni
Pb
Zn
*
Fe
Mn
DDT
PCB
Dry Weather Flow
1.4
4.3
84.
21.
0.39
24.
150.
230.
3.8
120.
0.27
1.6
Storm Runoff
2.2
13.
83.
130.
1.0
78.
940.
100.
18.
480.
0.93
2.6
Ratio
Wet/ Dry
1.6
3.0
.99
6.2
2.6
3.2
6.3
.43
4.7
4.0
3.4
1.6
1979-1980
Dry Weather Flow
0.4
6.2
30.
52.
0.25
43.
56.
200.
4.8
130.
0.05
0.21
Storm Runoff
0.8
4.0
100.
87.
0.8
61.
160.
960.
51.
650.
0.33
0.31
Ratio
Wet/ Dry
2.0
0.64
3.3
1.7
3.2
1.4
2.9
4.8
10.6
5.
6.6
1.5
Source: Coastal Water Research Project Biennial Report 1979-1980.
-------
TABLE 111-44
30NCENTRATIONS OF PAH IN MUNICIPAL UASTEUATER EFFLUENTS IN THE GFR
(ALL VALUES ARE IN THE yG/LITER)
Compound
Fluoranthene
Pyrene
Benz[a]anthracene
Benzo[b]fluoranthene
Benzo[i]fluoranthene
Benzo[k]fluoranthene
Benzo[a]pyrene
Benzo[ghi]perylene
Indeno[l ,2,3-cd]pyrene
Total identified PAH
Total unidentified PAH
Total PAH
Dry Weather
0.352
0.254
0.025
0.039
0.057
0.022
0.001
0.004
0.017
0.771
0.075
0.846
During Heavy Rain
16.350
16.050
10.360
10.790
9.910
1.840
3.840
4.180
4.980
78.300
9.200
87.500
Ratio
Wet/ Dry
46.4
62.2
414.4
276.7
173.9
83.6
3840.0
1045.0
292.9
103.4
Source: Neff, 1979
295
-------
•EXAMPLE II1-9
Washoff of Organic Urban Pollutants
Estimate the washoff of pyrene from the street surface of Laurel, MD
during the storm in Example III-6. Assume the deposition rate of pyrene
from Example III-7. Estimate the load of pyrene in runoff (dissolved) and
adsorbed to solids. The KQC for pyrene is approximately 1.2 x 105 and the
streets solids are 20 percent organic carbon.
Solution:
From problem III-6
A = 1000 ha
w
EDA = 5
Q = 0.32 in or 0.81 cm
Y(S) = 7210 Ib/day or 3277 kg/day
Y(pyrene)u = 4.3 x 10s ug/day
The washoff of solids is (Equation 111-51)
Y(S)U = 5 (3277)(.75)
= 12289 kg-solids
The mean concentration of solids is (Equation 111-63)
p_ = 10"5(12289)(0.81-1000)
s
= 0.15 kg/m3 or 1.5 x 10"* g/cm
To estimate the K , (partition coefficient) for pyrene Equation 111-30 can be
used
= 1.2xl05(20)
Kd 100
= 24000 cm3/9
Using Equations 111-61 and 111-62 the dissolved and adsorbed pyrene
fractions are
F,, = [1 + (1.5xlO"'4)(24010.)]"1
W
- 0.22 and
296
-------
[' * ! ]'
L 1.5 x 10'"(24000) J
i-l
.5 x 10'" (24000)
= 0.78
The dissolved and adsorbed loads are the product of the above fractions
and the total pyrene load (Equation 111-51)
Y(pyrene)w = 5(4.3xl05)(.75)
= 1.6xl06 yg
Y(pyrene)w = F 1.6xl06
w = 0.22(1.6xl06)
= 3.5x105 yg
Y(pyrene)w = Fg (1.6xl06)
5 = 0.78 (1.6 x 106)
= 1.25xl06 yg
Note that although the partition coefficient for pyrene is very high, a
substantial quantity of pyrene is in the dissolved form (22 percent). This
is due to the very low mean solids concentration (only 150 mg/1). The mean
pyrene concentration is about 0.20 yg/1 which is close to dry weather values
in Table 111-46.
END OF EXAMPLE III-9
3.4 POINT SOURCE WASTE LOADS
The purpose of this section is to discuss sources of information
concerning point source discharges and to provide a reasonable range of
values for discharge concentrations when no direct source data can be
located. When available, direct data concerning an existing or proposed
discharge is preferable to the use of the information presented here.
297
-------
3.4.1 Direct Sources of Data
Before using these guidelines and estimates a planner should exhaust
the sources of actual point source waste loading information available to
him. The discharger may be the best source of information since many states
require dischargers to maintain a self monitoring program. Pollutant load
per day and pollutant concentration data are usually included in this
information. Second, essentially all point source discharges are required
to obtain a discharge permit. The state or federal agency issuing these
permits will have on file maximum allowable limits for the discharge. These
limits can be used as an upper bound waste loading rate. Third, state water
quality or water resource agencies often have conducted sample collection
programs for significant discharges. Fourth, data for a similar facility
within the local region (same activity, same general size) may be used as an
estimate for an unknown waste load. If none of these are available the
following procedure may be used.
3.4.2 Estimation of Municipal Waste Loads
The equation for estimating municipal waste loads is shown below.
Basically this method involves computing an influent load and reducing that
load according to removal efficiency based on treatment type. The equation
is
Y(p)m =
where
Y(P) is the waste load being delivered from the treatment plant
m
to the stream (Kg/day)
Q is the flow per capita per day
P is the population being served by the treatment facility
C is the concentration of a particular pollutant in raw
domestic sewage (yg/1) and
e is the removal efficiency based on treatment type.
298
-------
3.4.2.1 Evaluation of Parameters in the Municipal Waste Loading Equation
The utilization of domestic water depends on a number of factors.
Among them are:
• geographic location
• climate
• degree of industrialization and
• size of community
A reasonable estimation of water comsumption in different parts of the
country is found in Table 111-45. Average water consumption per capita by
state and selected municipalities are shown. The user should note the large
variations in the table. These rates can be used for the parameter Q if
more site specific data is unavailable.
Typical concentrations of conventional wastewater constituents in
untreated domestic wastewater (mg/1) is shown in Table 111-46. If the
relative strength of the raw sewage is unknown, use of the higher values are
recommended.
Removal efficiencies vary greatly depending upon the constituent and
upon treatment process. Typical removal efficiencies are shown in
Table 111-47.
An example of the differences in loadings of untreated and treated
domestic sewage are shown in Table 111-48.
3.4.3 Industrial Waste Loads
Industry is a source of many types of pollutants. Of the conventional
pollutants, they are primarily sources of BOD, suspended solids, COD, and
oil and grease. They are also primary sources of toxic organic pollutants
and metals as will be discussed in the next section.
299
-------
co
o
o
TABLE 111-45
WATER WITHDRAWALS FOR PUBLIC SUPPLIES BY STATES AND BY SELECTED MUNICIPAL SYSTEMS, 1970
State, city
Alabama:
Birmingham
Alaska:
Anchorage
Arizona:
Phoenix
Arkansas:
Little Rock
California:
Los Angeles
San Francisco
Colorado:
Denver
Connecticut:
Hartford
Delaware
Florida:
Miami
Georgia:
Atlanta
Hawaii :
Honolulu
Idaho
Illinois:
Chicago
Indiana:
Indianapolis
Iowa:
Des Moines
Kansas:
Wichita
Kentucky:
Louisvil le
Louisiana:
Shreveport
L/
capita-d
806
576
1790
769
787
864
503
784
685
686
1424
746
955
541
564
700
617
1208
946
564
746
780
897
772
871
534
508
466
534
587
508
314
655
545
519
gal/
capita-d
213
152
473
203
208
228
133
207
181
181
376
197
252
143
149
85
163
319
250
149
197
206
237
204
230
141
134
123
141
155
134
83
173
144
137
State, city
Maine:
Portland
Maryland:
Baltimore
Massachusetts:
Boston
Michigan:
Detroit
Minnesota:
St. Paul
Mississippi :
Jackson
Missouri :
Kansas City
Montana:
Billings
Nebraska:
Omaha
Nevada:
Las Vegas
New Hampshire
New Jersey:
Elizabeth
New Mexico:
Albuquerque
New York:
New York City
Rochester
North Carolina:
Greensboro
North Dakota:
Fargo
Ohio:
Akron
L/
cap1ta-d
553
580
515
648
530
883
636
671
473
515
507
432
485
587
826
754
636
742
1154
1038
435
526
314
772
746
609
1046
663
644
492
477
515
594
492
gal/
capita-d
146
153
136
171
140
233
168
177
125
136
134
114
128
155
219
199
168
196
305
274
128
139
83
204
197
161
276
175
170
130
126
136
157
130
State, city
Oklahoma:
Tulsa
Oregon:
Portland
Pennsylvania:
Pittsburgh
Rhode Island
South Carolina:
Charleston
South Dakota:
Sioux Falls
Tennessee:
Memphis
Texas:
Dallas
Houston
Utah:
Salt Lake City
Vermont
V i rg i n i a :
Richmond
Washington:
Seattle
West Virginia:
Morgantown
Wisconsin :
Mi Iwaukee
Wyoming:
Chenne
District of Columbia
Puerto Rico
United States*
L/
capita-d
492
595
712
1129
685
485
462
916
652
549
587
488
549
587
610
947
1113
523
553
420
644
1200
1091
568
549
587
659
746
841
799
326
628
gal/
capita-d
130
157
188
298
181
128
122
242
172
145
155
129
145
155
161
250
294
138
146
111
170
317
288
150
145
155
174
197
222
211
86
166
Note: L x 0.2642 = gal.
Source: Metcalf and Eddy, 1979.
-------
TABLE III-46
TYPICAL MUNICIPAL WASTE CONCENTRATIONS
Concentration ma/1
Constituent
Solids, total
Dissolved, total
Fixed
Volatile
Suspended, total
Fixed
Volatile
Settlable solids, (ml/liter)
Biochemical oxygen demand, 5-day, 20° (BOD5-20°)
Total organic carbon (TOC)
Chemical oxygen demand (COD)
Nitrogen, (total as N)
Organic
Free ammonia
Nitrites
Nitrates
Phosphorus (total as P)
Organic
Inorganic
Chlorides*
Alkalinity (as CaC03)*
Grease
Strong
1,200
850
525
325
350
75
275
20
400
290
1,000
85
35
50
0
0
15
5
10
100
200
150
Medium**
720
500
300
200
220
55
165
10
220
160
500
40
15
25
0
0
8
3
5
50
100
100
Weak
350
250
145
105
100
20
80
5
110
80
250
20
8
12
0
0
4
1
3
30
50
50
*Values should be increased by amount in carriage water.
**In the absence of other data use medium strength data for planning purposes,
Source: Metcalf and Eddy, 1979
301
-------
TABLE III-47
MUNICIPAL WASTEWATER TREATMENT SYSTEM PERFORMANCE
Influent: "Raw-Medium Strength Domestic Sewage" see Scheme Number 0 for characteristics.
Effluent Concentrations (ma/1), (% Total Removal Efficiencies*)
Scheme Number** BOD,
o 200(rr)
Raw waste water v '
1 130(35%)
2 40
(80%)
3 25(88%)
4 18
5 18
61 ^
( 94»)
7 2/
COD
50V)
375(252)
125
100(80S)
70(86%)
70(86%)
60(88%)
«(97«
SS PT, (mgP/1) NT, (mgn/1 )
?nn IA An
ZOO(OS) 10(OS) 40(0i)
100(25S) 9(102) 32(202)
30 7.5 26
(85%) (25%) (355)
12(94%) ' 7(30S) 24(40%)
7(95^) ^90%) 22(45%)
7(95S) ^90%) 4(90%)
^gg.Si) ^gox) 3(92%)
^99. 5%) ^gOS) 2(95S)
* Efficiencies for wastewater treatment are for the approximate concentration range,
as measured by BOD5, of TOO < BODg S 400, (mg/1).
** Scheme No. Process
0 No treatment.
1 Primary
2 Primary, plus Activated Sludge (Secondary Treatment)
3 Primary, Activated Sludge, plus Polishing Filter (High Efficiency or
Super Secondary)
4 Primary, Activated Sludge, Polishing Filter, plus Phosphorus Removal
and Recarbonation
5 Primary, Activated Sludge, Polishing Filter, Phosphorus Removal, plus
Nitrogen Stripping and Recarbcnation
6 Primary, Activated Sluage, Poiisning Filter, Phosphorus Removal, Nitrogen
Stripping Recarbonation, plus Pressure Filtration
7 Primary, Activated Sludge, Polishing filter, Pnosphorus Removal, Nitrogen
Stripping Recarbonation, Pressure Filtration, plus Activated Carbon
Adsorption
Source: Meta Systems, 1973
302
-------
OJ
o
TABLE 111-48
POINT SOURCE LOADINGS OF SIX MAJOR WASTEWATER TREATMENT FACILITIES IN ONE NORTH CAROLINA 208 AREA
FACILITY NAME
Norths ide
Third Fork
New Hope
.Chapel Hill
Walnut Creek
Ilil Isborough
SUB-TOTAL
FLOW
(MGD)
8.4
3.4
3.4
3.8
20.3
0.56
39.9
BOD C
INFLUENT
16,673
7,231
5,473
5,166
64,673
957
100,173
b/day)
EFFLUENT
1,331
993
1,758
824
11,005
383
16,294
S.S.
INFLUENT
13,731
5,189
4,197
5,071
23,702
934
52,824
Ib/day)
EFFLUENT
2,522
2,410
1,361
1,363
7,111
234
15,001
Sludge Production
(Ib/day Dry Solids)
4,300
3,300
318
3,000
6,255
17,173
-------
Because of the difficulty in characterizing industrial wastes in
general, the user is advised to get as much data as possible locally.
Water use (flow rate) as well as variety in unit processes will have a
profound effect on pollutant loads. If local data is unavailable, the
best sources of information on industrial waste characterization is the
"Effluent Guidelines" series of reports by the U.S. EPA. A report is
available for each of the EPA point source categories. These reports
contain typical waste characteristics for various processes within the
point source category as well as process water usage and are listed in
the references at the end of the chapter. Effluent limitations are
also given which can be used as upper bound concentrations in water
quality assessments.
In addition to process variation, industries also employ various
types of waste reduction measures in the plant and in plant wash
treatment facilities. In-plant measures include
• recirculation of non-contaminated water
• segregation of contaminated and non-contaminated waters
prior to treatment
0 removal of semidry residues and
• flow reduction
Eckenfelder (1970) presents information on industrial waste
concentrations and loads, waste reduction measures and industrial waste
treatment.
Table 111-49 contains some typical pollutant loads which might
result from the industries shown. Table 111-50 also presents loads for
some industries and typical removal efficiencies and expected loads.
These waste water treatment processes are representative though not
exhaustive of techniques which may be used. Values in Tables 111-49
and 111-50 are for comparison only. They should not be used for load
projecting in other areas.
304
-------
OJ
o
in
TABLE III-49
TYPICAL INDUSTRIAL DISCHARGE POLLUTANT CONCENTRATIONS
r
Industry
Primary Metal2
Cu, Brass Rods3
Roofing Materials3
Steel Plate, Wire3
Petroleum (General)2
Oil Production^!)3
Oil Production(#2)3
Oil Production(#3)3
Paper (General)2
Paperboard3
Paper3
Primary Inorganic2
Alky Lead Fluoro
Hydrocarbons3
Inorg. Acids3
Primary Organic2
Caustic Chemicals3
Plant Food3
Flow
Rates ,
mg/1000 Ib1
0.2-1.6
0.04
0.01
0.004
0.005
0.003
0.003
0.001
0.015
0.017
0.024
0.002
0.002
0.004
0.002
0.021
0.001
BOD
lb/1000 Ib
_
0.1
13.6
0.56
1.3
0.57
0.45
0.45
18
19.6
12.6
0.2-3.5
0.39
0.08
U1.9
1.24
0.03
COD
lb/1000 Ib
_
0.5
25.7
2.7
3.7
2.1
1.3
2.9
55
64.8
43
-
0.89
0.52
-
4.9
1.43
TSS
lb/1000 Ib
„
0.02
14.2
5.1
-
0.58
0.86
• 0.65
28
37.9
33.1
5-30
0.15
5.37
_
19.9
0.01
Total N
lb/1000 Ib
32
0.07
0.34
0.04
0.4
0.33
0.16
0.24
-
0.05
0.03
0.03-0.7
0.03
0.04
3-7
1.27
1.17
Total P
lb/1000 Ib
15
-0
-0
0.01
_
0.01
0.01
0.01
-
_
..
0.8-9.0
„
0.02
0.15-0.3
0.3
0.14
Heavy
Metal
lb/1000 Ib
55-242
-
0.13
1.2
0.003
0.05
0.03
0.04
-
-
-
0.05-0.3
..
0.33
0.01-0.02
1.69
0.02
Oil &
Grease
lb/1000 Ib
-
-
0.68
0.12
0.15
0.14
0.09
0.31
-
-
-
0.06-2
0.1
0.06
0.05-0.08
0.24
0.02
1 Units are million gallons of pollutant per 1000 Ib. of finished product
2 Kaiser Engineers, 1969
3 Pearson, Storrs, Sellech, 1969
-------
TABLE 111-50
SUMMARY OF CURRENT AND PROJECTED WASTE LOADS IN ONE REGION 208 AREA (BY SIC CODE)
OJ
o
cr>
SIC GROUP
No.
201
202
204
205
208
211
22_
226
251
265
27
28_
32_
35_
36_
379
---
9999
Description
Meat Products
Dairy Products
Grain Mill Prods.
Bakery Prods.
Soft Drinks
Tobacco Man.
Textile Mill
Dyeing & Fin.
Furniture
Paperboard Con.
Print. & Pub.
Chem. & Allied P.
Stone, Clay P.
Machinery.
Elect. Equip.
Transp. Equip.
Non-Manuf.
Mun. W.W.T.P.
TOTALS
CURRENT LOADINGS
BOD
(Ib/day)
Sewer
1,523
973
180
935
330
2,024
2,530
0
0
245
0
64
0
32
659
100
1,374
0
10,469
SS
(Ib/day)
Sewer
1,059
400
50
910
40
1,750
2,173
0
0
150
0
29
0
79
402
100
170
0
7,312
BEST PRACTICABLE WASTE REDUCTION TECHNOLOGY
Description
Anaerobic Lagoon to Stabilization Pond
Anaerobic Digestion & Clarification
Oxidation Ditch & Clarification
Rotating Bio-Filters & Clarification
Fixed Activated Sludge
Activated Sludge (E.A.) & Clarification
Activated Sludge & Alum-Aided Clarif.
Carbon Adsorption & Clarification
—
Screening, Ext. Aeration, Clarification
--
Activated Sludge & Clarification
Stilling Ponds, Water Recycle
Oil & Grease Traps
Ion Exchange (for Plating Process)
Oi 1 & Grease Traps
See Text
Upgrade Six Largest Plants
--
Expected
Reductions
BOD ( %)
90
85
85
85
84
85
85
75
--
35
--
85
30
50
10
50
70
SS(X)
85
90
75
65
65
75
75
60
—
65
--
75
70
65
90
65
90
Varies for
Each Plant
--
PROJECTED LOADINGS
BOD
(lb/aay)
Sewer
152
71
27
140
53
304
380
159
10
16
593
50
412
2,367
SS
(Ib/day)
Sewer
117
40
13
319
14
438
543
53
18
28
40
50
17
1,690
-------
3.4.4 Priority Pollutants in Municipal and Industrial Waste Waters
The priority pollutants which appear in municipal wastewaters comes
from three main sources
• industrial effluents
• nonpoint source runoff and
• domestic uses
The proportion from each category will vary from location to location
as well. The types of pollutants in Table 111-51 are those occurring most
frequently in household wastewaters.
According to the data of Feiler (1980) of the 129 priority substances
only 27 occurred at least 30 percent of the time in his sampled treatment
plant influents (Table 111-52). Of those 27, eight were metals. Of the
five most frequently detected, three were metals (zinc, copper and
chromium). The loading of metals in the influent is substantially affected
by the percentage of industrial effluent as shown in Figure 111-18. The
slope of the regression line is 72.86 yg/1/percent.
The mixture of priority substances found in municipal influent will
depend primarily on the mixture of industries contributing flow.
Table 111-53 contains 42 of the 129 priority pollutants categorized by the
industrial effluents in which they will likely be found. This table is
based on screening data provided by U.S. EPA (Neptune, 1981). The 42 which
appear are those which most frequently appeared in the screening data. The
intent of the table is not to imply that these chemicals are necessarily the
most problematic (i.e. carcinogenic, toxic) but only that they are the most
ubiquitous. Characterization of influent concentrations for these
pollutants is not currently possible. Maximum and minimum values such as
those shown in Table 111-52 are more typical of the available data. Jf data
for a particular priority pollutant is necessary some sampling of the
influent and effluent of the treatment plant is recommended.
307
-------
TABLE 111-51
PREDICTED PRIORITY POLLUTANTS IN HOUSEHOLD WASTEWATER
Organ ics_ Inorganics
benzene arsenic
phenol cadmium
2,4,6-trichlorophenol chromium
2-chlorophenol copper
1,2-dichlorobenzene lead
1,4-dichlorobenzene mercury
1,1,1-trichloroethane zinc
naphthalene antimony
toluene silver
diethylphthalate
dimethylphthal ate
trichloroethylene
aldrin
dieldrin
Source: Hathaway, 1980
308
-------
TABLE 111-52°
OCCURRENCE OF PRIORITY POLLUTANTS
IN POTW INFLUENT SAMPLES
Parameter
Zinc
Copper
Cyanide
Chromium
Toluene
Tetrachl oroethyl ene
Chloroform
Methylene Chloride
Trichloroethylene
Bis(2-ethylhexyl) phthal
1,1,1-Trichloroe thane
Nickel
Ethylbenzene
Silver
Phenol
Lead
Cadmium
Mercury
Benzene
Di-n-butyl phthalate
Diethyl phthalate
Butyl benzyl phthalate
Number of
Samples
Analyzed
146
146
150
146
152
152
152
152
152
ate 152
152
146
152
146
152
146
146
146
152
152
152
152
1,2-Trans-dichloroethylene 152
Naphthalene
1,1-dichloroethane
1,1-dichloroethylene
1,2-dichlorobenzene
? Only those substances
° Source: Feiler, 1980
^ 11 nl 4- r* ^-C ,.,-,/ 1
152
152
152
152
detached at least
Percent
of Times
Detected
100
100
99
99
98
97
96
95
95
94
91
87
86
84
83
79
71
70
68
63
62
59
58
55
40
35
30
30% of the
c c
Minimum Maximum
23
34
3
8
2
2
1
1
1
2
1
11
1
2
1
16
1
200
1
1
1
2
1
1
1
1
2
time are
7680
1190
2500
2380
500
1100
430
11000
860
390
1600
1930
448
77
380
935
1800
3900
1560
105
33
140
97
150
24
243
440
included.
309
-------
10000
i r
i i
8000
6000
C
O
tO
s_
+->
c
QJ
O
c
o
4000
2000
» I L
J L
10
20 30
% Industrial Flow
4 0
5 0
FIGURE III-13 CORRELATION OF INFLUENT TOTAL METALS CONCENTRATION
TO PERCENT INDUSTRIAL FLOW
Source: Feiler, 1980
310
-------
TABLE 111-53
INDUSTRIAL CATEGORIES AND FREQUENTLY DETECTED
PRIORITY POLLUTANTS BY CATEGORY
Benzene
Carbon tetrachloride
Chlorobenzene
1,2 dichloroethane
1,1,1 trichloroethane
Chloroform
1,1 dichlosoethylene
1,2 trans-dichlosoethylene
2,4 dimethyl phenol
Ethylbenzene
Methylene chloride
Oi ch 1 orobromonethane
Tnchlorof luoromethane
Naphthalene
Pentachlorophenol
Phenol
Bis(2-ethylhexyl) phthalate
Butyl benzyl phthalate
Di-n-butyl phthalate
Di-n-octyl phthalate
Dimethylphthalate
Chrysene
Anthracene
Flourene
Phenanthrene
Pyrene
Tetrachloroethylene
Toluene
Trichloroethylene
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Copper
Cyanide
Lead
Mercury
Nickel
Silver
Thallium
Zinc
& Detergents
Q.
«
0
Ul
l/l
*->
C
«n
to
•o
t/>
OJ
>
QJ
.C
•o
<
L.
?
+J
ra
(U
•J
•
•
•
OJ
4-*
X
(J
aj
LU
•a
fO
at
4->
(/i
•
cn
c.
c
IH
S
E
3
0)
k.
4->
ai
Q.
0)
a;
4->
to
•a
c
o
t_
•
•
•
•
•
O
•
•
•
•
•
•
•
•
•
•
•
•
i/)
a>
i-
c
n
o
u_
•
•
•
a>
c
4->
O
J^
CL.
anic Chemicals
t_
o
c:
•
•
fO
L>
U
(J
0)
LU
•
& Other Laundries
o
3
<
JZ
0
c
D_
cn
c
(U
o
o
u
0-
u
z
)sives
a.
UJ
lain/Enamel ing
u
i-
o
a.
-a
c
•
U
•J3
XJ
c
ex
_c:
u
QJ
5
•
•
«*
•
•
i|
•
•
•
•
«J
O
O
I
i
1
~j
i
i
i
1
1
Source: Neptune, 1980
311
-------
The evaluation of priority pollutant loads in municipal wastes is the
same as for conventional parameters. First, the influent concentration is
estimated and then the removal of the substance is estimated. Data such as
that found in Table 111-54 can be used to estimate the removal efficiency
for some compounds.
The data of Table 111-55 give typical concentrations of some priority
pollutants in municipal primary and secondary effluent.
-EXAMPLE III-8
Estimation of Municipal Waste
Treatment Loads of Trichloroethylene
A municipal waste treatment plant has a mean influent concentration of
40 ug/1 of trichloroethylene. The plant employs a trickling filter for
secondary treatment. Estimate the daily trichloroethylene effluent load.
The plant flow rate is 1 mgd.
Solution:
From Table 111-54 the removal efficiency of trickling filters for
trichloroethylene (TCE) is 96 percent.
Equation 111-64 gives the effluent load. Since the plant flow rate is
known the estimate of flow (Q • P ) is not necessary. The flow rate of 1
P P
mgd (3.8 x 106 I/day) can be substituted for (Q • P ).
Y(TCE) = 3.8 x 10& (40)(l-.96) x 10"9
m
= 6 x 10"3 kg-TCE/day
END OF EXAMPLE 111-8
312
-------
TABLE 111-54
REDUCTION OF CONVENTIONAL AND PRIORITY POLLUTANTS BY POTW TREATMENT PROCESSES
OO
I—'
OJ
Fraction
Conventionals
Organlcs
Meta 1 s
( 1 ) Two Plant
Parameter
BOO
Total Suspended Solids
COD
01 1 and Grease
Benzene
1,1,1 - Tr Ichloroethane
Chloroform
1,2 - Trans-Dlchloroethylene
Ethyl benzene
Methyl ene Chloride
Tetrachloroethylene
Toluene
Tr Ich loroethylene
Phenol
Naphthalene
Bis (2-Ethylhexyl) Phthalate
Butyl Benzyl Phthalate
DI-N-Butyl Phthlate
Dl ethyl Phthlate
Cad ml urn
Chromium
Copper
Cyanide
Lead
Mercury
Nickel
Si Iver
Zinc
data base (3) Seventeen
(2) Three Plant data base (4) One plant
Pr imary
Treatment 1
17
39
13
52
23
47
23
68
25
14
-
-
30
55
0
-
36
99+
II
_
0
24
57
-
-
-
17
27
plant data
data base
Trickling
) niter<2)
93
94
62
69
96
92
60
98
92
47
78
87
96
96
99+
54
97
0
96
74
74
89
81
94
99+
25
95
80
base
Percent Removal
Activated
Sludge(3)
95
94
90
89
95
87
79
57
92
77
82
71
89
13
28
78
48
31
10
92
91
90
71
74
54
60
83
86
(5) Three plant
Activated
Sludge(4)
90
88
80
90
99+
38
11
99+
80
27
-
96
44
99+
99+
40
99+
99+
99+
99+
76
84
53
99+
86
18
99+
83
data base
Tert 1 ary
Treatment (5)
95
98
91
71
99+
99
25
99+
98
92
98
99
99
99
-
60
99+
64
99+
99+
87
88
-
97
86
51
99+
87
-------
TABLE III-55
CONCENTRATIONS (MEAN t STANDARD ERROR) OF EPA PRIORITY
POLLUTANTS IN THE LOS ANGELES COUNTY JWPCP EFFLUENTS3
Flow (liters/yr x 10")
PH
General constituents (mg/liter)
Total suspended solids
Oil and grease
Ammonia nitrogen
Nitrate nitrogen
Total (K) nitorgen
Total phosphorus
BOO
COD
Fecal coliform (MPN/lOOml)
Volatile organics (pg/liter)
Benzene
Carbon tetrachloride
Chlorobenzene
1,2-dichloroethane
1,1,1-trichloroethane
Chloroform
Ethyl benzene
Methyl ene chloride
Dichlordifluoromethane
Dichl orobronomethane
Tetrachl oroethyl ene
Toluene
Trichloroethylene
Extractable organics (yg/liter)
Acenaphthene
1,2-dichlorobenzene
1,4-di Chlorobenzene
2,4-dimethylphenol
Bis (2-chloroethoxy) methane
Naphthalene
Nitrobenzene
N-ni trosodiphenyl amine
Phenol
Pentachlorophenol
Bis (2-ethylhexyl) phthalate
Di-n-butyl phthalate
Diethyl phthalate
Dimethyl phthalate
Miscel laneous
Asbestos (lOVliter)
Cyanide tug/liter)
Phenol (mg/liter)
Trace metals (mg/liter}
Antimony
Arsenic
Beryll ium
Cadmium
Chromium
Copper
Mercury
Manganese
Nickel
Lead
Selenium
Si Iver
Thallium
Zinc
Primary
4.71
7.3
131
34
37
NA
49
20
200
460
7.5 x 10s
200 ± 34
12 ±2
12 ±1
<10
130 ± 15
34 ± 0
130 ± 6
24 ± 2
6
<10
54 ± 14
310 ± 24
140 ± 5
<10
<10
<10
14 ± 2
<10
-------
REFERENCES
American Public Works Association, 1974. Nationwide Characterization,
Impacts and Critical Evaluation of Stormwater Discharges, Non-sewered
Urban Runoff and Combined Sewered Overflows. Monthly Progress Report
to the U.S. Environmental Protection Agency.
Amy, G., R. Pitt, R. Singh, W.L. Bradford and M.B. LaGraff, 1974. Water
Quality Management Planning for Urban Runoff. U.S. Environmental
Protection Agency, Washington, D.C., (EPA 440/9-75-004) (NTIS PB 2<"
689/AS)
Arthur, R.D., J.D. Cain, and B.F. Barrentine, 1976. Atmospheric Levels of
Pesticides in the Mississippi Delta. Bulletin of Contamination and
Toxicology 15(2):129-134.
Bave, J.L., P. Danlen, and V.P. Kukreja, 1978. Airborne di-Butyl and
di-(2-Ethylhexyl) phthlate at three New York City Air Sampling
Stations. International Journal of Environmental Analytical
Chemistry. 5:189-194.
Bower, C.A., G. Ogata and J.M. Tucker, 1968. Sodium Hazard of Irrigation
Waters as Influenced by Leaching Fraction and by Precipitation on
Solution of Calcium Carbonate. Soil Sci. Soc. Am. Proc. 29-34.
Buckman, H.O., and N.C. Brady, 1969. The Nature and Properties of Soil.
7th ed., The MacMillan Company New York. ~~
Carlston, C.W., and W.C. Langbein, 1960. Rapid Approximation of Drainage
Density: Line Intersection Method. U.S. Geological Survey, Water
Resource Division, Bulletin 11.
Cautreels, W. and K. Vau Canwenberghe, 1978. Experiments on the
Distribution of Organic Pollutants Between Airborne Particulate Matter
and the Corresponding Gas Phase. Atmospheric Environment
12:1133-1141.
Eckenfelder, W.W., 1970. Water Quality Engineering for Practicing
Engineers. Barnes and Noble, Inc., New York.
Eisenreich, S.J., B.B. Looney, and J.D. Thornton, 1981. Airborne Organic
Contaminants in the Great Lakes Ecosystem. Environmental Science and
Technology 15(l):30-38.
El Beit, I.O.D., T. Verner Wheelock and D.E. Cotton, 1981. Factors
Affecting Soil Residues of Dieldrin, Endosulfan, -HCH, Dimethoate and
Pyrolan. Ecotoxicology and Environmental Safety 5:135-160.
Feiler, H., 1980. Fate of Priority Pollutants in Publicly Owned Treatment
Works: Interim Report. EPA 440/1-80-301. October.
Frere, M.H., J.D. Ross, and L.J. Lane, 1980. Chapter 4. The Nutrient
Submodel. I CREAMS: Field Scale Model for Chemicals, Runoff, and
Erosion from Agricultural Management Systems. USDA Conservation
Research Report Number 26.
315
-------
Foster, G.R. 1980. Soil Erosion Modeling: Special Considerations for
Nonpoint Pollution Evaluation of Field-Sized Areas. In Environmental
Impact of Nonpoint Source Pollution. (M.R. Overcash and
J.J. Davidson, eds.) Ann Arbor Science, Ann Arbor, Michigan.
Garstka, W.U., 1964. "Snow and Snow Survey," in Handbook of Applied
Hydrology, ed: V.T. Chow McGraw-Hi 11, Inc.; New York, NewTdrk.
Giam, C.S., E Atlas, H.S. Chan, and G.S. Neff., 1980. Phthalate Esters,
PCB and DDT Residues in the Gulf of Mexico Atmosphere. Atmospheric
Environment 14:65-69.
Haith, D.A., 1980. A Mathematical Model for Estimating Pesticide Losses in
Runoff. Journal of Environmental Quality. 9(3):428-433.
Hathaway, S.W., 1980. Sources of Toxic Compounds in Household Wastewater.
EPA-600/2-80-128. Municipal Environmental Research Laboratory.
Cincinnati, OH.
Hem, J.D., 1970. Study and Interpretation of the Chemical Characteristics
of Natural Water. Geological Survey Water Supply Paper 1473. 2nd ed.
Jenny, H., 1930. A Study on the Influence of Climate Upon the Nitrogen and
Organic Matter Content of the Soil. Missouri Agr. Exp. Sta., Res.
Bui. 152.
Kaiser Engineers, 1969. Final Report to the State of California San
Francisco Bay Delta Water Quality Control Program. Chapter VII.
California Water Quality Control Board, Sacramento, Cal.
Kaufman, D.D., 1974. Degradation of Pesticides by Soil Microorganisms.
Chapter 8 Jjn Pesticides in Soil and Water. W.D. Guengi, ed. Soil
Sci. Soc. Am. Madison, W.I.
Knoblauch, H.C., L. Koloday, and G.D. Brill, 1942. Erosion Losses of Major
Plant Nutrients And Organic Matter from Collington Sandy Loam. Soil
Sci. 53:369-378.
Lager, J.A. and W.G. Smith, 1974. Urban Stormwater Management and
Technology: An Assessment. USEPA National Environmental Research
Center, Cincinnati, Ohio.
Leonard, R.A., G.W. Langdale, and W.G. Fleming, 1979. Herbicide Runoff
from Upland Piedmont Watersheds - Data and Implications for Modeling
Pesticide Transport. J. Environmental Quality. 8(2):223-229.
Logan, T.J., 1980. The Role of Soil and Sediment Chemistry in Modeling
Nonpoint Sources of Pollution, lr\_ Environmental Impace of Nonpoint
Source Pollution. (M.R. Overcash and J.M. Davidson, eds.) Ann Arbor
Science, Ann Arbor, Mich.
Manuel, A.D., R.H. Gustafson, and R. B. Welch, 1968. Three Land Research
Studies. National Commission on Urban Problems, Washington, D.C.,
Report No. 12.
Massey, H.F. and M.L. Jackson, 1952. Selective Erosion of Soil Fertility
Constituents. Soil Sci. Soc. Am. Proc. 16:353-356.
316
-------
Massey, H.F., M.L. Jackson, and O.E. Hays, 1953. "Fertility Erosion on
Two Wisconsin Soils" J. Agron., 45:543-547.
Maxwell, J.C., 1960. Quantitative Geomorphology of the San Dimas
Experimental Forest, California. Columbia University, Department of
Geology, New York, Project No. NR389-042, Report No. 19.
McElroy, A.D., S.Y. Chiu, J.W. Nebgen, A. Aleti, and F.W. Bennett, 1976.
Loading Functions for Assessment of Water Pollution from Nonpoint
Sources. U.S. Environmental Protection Agency, Washington, D.C.,
(EPA-600/2-76-151).
Melton, M.A., 1957. An Analysis of the Relations Among Elements of Climate,
Surface Properties and Geomorphology. Columbia University, Department
of Geology, New York, Project No. NR389-042, Technical Report No.
11.
Menzel, R.G., 1980. Enrichment Ratios for Water Quality Modeling. Chapter
12. Iji CREAMS: A Field Scale Model for Chemicals, Runoff, and
Erosion fronm Agricultural Management Systems. W.G. Knisel, ed. U.S.
Dept. of Agriculture, Conservation Research Report No. 26. p. 640.
Meta Systems, Inc., 1973. Effluent Changes. Prepared for the
U.S. Environmental Protection Agency, Contract No. 68-01-0566.
Metcalf and Eddy, Inc., 1979. Wastewater Engineering: Treatment, Disposal,
Reuse. McGraw-Hill Book Company; New York.
Mockus, J., 1972. Estimation of Direct Runoff from Storm Rainfall. JJT_
National Engineering Handbook, Sec. 4, Hydrology. U.S. Soil
Conservation Service, Washington, D.C.
Murphy, B.D., W.C. Johnson, and E.G. Schlatter, 1980. Simulated
Deposition Rates for SO on a Southeastern U.S. Landscape.
Agricultural Meteorology. 21:179-103.
Nash, R.G., 1974. Plant Uptake of Insecticides, Fungicides and Fumigants
from Soils. Chapter 11 Jji Pesticides in Soil and Water. W.D.
Guengi, ed. Soil Sci. Soc. Am., Madison, WI.
Nash, R.G., 1980. Dissipation of Pesticides from Soils. In CREAMS: A
Field Scale Model for Chemicals, Runoff, and Erosion from Agricultural
Management Systems. W.G. Knisel, ed. U.S. Dept. of Agriculture.
Conservation Research Report No. 26. p. 640.
Neal, O.R., 1944. Removal of Nutrients from the soil by Crops and Erosion.
Agronomy Journal. 36:601-607.
Neff, J.M., 1979. Polycylic Aromatic Hydrocarbons in the Aquatic
Environment: Sources, Fates and Biological Effects. Applied Science
Publishers, Ltd. London.
Neptune, D., 1980. Priority Pollutant Frequency Listing Tabulations and
Descriptive Statistics. Internal Report to R.B. Schaffer, Director
Effluent Guidelines Divison, USEPA, November 4, 1980.
Onstad, C.A., R.A. Young, M.A. Otterby, and R.F. Holt, 1979. Chapter 4.
Sediment Yield Modeling for 208 Planning. J^n Universal Soil Loss
317
-------
Equation: Past, Present, and Future. Soil Sci. Soc. Am. Special
Publication No. 8.
Parker, C.A., e_t al_., 1946. Fertilizers and Lime in the United States.
USDA Misc. Pub. No. 586.
Pearson, E., P. Storrs, and R. Sellech, 1969. Final Report, A
Comprehensive Study of San Francisco Bay. Volume IV: Waste
Discharges and Loadings. University of California Sanitary
Engineering Research Laboratory.
Pionke, H.B. and R.J. DeAngelis, 1980. Method for Distributing Pesticide
Loss in Field Runoff between the Solution and Adsorbed Phase. lr\_
CREAMS: A Field Scale Model of Chemicals, Runoff, and Erosion from
Agricultural Management Systems. USDA. Conservation Research Report
No. 26.
Rao, P.S.C. and J.M. Davidson, 1980. Estimation of Pesticide Retention
and Transformation Parameters Required in Nonpoint Source Pollution
Models. _I_n Environmental Impact of Nonpoint Source Pollution. M.R.
Overcash and J.M. Davidson, eds. Ann Arbor Science Publishers, Inc.
Ann Arbor, Michigan.
Reddy, K.R., 1980. Land Areas Receiving Organic Wastes: Transformations
and Transport in Relation to Nonpoint Source Pollution. In
Environmental Impact of Nonpoint Source Pollution. (M.R. Overcash
and J.M. Davidson, eds.) Ann Arbor Science, Ann Arbor, Michigan.
Rogers, M.T., 1941. Plant Nutrient Losses by Erosion from a Corn, Wheat,
Clover Rotation on Dunmore Silt Loam. Soil Sco. Soc. Am. Proc.
6:162-271.
Startor, J.D. and G.B. Boyd, 1972. Water Pollution Aspects of Street
Surface Contaminants. U.S. Environmental Protection Agency,
Washington, D.C. EPA-Rl-72-081.
Schuman, G.E., R.6. Spomer, and R.F. Piest, 1973. Phosphorus Losses from
Four Agricultural Watersheds on Missouri Valley Loess. Soil Sci.
Soc. Am. Proc. 37(3):424-417.
Schumm, S.A., 1956. "The Evolution of Drainage Systems and Slopes in
Badlands at Perth Amboy, New Jersey." Geo. Soc. Amer. Bull.,
67:597-646.
Schwab, G.O., R.K. Frevert, T.W. Edminster, and K.K. Barnes, 1966. Soil
and Water Conservation Engineering. John Wiley and Sons, New York.
2nd edition.
Sharpley, A.N., 1980. The Enrichment of Soil Phosphorus in Runoff
Sediments. Journal of Environmental Quality. 9(3):531-526.
Skogerboe, G.V. and J.P. Law, Jr., 1971. Research Needs for Irrigation
Return Flow Quality Control. U.S. Environmental Protection Agency,
Report No. 13030-11/71.
Smith, K.G., 1950. "Standards for Grading Texture of Erosional Topography,"
Amer. J. Sci., 248:655-668.
318
-------
Smith, K.G., 1958. "Erosional Processes and landforms in Badlands National
Monument, South Dakota," Geo. Soc. Amer. Bull., 69:975-1008.
Smolenyak, K., 1979. Urban Wet Weather Pollutant Loadings. Masters Thesis.
University of Florida, Dept. of Environmental Engineering.
Stoltenberg, N.L. and J.L. White, 1953. "Selective Loss of Plant
Nutrients by Erosion," Soil Science Society of America, Proceedings.
17:406-410.
Strahler, A.N., 1964. Quantitative Geomorphology of Drainage Basin and
Channel Network. Handbook of Applied Hydrology, ed: V.T. Chow,
McGraw-Hill, Inc.: New York, New YOrk, pp.4-39 to 4-76.
Taylor, A.W., 1967. "Phosphorus and Water Pollution," J. Soil and Water
Conserv., 22-228-231.
Timmons, D.R., R.F. Holt, and J.J. Latterell, 1970. Leaching of Crop
Residues and a Source of Nutrients in Surface Runoff Water. Water
Resources. 6:1367-1375.
U.S. Department of Agriculture, 1954. Diagnosis and Improvement of Saline
and Alkali Soils. USDA Handbook No. 60.
U.S. Department of Agriculture, 1973. Engineering Technical Note No. 16.
Soil Conservation Service, Des Moines, Iowa.
U.S. Department of Agriculture, 1974a. Soils Technical Note No. 3. Soil
Conservation Service, Honolulu, Hawaii.
U.S. Department of Agriculture, 1974b. Agronomy Technical Note No. 32.
Soil Conservation Service, West Technical Service Center, Portland,
Oregon.
U.S. Environmental Protection Agency, 1974. Standards for the Steam
Electric Power Generating Point Source Category. EPA-440/1-74-029.
U.S. Environmental Protection Agency, 1975. Control of Water Pollution
from Cropland. Volumes I and II. EPA-600/2-75-026.
U.S. Environmental Protection Agency, 1975. Standards for the
Pharmaceuticals Manufacturing Point Source Category.
EPA-440/1-75-060.
U.S. Environmental Protection Agency, 1976. Standards for the Explosives
Manufacturing Point Source Category. EPA-440/1-76-060.
U.S. Environmental Protection Agency, 1976. Standards for the Mineral
Mining and Processing Industry. EPA-440/1-76-059.
U.S. Environmental Protection Agency, 1979. Standards for the
Electroplating Point Source Category. EPA-440/1-79-003.
U.S. Environmental Protection Agency, 1979. Standards for the Gum and Wood
Chemical Manufacturing Point Source Category. EPA-440/1-79-078B.
U.S. Environmental Protection Agency, 1979. Standards for the Iron and
Steel Point Source Category. EPA-440/1-79-024A.
319
-------
U.S. Environmental Protection Agency, 1979. Standards for the Leather
Tanning and Finishing Point Source Category. EPA-440/1-79-016.
U.S. Environmental Protection Agency, 1979. Standards for the Petroleum
Refining Industry. EPA-440/1-79-049B.
U.S. Environmental Protection Agency, 1979. Standards for the Textiles
Mills Point Source Category. EPA-440/2-79-020.
Water Resources Administration, 1973. Technical Guide to Erosion and
Sediment Control Design (Draft). Maryland Department of Natural
Resources, Annapolis, Maryland.
Weed, S.B. and J.B. Weber, 1974. Chapter 3. Pesticides—Organic Matter
Interactions. I_n Pesticides in Soil and Water. W.D. Guenzi, ed.
Soil Ci. Soc. Am., Madison, WI.
Weimer, W.C., H.E. McGuire and A.F. Gasperinso, 1976. A Review of Land
Use Nutrient Loading Rate Relationships. Battelle Northwest
Laboratories, Richland, WA.
Williams, J.R., 1975. Sediment-Yield Prediction with Unviersal Equation
Using Runoff Ener;gy Factor. Jji Present and Prospective Technology
for Predicting Sediment Yields and Sources. U.S. Department of
Agriculture. ARS-S-40.
Wischmeier, W.H. and D.D. Smith, 1965. "Predicting Rainfall—Erosion
Losses from Cropland East of the Rocky Mountains," Agriculture
Handbook 282. U.S. Department of Agriculture, Agriculture Research
Service.
Wischmeier, W.H., 1972. "Estimating the Cover and Management Factor or
Undisturbedf Areas," Proceedings, USDA Sediment Yield Workshop. U.S.
Department of Agriculture, Oxford, Mississippi.
Young, D.R., 1978. Priority Pollutants in Municipal Wastewaters. lr\_
Coastalk Water Research Project Annual Report (W. Bascom, editor).
Southern California Coastal Water Research Project, El Segundo,
California.
320
-------
CHAPTER 4
RIVERS AND STREAMS
4.1 INTRODUCTION
The purpose of this chapter is to present simplified tools which can be
used to predict responses of rivers and streams to the impact of pollutants.
The introductory sections to the chapter should be read prior to solving any
problems in order to become familiar with the topics that will be covered
and the limitations of the formulations presented.
Rivers throughout this country are subject to a wide spectrum of
geological, biological, climatological , and anthropogenic impacts which
produce a variety of water quality problems. Approaches which provide
guidance to the solution of these problems, especially ones restricted to
hand calculations, must be limited in scope. The following guidelines have
been used in selecting topics to be considered in this chapter: 1. widely
occuring problems, 2. those amenable to hand calculations, and 3. those
for which planners can obtain sufficient data.
4.1.1 Scope
The major problem areas to be considered are:
• Carbonaceous (CBOD) and Nitrogenous (NBOD) Biochemical Oxygen
Demand
» Dissolved Oxygen
• Temperature (with a discussion of low flow)
• Nutrients and Eutrophication Potential
• Coliform Organisms
• Conservative Constituents
• Sedimentation and Suspended Solids
• Toxic Substances
321
-------
Beginning in 1974, the U.S. Environmental Protection Agency has for
several years published the National Water Quality Inventory which is a
compilation of current water quality conditions and recent trends in the
nation's rivers and lakes. Several of the tables in that report series are
relevant to this document and are included here. Table IV-1 illustrates
reference water quality levels used to define acceptable pollutant limits in
U.S. waterways. Table IV-2 shows water quality conditions in eight major
waterways in the United States, while Table IV-3 summarizes the most widely
observed water quality problems in the U.S. These tables will be cited
throughout this chapter.
Local water quality standards, when they exist, are preferable to the
general guidelines provided in Table IV-1. Table IV-4 shows example
standards for dissolved oxygen and water temperature for the states of
Virginia and Maryland. Parts of the standards are significantly different
from the reference levels in Table IV-1. For example the daily average
dissolved oxygen standard for natural trout water for the state of Virginia
is 7.0 mg/1, while 5.0 mg/1 is recommended for the protection of aquatic
life (Table IV-1). Thus, when local standards exist, they should be used in
lieu of general reference levels.
4.1.2 Significance of Problem Areas
Oxygen depletion is often the result of excessive CBOD and NBOD
loadings particularly in combination with high temperature and low flow
conditions. Increased nutrient loadings to streams which produce elevated
ambient concentrations can pose substantial potential for eutrophication.
The nutrient problem is currently one of the most widespread areas of
concern regarding river water quality. The health hazards category in Table
IV-3 lists elevated coliform levels as a problem of particular concern in
northeastern and Great Lakes States. Salinity has been identified as a
major problem in the central and southwestern states.
322
-------
TABLE IV-1
REFERENCE LEVEL VALUES OF SELECTED WATER QUALITY
INDICATORS FOR U.S. WATERWAYS (U.S. EPA, 1976)
Parameter
Reference Level
Ammonia
Color
Dissolved Oxygen
Dissolved Solids
Fecal Coliforms
Nitrate-N
pH
Phenols
Suspended Solids and
Turbidity
Total Dissolved Gases
£ 0.02 mg/1 as unionized ammonia
(for freshwater aquatic life)
£ 75 platinum-cobalt units (for
water supply)
>_ 5.0 mg/1 (to maintain fish
populations)
£ 250 mg/1 (for water supply)
log mean £ 200 per ml over 30 days
and 90 percent £ 400 per ml (for
bathing waters)
£ 10 mg/1 (water supply)
between 6.5 and 9.0 (for freshwater
aquatic life)
£ 1 yg/1 (for water supply)
shall not reduce the depth of the
compensation point by more than
10 percent (aquatic life)
£ 110 percent saturation (aquatic
life)
323
-------
TABLE IV-2
CONDITION OF EIGHT MAJOR WATERWAYS (EPA,1974)
River
Harmful
Substances
Physical
Modification
Eutrophication
Potential
CO
ro
Mississippi
Missouri
Ohio
Tennessee
Detroit area
rivers
Columbia
Trace metals
present in middle
river
High*, increasing
iron and
manganese
Cyanide present
but dimishing
Severe gas super
saturation; some radio-
activity in lower river
High* turbidity and
solids below
Missouri River
High* suspended solids,
turbidity in middle and
lower river
High* suspended solids
in lower river, some
improvements
Suspended solids
improving, local
temperature effects
from discharges
Occasional high*
temperatures
High*, increasing
nutrients but no
algae
High*, increasing
nutrients but no
algae
High* nutrients but
no algae
Small increase in
nutrients but no
algae
Nutrients discharged
to Lake Erie
decreasing
High* nutrients but
no algae, except for
slime growths in
lower river
Snake
Severe gas super-
saturation, signif-
cant pesticides
Turbidity from
natural erosion,
agricultural practices,
reservoir flushing
Nuisance algal
blooms each
summer
-------
TABLE1 IV-2 (continued)
River
Willamette
Harmful
Substances
Significant sulfite
Physical
Modification
High* turbidity at
Eutrophication
Potential
High* level of
waste liquor from
pulp and paper wastes
high flow, high
temperature in summer
nutrients but
not excessive algae
River
GO
ro
Mississippi
Missouri
Ohio
Tennessee
Detroit area
rivers
Salinity, Acidity,
and Alkalinity
Oxygen
Depletion
High* salinity, acidity
below major
tributaries
High* dissolved salts
in middle and lower
river
Low* alkalinity
especially in upper
river
Health Hazards and
Aesthetic Degradation
Acids and chloride
improving despite
large discharges
low,1
Oxygen-demanding
loads from large
cities evident
High* organic loads
from feedlots,
improved near cities
Occasional low*
dissolved oxygen near
Cincinnati and Pittsburgh
Low* BOD and
decreasing COD in
reservoirs
Low* dissolved oxygen
only at mouths of
area tributaries
Commercial fishing
eliminated in lower
river by phenols,
bacteria near cities
High* bacteria and
viruses present in wet and
dry periods
High* bacteria especially
in high population
areas
High* bacteria in small
areas near cities, low
radionuclides
Phenols decreasing,
bacteria unchanged-
to-higher
-------
TABLE IV-2 (continued)
Salinity, Acidity, Oxygen Health Hazards and
River and Alkalinity Depletion Aesthetic Degradation
Columbia Approaches ideal Dissolved oxygen Very low* bacteria
for fresh waters close to saturation
Snake High* dissolved Dissolved oxygen High* bacteria
solids from irrigation close to saturation below population
in middle river centers
Willamette Low* dissolved mineral Improved dissolved High* bacteria, but
salts, improved pH oxygen, no standards improving
violations
*High (or low) relative to other rivers, or relative to other sections of river, or to
national reference levels. Does not necessarily imply standards violations or
dangerous condition.
-------
TABLE IV-3
co
ro
WATER QUALITY PROBLEM AREAS REPORTED BY STATES*
NUMBER REPORTING PROBLEMS/TOTAL (EPA,1975)
Oxygen
depletion
Eutrophi-
cation
potential
Health
hazards
Salinity,
acidity,
alkalinity
Physical
modification
Harmful
substances
Middle
Atlantic,
Northeast
11/13
11/13
11/13
3/13
7/13
6/13
South
9/9
6/9
8/9
6/9
3/9
6/9
Great
Lakes
6/6
6/6
5/6
2/6
3/6
5/6
Central
6/8
8/8
8/8
6/8
8/8
4/8
Southwest
4/4
2/4
3/4
4/4
3/4
4/4
West
6/6
6/6
5/6
4/6
6/6
2/6
Islands
4/6
4/6
5/6
2/6
5/6
3/6
Total
46/52
43/52
45/52
27/52
35/52
30/52
* Localized or statewide problems discussed by the States in their reports.
-------
TABLE IV-4
EXAMPLE RIVER WATER QUALITY STANDARDS
VIRGINIA
CO
ro
oo
CLASS
III
IV
V
VI
DESCRIPTION
COASTAL AND PIEDMONT
MOUNTAINOUS
PUT AND TAKE TROUT WATERS
NATURAL TROUT WATERS
DISSOLVED
MINIMUM
4,0
4.0
5.0
6.0
OXYGEN
AVERAGE
5.0
5.0
6.Q
7.0
TEMPERATURE,
*TN
5
5
5
5
°F
'MAX
90
37
70
68
MARYLAND
CLASS
I
III
IV
DESCRIPTION
WATER CONTACT, RECREATION
NATURAL TROUT WATERS
RECREATIONAL TROUT WATERS
DISSOLVED OXYGEN*
MINIMUM AVERAGE
4.0 5.0
5.0 6.0
4.0 5.0
TEMPERATURE
MAXIMUM, °F**
90
63
75
*THESE VALUES APPLY EXCEPT WHERE LOWER VALUES OCCUR NATURALLY.
**THESE APPLY OUTSIDE THE MIXING ZONE, IF NATURAL TEMPERATURE OF RECEIVING WATER
IS GREATER THAN THE STANDARD, THEN THAT BECOMES THE STANDARD.
-------
Because of their importance, each of the problem areas described will
be addressed in this chapter. As shown in Table IV-5, many states routinely
measure the parameters associated with these problems. The total number of
states responding to the survey was 47. Because of the routine surveys
conducted, data are commonly available for performing hand calculations.
NBOD, though not directly measured, can be found from measurements of
organic and ammonia nitrogen. Chloride concentration measurements can be
directly converted to salinity.
4.1.3 Applicability to Other Problems
The tools which are presented in this chapter are designed to address
specific water quality problems. However, a number of the tools, which are
based on the law of mass conservation, can be directly applied to other
problems with little or no modification. In the case of temperature
prediction, an energy balance is used (which is analogous to a mass
balance).
The degree of commonality of source and sinks of a particular pollutant
e.g. a nutrient) or water quality indicator (e.g. dissolved oxygen) is
responsible for the similarities and differences among the specific
equations. For example, CBOD and NBOD produce a similar general effect
(oxygen depletion), generally have similar sources and sinks, and for
purposes of this study, are assumed to follow first order decay kinetics.
Coliforms, also assumed to decay by first order kinetics, are handled by the
mass-balance approach. Conservative substances are different from BOD and
coliforms in that they do not decay. Finally, there are some instances
where a more subjective analysis is indicated, and neither a mass nor energy
balance is presented.
Once the similarities among water quality parameters are understood,
handling two seemingly different problems can often be accomplished in a
straightforward and similar fashion. For example, the distribution of toxic
substances that are either conservative or follow a first order decay may be
evaluated using techniques described for conservative substances and
coliforms, respectively.
329
-------
TABLE JV-5
WATER QUALITY PARAMETERS
COMMONLY MONITORED BY STATES* (EPA,1975)
Number
Parameter of States
Flow 47
Dissolved oxygen 47
Coliform bacteria 45
Nitrogen (any form) 39
Phosphorus (any form) 35
PH 35
BOD/COD/TOC 27
Water temperature 29
Turbidity 26
Solids (any type) 27
Metals (any type) 17
Chlorides 19
Alkalinity 15
Conductivity 16
Color 11
Sulfate 14
*0nly parameters listed by at least 10 States and specified as being
part of each State's monitoring program are included.
330
-------
4.1.4 Sources of Pollutants
Pollutant loadings originate from three general sources: point,
nonpoint, and natural. Each of these can constitute a major hurdle in
meeting the 1983 goals of fishable and swimmable waters. Specifically,
point sources (30 states), nonpoint sources (37 states), and natural
conditions (21 states) are all major contributors to water quality problems
(EPA, 1975).
It is imperative that the capacity to assess impacts of nonpoint
sources be a part of the hand calculation methodology for rivers. Table
IV-6 illustrates the importance of nonpoint source nutrient loading for
selected rivers in Iowa. Up to 96 percent of the annual phosphorus load and
up to 99 percent of the total nitrogen load are from nonpoint sources.
Admittedly, quantification of nonpoint source loads is often difficult.
Nevertheless, simplified nonpoint source terms will be included in some of
the mass-balance formulations. The methodology supplied in Chapter III can
be used to estimate the nonpoint source loading rates.
4.1.5 Assumptions
In deriving the mass-balance equations, a number of assumptions were
made. Users should be aware of each assumption so that the tools are not
misapplied. The most important assumptions are:
• The system is at steady-state
t Dispersion is small compared to advection (i.e. plug flow is
assumed)
• The river system is vertically and laterally mixed
• When pollutants decay, the rates are first order.
The steady-state assumption means that conditions are not changing with
time, but only as a function of distance along the river. The time scale
for steady-state generally should be on the order of a week or longer. For
331
-------
TABLE IV-6
ANNUAL PHOSPHORUS AND NITROGEN LOAD FOR SELECTED IOWA RIVER BASINS (EPA,1975)
CO
CO
r\>
River
Floyd
Little Sioux
Char i ton
Des Moines
Iowa
Cedar
*0rthophosphate
River
Floyd
Little Sioux
Chariton
Des Moines
Iowa
Cedar
Total
(Ibs/year)
720,207
1,851,632
879,916
5,621,007
1,723,975*
5,099,507
Total
(Ibs/year)
1,705,984
9,609,556
1,585,427
41,334,897
2,075,830
6,804,881
ANNUAL PHOSPHORUS LOAD
Point Sources
(Ibs/year)
29,807
129,088
48,203
586,015
103,445*
1,526,775
ANNUAL NITROGEN LOAD
Point Sources
(Ibs/year)
65,171
85,308
24,795
695,235
91,287
1,552,334
Nonpoint Sources
(Ibs/year)
690,400
1,722,544
831,713
5,034,992
1,620,530*
3,572,732
Nonpoint Sources
(Ibs/year)
1,640,813
9,522,248
1 ,560,632
40,639,662
1,984,543
.5,252,547
Percent of
Total from
Nonpoint Sources
95.9
93.0
94.5
89.6
94.0
70.1
Percent of
Total from
Nonpoint Sources
96.2
99.1
98.4
98.3
95.6
77.2
-------
example, the summer low flow period generally represents a steady-state
situation. However, storm events, and the dynamic responses of a river to
them, must be considered a transient phenomenon.
Dispersion effects can usually be neglected when pollutant input into a
river is continuous. Under these conditions the plug flow assumption is
reasonable because the net dispersive transport is small. However, when a
slug of pollutant is discharged instantaneously, dispersive transport is
important since high concentration gradients exist around the centroid of
the discharged pollutant.
The fully-mixed assumption presupposes that concentration gradients
exist only in the direction of flow (longitudinal direction) and not in
either the vertical or lateral direction. The final major assumption is
that all decay rates can be approximated by first order kinetics. This
means that the decay rate of a substance is proportional to the amount
present. First order decay is traditionally used in CBOD computations, and
occasionally in nitrogen oxidation. The oxidation of inorganic nitrogen
actually proceeds in stages from ammonia-N to nitrite-N to nitrate-N.
However, for purposes of this report, the first order decay rate is
acceptable for NBOD and coliforms, as well as CBOD. Before applying first
order decay to other substances, however, care should be taken to determine
the validity of the assumption.
For a few of the analyses which follow, several of the aforementioned
assumptions are relaxed. In the discussion of mixing zones, Section 4.1.9,
partial mixing is discussed for wide rivers. In the discussion on
toxicants, Section 4.9, the spill analysis requires that an unsteady-state
situation be analyzed where the effects of dispersion are included.
4.1.6 Data Requirements
Required in the analysis of most water quality problems are one or more
types of data. For example, stream velocity (U), volumetric flow rate (Q),
and stream temperatures (T) are commonly needed. Decay rates, specific to
the particular problem at hand, are also required.
333
-------
The U.S. Environmental Protection Agency has published a document
(Zison et^ aj^., 1978) intended to provide water quality modelers with a
comprehensive source of information on rate constants and coefficients. The
document provides extensive information on both biological and water quality
parameters commonly used in surface water modeling. The contents of the
document will be useful to the users of this document, who are often faced
with limited information on process rates for the water bodies being
analyzed.
Stream velocity is the most basic hydraulic parameter needed for the
analyses presented in this chapter. Ideally, the appropriate stream
velocity is the travel time of neutrally buoyant particles over the reach
being investigated divided by the distance traveled. Note that this concept
of velocity is different from the velocity determined by:
As defined by Equation IV-1, this concept of velocity exists only at the
point in the river where the cross-sectional area is A. If the point of
measurement is not typical of the reach being investigated, then neither
will the velocity be typical. Consequently, should the user predict stream
velocity based on cross-sectional area, a location typical of the river
reach being investigated should be chosen.
An alternate method of predicting stream velocity, which does not
depend on either flowrate Q or cross-sectional area A is Manning's Equation.
A complete description of the use of this approach is given in many texts on
surface water hydraulics, one of the best being Chow (1959).
According to Manning's Equation stream velocity under uniform flow
conditions is expressible as:
U=L49 sl/2 2/3 (IV.2)
n n
where
n = Manning's n
S = slope, ft/ft
334
-------
R = hydraulic radius, ft
H
Manning's n is the roughness of the stream bed, and can be predicted as
outlined in Chow (1959). Barnes (1967) provides roughness data for 90
streams in the United States, and includes cross-sectional areas and
photographs of the streams investigated. The slope can be estimated using
topographic maps to predict elevation changes between two locations, and
then overlying a string over the stream path to predict distance. The
hydraulic radius (which is the ratio of the cross-sectional area to wetted
perimeter) can be estimated in terms of depth when the stream width is much
greater than the depth. Specifically,
/depth, if channel cross-section is rectangular
V
v/3 maximum depth, if channel cross-section is parabolic
4.1.7 Selecti on of Season
It is reasonable to expect that a particular water quality problem may
be more severe at one time of the year than another. Table IV-7 shows that
pollutant levels can depend on season (summer or winter) and flow rate (high
flow or low flow). Dissolved oxygen problems, for example, are clearly
associated with summer, low flow conditions. Consequently, for any
particular pollution problem, users should strive to perform the analysis
under critical conditions. Where planning is performed with consideration
of the aggravated situation, and where proper abatement action is taken, it
is likely that pollution concentrations will be below problem levels during
other times of the year. If a problem in fact exists, then it is under
these conditions that it will be most pronounced.
In the following sections, hand calculation methods for each problem
area are described with illustrative examples. Table IV-8 provides a
summary of the material presented.
335
-------
TABLE IV-7
co
CO
MAJOR WATERWAYS: SEASONAL AND FLOW ANALYSIS, 1968-72 (EPA, 1974)
Parameters
Suspended solids
Turbidity
Color
Ammonia
Nitrite
Nitrate(as N)
Nitrate(as N03)
Nitrite plus nitrate
Organic nitrogen
Total Kjeldahl nitrogen
Total phosphorus
Total phosphate
Dissolved phosphate
Dissolved solids(105°C)
Dissolved solids(180°C)
Chlorides
Sulfates
Al kal ini ty
ph
Dissolved oxygen
BOD5
COD (.025N)
Total coliforms(MFD)*
Total coliforms(MFI)*
Total coliforms(MPN)*
Fecal col iforms(MF)*
Fecal coliforms(MPN)*
Phenol s
Odor
Winter,
High Flow
9
13
11
14
3
12
8
2
3
3
10
8
6
4
3
4
5
6
15
0
11
6
4
8
4
6
4
5
4
Summer,
Low Flow
(number of
5
4
6
3
7
4
3
3
6
5
3
3
3
7
8
15
13
12
4
19
6
5
10
6
2
6
0
0
0
Winter,
Low Flow
reaches exceeding
0
1
3
7
5
8
6
2
0
0
5
5
4
6
6
10
5
10
6
0
8
3
2
2
3
3
1
1
0
Summer,
High Flow
reference
4
7
4
1
1
1
1
1
3
3
2
1
0
3
2
0
5
0
4
9
1
2
5
4
3
4
0
0
0
Dominant
Effect
levels)**
High flow
High flow
High flow
Cold weather
Low flow
Cold weather
Cold weather
Inconclusive
Warm weather
Warm weather
Cold weather
Cold weather
Cold weather
Low flow
Low flow
Low flow
Warm weather, low flow
Low flow
Cold weather, high flow
Warm weather
Cold weather
Cold weather
Warm weather
High flow, warm weather
Inconclusive
Inconclusive
Cold weather
Inconclusive
Inconclusive
['•mbrane filter delayed, membrane filter immediate, most probable number, membrane filter.
'"'. ference levels are available in Table IV-1. Thirty reaches were analyzed during each season.
-------
TABLE IV-8
WATER QUALITY ANALYSES FOR RIVER SCREENING METHODOLOGY
Water Quality Constituent
Computational Procedures
Supporting Information Included
Water temperature
equilibrium temperature
mixing temperature
temperature profile for point sources
shortwave solar radiation
longwave solar radiation
vapor pressure
Carbonaceous and nitrogenous
biochemical oxygen demand
BOD profiles for point sources
BOD profiles with benthic sources added
BOD profiles with both benthic and nonpoint
sources added
graphs, tables, and equations
tor decay rates
Dissolved oxygen
U)
CO
- CB8D-NBOD-DO profile for point sources
- DO profiles with photosynthetic oxygen
production and benthic uptake added
- critical dissolved oxygen conditions
- waste assimilative capacity
reaeration rates for shallow and
deep streams
saturation dissolved oxygen levels
corrected for temperature, altitude,
salinity
photosynthetic oxygen production and
benthic uptake data
tabulated solutions for critical
dissolved oxygen levels
Nutrients
growth limiting nutrient
nutrient profiles for point sources
nutrient profiles for nonpoint sources
nitrogen/phosphorus ratios tor
growth 1 imitation
nonpoint source loading rates by
land use type
Coliform organisms
coliform profiles for point sources
coliform profiles for nonpoint sources
- decay rates
Sediment
bed load
suspended load
total load
median bed particle sues for
numerous rivers
critical shear stress
sediment transport propensity factor
approximate bed load/suspended load
relationship
Toxicants
toxicant profiles for point and
nonpoint sources
• mass flux volatilized, advected, and
transformed
- spill analysis of low and high density
toxicants
• time required to desorb toxicant from
bedded sediments
vnpor pressure, solubility,octanol -
water portion coefficient for
priotity pollutants
Henry's I dw Constants
-------
4.1.8 River Segmentation
Although the tools presented in this chapter are of a simplified nature
they can be used to analyze complex river systems (i.e. those which have a
number of different point and nonpoint sources of pollution, tributaries and
withdrawals). Analysis of these systems is accomplished by dividing the
river into segments. The basic tenet which must be followed is simply this:
Segments are created so that one of the analytical tools presented in this
chapter can be used to predict the pollutant concentration profile within
the segment.
Analyses of river systems normally begin at a segment where the
boundary conditions are known, and proceed sequentially downstream. Thus
the results found for one segment are used as the upstream boundary
condition for the next segment. Based on the tools in this chapter, the
following rules should be followed when segmenting:
1. Point sources of pollutants enter the river just above the
upstream boundary of a segment. Tributaries are treated as
point sources.
2. Nonpoint sources enter a river throughout the length of a
segment.
3. Pollutant concentrations at the upstream end of segments are
obtained by mixing the pollutant concentration in the river
with the contribution of the point source at that location (if
one exists). The location where mixing occurs is called a
mixing zone.
4. Generally constant hydraulic variables (e.g. depth and
velocity) are used throughout a segment. If there is a
gradual change in the hydraulic variables over distance, an
average value can be used. If there is an abrupt change in
the variable, such as a velocity change caused by a
significant deepening of the channel, then a new segment can
be created at this boundary.
338
-------
5. Decay rates, reaeration rates, and other rate coefficients
remain constant within a segment. If rate coefficients are
known to change significantly from one location to another in
a river, then different segments should be created. This rule
is consistent with rule (4), since rate coefficients are often
functions of hydraulic variables.
EXAMPLE IV-1
Figure IV-la shows a stretch of the James River, located in Virginia.
On the stretch of the river shown, there is a tributary (Falling Creek), a
sewage treatment plant (STP), and a nonpoint source of runoff
(agricultural). Segment the river between locations A and B in order to
determine the profile of a pollutant which is discharged from each of the
three sources.
First, it should be noted that often there is more than one way to
segment the river to successfully solve the problem. The most obvious
method will be illustrated here. Figure IV-lb shows the proposed solution.
There are two mixing zones - the first around the treatment plant and the
second around the tributary which is treated as a point source. The first
segment is located from below the first mixing zone to above the second
mixing zone, and has a nonpoint source discharging throughout the length of
the segment, consistent with rule (2). The second segment is located below
the second mixing zone and continues downstream to location B, which is the
end of the nonpoint source. If Falling Creek had not been present, a single
segment and a single mixing zone would have been sufficient to analyze the
problem.
END OF EXAMPLE IV-1
A second, more comprehensive example will illustrate a number of points
about segmentation not covered in the previous example. One of these points
is that the segmentation scheme used can vary depending on the pollutant
being analyzed.
339
-------
JAMES RIVER
> B
AGRICULTURAL RUNOFF
(a) River Segment Being Analyzed
(b) Proposed Segmentation Scheme
FIGURE IV-1 ILLUSTRATION OF RIVER SEGMENTATION
PROCEDURE ON THE JAMES RIVER,
340
-------
EXAMPLE IV-2
Segment the river shown in Figure IV-2 beginning at location A and
continuing to location B in order to determine the instream BOD
distribution. How would the segmentation differ when predicting the
dissolved oxygen profile?
Both point and nonpoint sources discharge to the river in Figure IV-2.
Several flows are diverted, and the river width changes over parts of the
reach being investigated. Each of the rules stated earlier will be utilized
to segment the river system. Figure IV-3 shows one solution to the problem.
Depending on the distances between the various sources of pollutants, which
are not given in the problem, it might be possible to combine some of the
segments. The reservoir is assumed to be analyzed using the methods in
Chapter 5, and so is not segmented.
Mixing zones are included around the four point sources: the food
processing plant, the tributary, the sewage treatment plant, and the pulp
mill. In segments 9 and 11 there appear to be a number of point sources and
diversions. Strictly speaking, segments 9 and 11 do not follow the rules
presented earlier, which require mixing zones around each point source.
However, the point sources and sinks within segments 9 and 11 are assumed to
represent equivalent nonpoint sources, which act over the length of each
segment. This approach can obviously simplify the analysis of complex river
systems by decreasing the number of segments analyzed. However, the
analysis is more approximate because the nonpoint source is assumed to be
uniformly distributed throughout the segment. Example IV-5 presented later
shows a specific application of the concept of an equivalent nonpoint
source.
In segment 2 the presence of the small dam is assumed not to influence
the BOD profile, so that its presence does not require a mixing zone.
However if the dissolved oxygen profile were being calculated, segment 2
would be divided into two segments, with a mixing zone around the dam. This
division is required because the dissolved oxygen concentration can rapidly
341
-------
J
SMALL DAM
n
AGRICULTURE
(CONT.)
VJIOENS)
ATTACHED
ALGAE
(CON'T)
RESERVOIR
n
DIVERTED FLOWS
FIGURE IV-2
HYPOTHETICAL RIVER HAVING A VARIETY OF
POLLUTANT SOURCES AND SINKS,
342
-------
SMALL DAM
i
I
1
^G
E
©
1 | •
*'*!* /^\
1 ©
MIXING
7nwct;
\t
H
\
1
' i
I J
' J
k J
AGRICULTURE
1 — >- t-r
t J | (CONT.)
i'®"!' r
/T\ Z°NE
(CON'T.)
DIVERTED FLOWS
FIGURE IV-3 RIVER SEGMENTATION FOR BOD DISTRIBUTION,
343
-------
change (almost instantaneously) as the water flows over the dam. The
dissolved oxygen concentration just below the dam should be used as the
upstream boundary conditions for the next segment. The specific information
required to accomplish this is discussed later in Section 4.3.
A second difference in segmenting for dissolved oxygen occurs in
Segment 8. The presence of the attached algae is assumed not to influence
the BOD profile, but the algae are internal sources of oxygen. Thus segment
8 would be subdivided at the upstream location of the attached algal growth.
END OF EXAMPLE IV-2
4.1.9 Mixing Zones
A mixing zone, as used in this document, is nothing more than a short
reach of a river where a point source and river water mix. It is often
assumed, for both simple and more complex approaches (e.g. QUAL-II computer
model), that mixing is instantaneous and complete across the entire width of
the channel. With several exceptions, such an approach is used in this
document.
Assuming complete mixing, the concentration of a pollutant in a river
after mixing is:
C Q + C Q
r u u wxw
Q+Q (IV-3a)
(iv-3b)
Qw + Qu
where
C = concentration of pollutant in river following mixing, mg/1
C,, = concentration in point source, mg/1
w
C = concentration in river above point source, mg/1
Q = discharge rate of point source, ft /sec
W
Qu = flow rate of river above point of discharge, ft3/sec
W = pollutant mass emission rate, Ibs/day
344
-------
The concentration C is the pollutant level in the mixing zones shown in the
earlier Figures IV-1 and IV-3. These concentrations become the upstream
boundary conditions for the adjacent downstream segment.
Although it is convenient to assume that complete mixing occurs, this
assumption may be inaccurate for wide rivers, depending on the
characteristics of the point source outfall and diffuser. Figure IV-4
illustrates such a case. The river is wide enough so that the wastewater is
initially mixed with only a fraction of the total river flow. As the
pollutant-riverwater mixture is transported downstream mixing continues
until the pollutant is completely mixed across the channel.
The initial pollutant concentration at the point of discharge is
Y
c = W Qu Cu * Qw Cw (IV-4)
QW + tf Qu
where
V
jj- = fractional distance across river where initial mixing occurs.
and all other variables have been previously defined.
The significance of incomplete initial mixing is that pollutant
concentrations can be initially much higher than if complete mixing occurs.
For example, if the upstream contribution of the pollutant is negligible
(C = 0) and if the fraction of river flow which initially mixes is far
Y
greater than the wastewater flow (~ Q »Q ), then
r = — r fIV-5)
L Y Ccm U '
where
C = concentration of pollutant if there is incomplete initial
mixing
Ccm = concentration of pollutant if there is complete initial
mixing
345
-------
I I I I I I I I I I I I I I II I I I I I I I I I 11 I I I I 11 I I I I I I I I I I I I I
CONTAMINATED '.
UNCONTAMINATED
•;.-CONTAMINATED .
T
W
U
I I I I II I I l\l I I I I I I I I I I I I I I I I I I I I I I
LJ I I 1 I I I I II III
FIGURE IV-4 POLLUTANT DISCHARGE WHERE INITIAL MIXING
OCCURS A FRACTIONAL DISTANCE ACROSS THE
RIVER,
346
-------
If Y/W = 0.1, then the pollutant concentration following incomplete mixing
is 10 times higher than if complete mixing were to occur.
The distance L to complete mixing (see Figure IV-4) can be estimated
(as an upper limit) by the following expression:
L = 0.4 W2 U (IV_6)
where
L = distance below point source where complete mixing occurs
W = width of river
U = river velocity
e = lateral diffusion coefficient
Values of the lateral diffusion coefficient can be estimated from the data
given in Table IV-9. Also, the following predictive formula can be used:
,0.1-0.2, for a straight rectangular flume
4- 1
f = <0.25, for irrigation channels (IV-7)
^0.4-0.8, many natural channels
where
D = mean depth of flow
u* = friction velocity = VgDS
S = slope of channel
The actual distance L is probably less than that calculated from Equation
IV-6 because of secondary mixing, river curvature, and initial momentum of
the discharge. It is also sensitive to river width.
4.1.10 Hater and Pollutant Balances
Many river systems are hydrologically complex. Flow patterns are
influenced by tributaries, nonpoint sources of runoff, flow withdrawals, as
well as point sources of pollution. If the planner intends to perform water
quality analyses on a basin wide scale, it is probably prudent that a water
347
-------
TABLE IV-9
EXPERIMENTAL MEASUREMENTS OF TRANSVERSE MIXING IN
OPEN CHANNELS WITH CURVES AND IRREGULAR SIDES
co
-P>
CO
Channel
Missouri River near
Blair, Nebraska
Laboratory
Laboratory model
of the Ussel River
IJssel River
Mackenzie River
from Fort Simpson
to Norman Wells
Missouri River
downstream of
Cooper Nuclear
station, Nebraska
Potomac River;
29 km reach below
the Dickerson
Power Plant
Channel
geometry
Meandering river
Smooth sides and
bottom; 0.15 m
long groins on
both sides
Smooth sides and
bottom; 0.5 m
long groins on
both sides
Groins on sides
and gentle curvature
Groins on sides
and gentle curvature
Generally straight
alignment or slight
curvature; numerous
island and sand bars
Reach includes one
90° and one 180°
bend
Gently meandering
river with up to
60° bends
Channel
width,
W
(m)
200
2.2
2.2
1.22
69.5
1240
210-270
350
Mean depth
of flow,
d
(m)
2.7
0.097
0.097
0.9
4.0
6.7
4
0.73-1.74
Mean
velocity,
(m/s)
1.75
0.11
0.11
0.13
0.96
1.77
5.4
0.29-0.58
Shear
velocity,
u*
(m/s)
0.074
-
-
0.0078
0.075
0.152
0.08
0.033-0.051
Transverse
mixing
coefficient
(m2/sec)
0.12
-
-
.
-
0.67
1.1
Et
Du*
0.6
0.36-0.49
0.3-0.4
0.45-0.77
0.51
3.4
3.4
0.52-0.65
from: Fischer, H.B., E.J. List, R.C.Y. Kob, J. Imberger, and N.H. Brooks, 1979.
Mixing in Inland and oastal Waters. Adademic Press, New York.
-------
budget be first completed. A water budget is a statement that
jjf = £ Inflows - £ Outflows = 0, for steady-state (IV-8)
where
S = storage in the river channel
For the steady-state situations, which are examined here, the water budget
simply states that inflows to the system equal outflows from the system. A
water budget thus provides a method of determining whether the major flow
contributions have been accurately assessed or not. If a large imbalance in
the water budget exists, accurate evaluation of the major sources of
pollutant might be difficult to achieve. An accurate water balance helps to
minimize the possibility of inaccurate assessment of pollutant concentra-
tion. It does not eliminate the possibility.
Once a water balance has been completed, then a pollutant balance of a
conservative pollutant can be developed based on the following relationship:
in
Flux = £ Flux
out
,at steady-state
(IV-9)
where the fluxes are the rates of entry and loss of the pollutant into and
out of the system, respectively. One of the following two expressions can
be used to predict the mass loading rates:
M = 5.38 C Q
(IV-10)
where
M = mass loading rate, Ibs/day
C = concentration, mg/1
Q = flow rate, ft3/sec
and
M = 86.4 C Q
(IV-11)
where
M = mass loading rate, kgs/day
Q = flow rate, m3/sec
349
-------
When nutrient and water balances are developed, the following considerations
should be kept in mind:
1. In most instances it is probably not possible to develop water
or nutrient balances where inflows and outflows balance to
within less than 10 percent of each other.
2. The system's upstream boundary must be included in the balance
as a source and the downstream boundary as a loss.
3. All sources and losses should be mutually exclusive of each
other.
4. Choose system boundaries to coincide with locations of gaging
stations when possible.
5. Try to use comparable periods of record of data. This will
help to minimize the impacts of trends which could be present
in one record but not in another.
6. It is typically easier to develop water and mass balances on
an annual basis, although balances can be developed for each
season of the year. However, if the system is not at steady
state, inflows and outflows should not balance.
Table IV-10 shows a suggested method of tabulating the results of water
and pollutant balances. Total nitrogen (TN) and total phosphorus (TP) are
the pollutants. All flow rates and loading rates are tabulated
individually. Once total loading rates have been tabulated, the percent
contribution from each source can be determined. Percent contributions help
to determine the relative importances of each source as a contributor to
pollution, and can provide a method to prioritize pollution abatement
efforts.
350
-------
TABLE IV-10
SUGGESTED CONFIGURATION FOR WATER AND NUTRIENT BALANCE TABLE
SOURCES
- UPSTREAM
- TRIBUTARIES
- IRRIGATION
RETURNS
- MUNICIPAL
- INDUSTRIAL
TOTAL
LOSSES
- DOWNSTREAM
- DIVERSIONS
i
TOTAL
SOURCES-LOSSES nnn
X JLUU
LOSSES
FLOW RATE
LOADING RATE
TN %
TP %
351
-------
EXAMPLE IV-3
Figure IV-5 shows a hypothetical river which has three tributaries, a
nonpoint source of runoff, and two diversions. Develop a water balance for
this system. The known flow rates are
Identification Number Flow rate (cfs)
1 2000
2 4000
3 1200
4 200
5 800
6 1000
7 2000
8 6000
The flowrates at locations 1,2,3, and 5 are assumed to comprise the inflow
rates to the system. The total inflows are:
Identification Number Inflows
1 2000 cfs
2 4000 cfs
3 1200 cfs
5 800 cfs
Total 8000 cfs
The inflow from gage 4 is not needed because gage 5 is located further
downstream on the same tributary. The outflows consist of diversions 6 and
7 and the downstream outflow past gage 8:
352
-------
FIGURE IV-5 ILLUSTRATION OF WATER BALANCE
353
-------
Identification Number Outflows
6 1000 cfs
7 2000 cfs
8 6000 cfs
9000 cfs
The inflows and outflows differ by 1000 cfs. There are several reasons
for the imbalance. One, the flow rate past each gage is not measured
perfectly, but differs by some degree from the actual flow rate. Two, the
gage at location 5 does not catch all of the nonpoint source runoff, so
there is an additional inflow to the system which has not be quantified.
Three, depending on the size of the reservoir, direct precipitation and
-evaporation might be significant.
END OF EXAMPLE IV-3
The following example illustrates both a water and nutrient balance,
and is based on work performed by Tetra Tech on the Snake River in Idaho
(Mills and Desvoigne, 1978).
EXAMPLE IV-4
Develop annual water and phosphorus balances for water year 1976 for
the Snake River from Heise, Idaho, to below American Falls Reservoir, a
distance of 150 miles. A sketch is shown in Figure IV-6. Estimate the
phosphorus retention coefficient for American Falls Reservoir. The
retention coefficient is defined as:
R ^ Flux Input - Flux Output
p Flux Input
The required data are shown below.
• Surface area of American Falls Reservoir = 56,600 acres
• Evaporation rate in this part of United States = 33
inches/year
354
-------
Snake River
near Heise
RM861
Henry's Fork
near Rexburg
Blackfoot River
near Blackfoot
(include bypass canal
Portneuf River at Pocatello
American Falls Reservoir
Snake River at Neeley
RM713
FIGURE IV-6
SKETCH OF SNAKE RIVER FROM HEISE TO NEELEY,
IDAHO,
355
-------
• Precipitation = 11 inches/year
• Ground water inflow into Snake River: 500 cfs
• Ground water inflow into American Falls Reservoir: 2,100 cfs.
The total phosphorus concentrations were generated during the study of Mills
and Desvoigne (1978) and are provided here:
Source mg/1
In rainwater 0.03
Snake River near Heise 0.05
Henrys Fork 0.11
Blackfoot River 0.26
Portneuf River 0.68
Groundwater inflow 0.23
Snake River near Neeley 0.08
The surface inflow rates are gaged by the U.S. Geological Survey and
are reported in the U.S. Geological Survey Water Data Report for Idaho
(1976). An example of how the information is tabulated in these reports is
shown in Figure IV-7. From an entry in the Table, the mean flow rate for
water year 1976 is 8549 cfs at USGS 30307500, near Heise. Rather than
showing the remaining tabulations from the USGS report the flow rates from
water year 1976 will simply be tabulated, as contained in the report.
Source Flow rate
Blackfoot River 453 cfs
Henrys Fork 3,235 cfs
Portneuf River 412 cfs
USGS 13059500 (diversion) 2,333,700 ac-ft/yr
USGS 13069000 (diversion) 800,900 ac-ft/yr
Based on this information the water and total phosphorus balances are
calculated and shown in Table IV-11. The flow rates are all converted to
units of cfs. This requires converting the precipitation, evaporation, and
diversions to these units. A precipitation rate of 11 inches per year is
equivalent to 71 cfs:
356
-------
LOCATION. --Lat 43°36'45", long 111°39'33", in SE
-------
TABLE IV-11
SOLUTION TO SNAKE RIVER WATER AND PHOSPHORUS BALANCE PROBLEM
Sources
Snake River at Heise
Blackfoot River
Henrys Fork
Portneuf River
Ground water inflow into
Snake River
Ground water inflow into
American Falls Reservoir
Precipitation on American
Falls Reservoir
Losses
USGS 13059500
USGS 13069000
Snake River at Neeley
Evaporation
2-f Losses
Flow
Rate (cfs)
8,549
453
3,235
412
500
2,100
71
15,320
Flow
Rate (cfs)
3,214.
1,103
11,360.
215.
15,892.
TP Loading (Ibs/day)
2,300.
1,915.
634.
1,510.
619.
2,600.
11.
9,589.
TP Loading (Ibs/dav)
865
415
4,890
_
6,170
r r. /Losses-Sources\ inn _ ,al
Losses
358
-------
11 f 12 x 56600 x 43560 f 366 :- 24 f 3600 - 71 cfs
The diverted flow in ac-ft/yr is converted to cfs as shown:
USGS 13059500: 2333700 x 43560 - 366 * 24 T 3600 = 3214
The percent difference between inflow rates and outflow rates is 4 percent.
Based on these flow rates, and the concentrations of total phosphorus
presented earlier, the sources and losses of total phosphorus can be
tabulated. For example, the mass flux of total phosphorus flowing past
Heise can be calculated using Equation IV-10:
M = 5.38 x 8549 x 0.05 = 2300 Ibs/day
Continuing in this manner, the sources and losses are as tabulated in Table
IV-11. The large imbalance is caused by retention at American Falls
Reservoir. The phosphorus loading to the reservoir is:
9589 - 865 - 415 = 8309 Ibs/day
Since the phosphorus leaving the reservoir is 4890 Ibs/day, the retention
coefficient is:
890
8309
American Falls Reservoir retains a significant quantity of the phosphorus
which enters the reservoir and consequently tends to keep phosphorus levels
in the Snake River below the dam depressed compared with what they would
otherwise be.
END OF EXAMPLE IV-4-
359
-------
4.1.11 Hand Held Calculator Programs
It has become apparent that, after applying the river screening
techniques contained in the original manual (Zison e_t a]_., 1977) to real
systems, a substantial savings of both time and effort could be realized by
programming the major computational sequences. To this end, these
algorithms have been programmed on the Texas Instrument TI-59 calculator and
are available upon request in a document prepared by Tetra Tech
(Mills erb _al_., 1979)*. To date the algorithms contained in Mills eit a 1.
(1979) predict:
• equilibrium temperature
• longitudinal instream temperature distribution
0 mixing temperatures
• BOD profiles for point and nonpoint sources
• reaeration rates
• dissolved oxygen profiles
• waste assimilative capacity and critical dissolved oxygen
levels
• coliform profiles for point and nonpoint sources
• bed material sediment transport
For each program contained in the document the following information is
provided for the user:
t a detailed set of user instructions,
0 a program listing, and
0 a sample input/output sequence.
An example set of users instruction is shown in Figure IV-8. The first 6
steps are for data entry and the seventh is for calculation of the required
information.
* Attention: W.B. Mills
Tetra Tech, Inc.
3746 Mt. Diablo Blvd., Suite 300
Lafayette, California 94549
360
-------
IICLE
PI-.OGHAf.1MER DATE
Partitioning (Op 17; JL6_Q.J> .Oj Library Module
FAGe_J OF J
.Printer .OptlOnaLClrcsI
PROGRAM DESCRIPTION
Program: Critical Dissolved Oxygen Calculations
This program calculates the critical dissolved oxygen deficit downstream from a
point source of pollution. It also calculates the travel time to the critical
deficit.
Note that if the travel time turns out to be negative, then the critical deficit
occurs at the point where pollution enters the stream.
SSTEPI
USER INSTRUCTIONS
PROCEDURE
I 1 | Enter program - see listing following
these instructions
Enter reaeration rate at 20°C, k
(I/day)
'a 20
Enter deoxygenation rate at 20°C, k
(I/day)
'd20
Enter BOD concentration in stream just
below source of pollution, L (mg/S.)
Enter dissolved oxygen deficit in stream
just below source of pollution, D (mg/1)
Enter stream temperature, T (°C)
7 i Calculate and display:
- reaeration rate at stream temperature,
ka (I/day)
- deoxygenation rate at stream
temperature, k. (I/day)
- travel time to critical deficit,
tc (days)
- critical deficit, D (mg/S.)
program j D
ka20
kd20
Lo
Do
T
I
R/S
R/S
R/S
R/S
R/S
R/S
R/S
R/S
R/S
\
1
1
1
FIGURE IV-8 EXAMPLE SET OF USER'S INSTRUCTIONS
FOR HAND HELD CALCULATOR PROGRAMS
361
-------
4.2 CARBONACEOUS AND NITROGENOUS OXYGEN DEMAND
4.2.1 Introduction
Many wastes discharged into waterways contain biologically oxidizable
materials that exert an oxygen demand on waterway resources. This
biochemical oxygen demand (BOD) can be subdivided into carbonaceous (CBOD)
and nitrogenous (NBOD) components. Table IV-12 illustrates typical
concentrations of NBOD and CBOD in untreated municipal waste.
CBOD represents the amount of oxygen required by bacteria to stabilize
organic matter under aerobic conditions. The reaction can be approximated
by
C.H.OtNc + n + - - c 0, -* nCO, + ? - c\ H,O + cNH,
This reaction assumes that the available organic matter is completely
oxidized. Bacteria, however, might not be able to completely oxidize all of
the available organic matter. Equation IV-13 does illustrate that oxidation
of the nitrogen is not included as part of CBOD. The reduced nitrogen is
oxidized to nitrate in a two step process as follows:
2NH, + 30, ^l!^ 2NOl- + 2H+ + 2H,0 ( IV-13)
b&eleru
nitral^-forminc / T., •% n \
2NOz~ + O, + 2H+ -- + 2NOr + 2H+ ( IV-14)
bacteria
Based on Equations IV-13 and IV-14 the NBOD is
NBOD = 4.571 lOrg-NJ + [NH^ - N| I + 1.14 |NO~ - N| (IV-15)
Typically the nitrite concentration is negligible so that
NBOD = 4.57 (TON) (IV-16)
where TON represents total oxidizable nitrogen, the sum of organic and
ammonia nitrogen. A typical value of TON from Table IV-12 is 20+28 = 48
mg-N/1, which corresponds to an NBOD of 220 mg/1.
362
-------
TABLE IV-12
MUNICIPAL WASTE CHARACTERISTICS
BEFORE TREATMENT (THOMANN, 1972)
Variable
Average Daily Flow
Solids
Total
Total Volatile
Total Dissolved
Total Suspended
Volatile Suspended
Settleable
BOD
Carbonaceous (5 day)
Carbonaceous (ultimate)
Nitrogenous*
Ni trogen
Total
Organic
Ammonia
Nitrite + Nitrate
Phosphate
Total
Ortho
Poly
Col i forms
Total mi
Fecal mi
Unit
gal/cap/day
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1 N
mg/1 N
mg/1 N
mg/1 N
mg/1 PO^
mg/1 PO^
mg/1 PO
11 ion org./lOO ml
11 ion org./lOO ml
Approx.
Averaqe
125
800
400
500
300
130
150
180
220
220
50
20
28
2
20
10
10
30
4
Normal
Range
100-200
450-1200
250-800
300-800
100-400
80-200
-
100-450
120-580
-
15-100
5-35
10-60
0-6
10-50
5-25
5-25
2-50
0.3-17
*Ultimate, Nitrogenous oxygen demand, exclusive of CBOD.
363
-------
Typically in the bottle determination of CBOD and NBOD, the
carbonaceous demand precedes the nitrogenous demand by 5 to 10 days, as
shown in Figure IV-9. This had led workers to believe that nitrification
can be ignored in river environments below a source of pollution up to a
distance corresponding to a travel time of five to ten days. Such an
assumption might be invalid for several reasons. Given that there are
numerous sources of pollution along many rivers a viable population of
nitrifying bacteria may already be present within the water column. Second,
nitrifers can grow attached to the bottom substrate. Consequently,
significant numbers can exist just below the discharge location and
nitrification can proceed immediately. Nitrification by attached bacteria
is more likely to be of significance in relatively shallow, wide rivers,
which have stable bottom substrate (Mills, 1976).
urve for combined demand
(Carbonaceous plus nitrification)
Curve for carbonaceous demand at 20°C
FIGURE IV-9 THE BOD CURVE, (A) CURVE FOR OXIDATION OF
CARBONACEOUS MATTER, (B) CURVE SHOWING
INFLUENCE OF NITRIFICATION,
CBOD is a commonly measured characteristic of waste water. The CBOD
used in the formulations presented below is the ultimate CBOD. Often CBOD
is expressed as CBOD5, the oxygen utilized in a 5 day test. The
relationship between ultimate (CBODL) and 5-day CBOD can be approximated by:
CBODL -
CBOD
OT5ET
364
-------
This relationship assumes a decay rate of 0.23/day, and may be different for
effluents from advanced wastewater treatment plants.
The mass balance equation used in the CBOD analysis is exactly
analogous to the NBOD equation. The first order decay rate assumption for
NBOD stabilization is necessary to maintain this analogy, and is sufficient
for hand calculations.
Nitrification (the process by which ammonia is oxidized to nitrite, and
nitrite to nitrate) is pH dependent with an optimum range of 8.0 to 8.5
(Wild, 1971). If the pH of the river is below 7.0, nitrification is not
likely to be important.
4.2.2 BOD Decay Rate
The decay rate for CBOD will be denoted by kL and for NBOD by k^.
Typical values of both kL and kN lie between 0.1 and 0.6/day, with 0.3/day
being typical, k. values can, however, exceed the range given here. Values
of 1 to 3/day have been computed for shallow streams (Thomann, 1972). A
figure to be presented shortly will show how k, depends on depth. The
following discussion will be directed toward kL, but in general will also
apply to kN.
The disappearance of BOD from a river is a reflection of both settling
and biochemical oxidation, as shown in Figure IV-10. Biochemical oxidation
can consist of instream oxidation (kjL) as well as absorption by attached
organisms (k4L). The total oxidation rate then, is k , where
k, = kx + k4
d
The total loss rate k is
"L-V1"
where k3 reflects settling losses.
365
-------
FLOW
Sedimentation
Ax\
K4L
ww
<
r
KiL- Instream
deoxygenation
L- ULTIMATE B.O.D.
Absorption by
attached organisms
FIGURE IV-10
MECHANISMS OF BOD REMOVAL
FROM RIVERS
Settling of BOD is generally more prevalent just below a sewage discharge
where the discharged material may contain a large suspended fraction. As
this material is transported downstream the settling component becomes less
important and the reaction rate k approaches the oxidation rate k ,. In
this chapter, the settling component will not be explicitly considered.
Neglecting settling will tend to cause estimated instream BOD levels to be
somewhat higher than they actually might be along certain portions of a
river. It should be noted that if instream BOD data are used to determine
k (one such method will be explained in Figure IV-12) then the effect of
settling is automatically included in k .
Figure IV-11 illustrates the dependence of k. on river depth. The highest
deoxygenation rates occur in shallow streams with stable, rocky beds,
reflecting the significance of attached biological organisms. Appendix C
contains observed and predicted values of k. for various natural streams.
The decay coefficients k and kN are both temperature dependent and this
dependence can be estimated by:
= k20 1.047
(T-20)
(IV-17)
366
-------
10.0
1.0
o
o
o
CVJ
O
0.05
Stable, Rocky Bed
Moderate Treatment
Some Ammonia
MEAN
Unstable, Sandy Channel
Highly Treated Effluent
with Nitrification
10.0
100.0
DEPTH (FT.)
FIGURE IV-11 DEOXYGENATION COEFFICIENT AS A FUNCTION OF
DEPTH, (AFTER HYDROSCIENCE, 1971)
367
-------
where
kon = k, at 20°C
20
k = k at T°C
T = water temperature, C
Numerous methods for computing k from observed data are available
(Nemerow, 1974). One method entails the use of a semi-log plot. The
stretch of river containing the data to be plotted must have a constant
stream area and flow rate, and the BOD loading must be from a point source
located at a position that will be called x = 0. Plotting the log of BOD
concentration versus distance generally produces a straight line with slope
of -k /U. An example is shown in Figure IV-12. Either CBOD5 or CBOD. can
be plotted as the ordinate. The slope should be converted from base 10
logarithms as given in the semi-log plot to base e logarithms as needed in
the formulations used in this chapter. The conversion is made by
multiplying the value for log base 10 by 2.303.
Wright and McDonnell (1979) have more recently developed an expression for
instream BOD decay rate based on the flow rate of the river. The expression
is:
i klab if Q>800 cfs (IV-18a)
kj '^YJ =
10'3 if Q<800 cfs (IV-18b)
This expression is particularly attractive because the only hydraulic
variable required is flow rate. Other predictive techniques and rate data
from rivers around the country are contained in Zison et al. (1978).
368
-------
10.0
-Slope x U
0.017* ,4Miles
Miles' * Day
0.16/Day
o
| DISTANCE (MILES)
INPUT
FIGURE IV-12 EXAMPLE OF COMPUTATION OF KJ_ FROM STREAM
DATA (FROM HYDROSCIENCE, 1971)
369
-------
4.2.3 Mass Balance of BOD
The general mass-balance equation for BOD in rivers is
If (QL) - kL L + Lr
-------
For any particular reach of a river under investigation the stream
cross-sectional area can be expressed by:
/A, -A \
A = AQ + x = AQ + AAx (iv-21)
where
A XL
A = stream cross-sectional area at upstream end of the reach
Af = stream cross-sectional area at downstrean end of reach
x = distance downstream from beginning of reach
x. = length of reach
The cross-sectional area need not be measured directly, but can be
computed from:
The cross-sectional area change can reflect a change in
stream velocity, perhaps due to a bed slope increase or decrease.
The length of the reach under investigation, x, , is measured in river
miles along the river's centerline. If use of a constant stream area
is assumed, then Aft = 0 and A = AQ throughout the reach.
4.2.4 Typical Solutions
Ca_s_e_]_: The only source of CBOD occurs as a point source at
x = 0. The CBOD distribution is then expressed by:
(IV-22)
where .
i - A
L " Uo
371
-------
U = stream velocity at x = 0
o
L = ultimate BOD at the upstream end of the reach
o
L = ultimate BOD at a distance x downstream
The other terms have previously beer, defined. The initial CBOD, L ,
must reflect both CBOD upstream of the reach as well as that contributed
by the point source in question. It is given by:
L.O + W/5.38
L = (IV-23)
where
W = mass rate of discharge of CBOD, Ib/day
Q = upstream river flow, cfs
Qw = waste flow rate, cfs
L = upstream CBOD concentration, mg/1
Ca_se_2: For a point source of CBOD at x = 0 and a distributed
mass influx of CBOD (with no associated flow) entering the river
throughout the reach, the solution is
(IV-24)
where
L , = mass rate of CBOD entering the reach per unit
rd
volume of river water, mg/1/day
Case 3: A distributed flow enters the river carrying CBOD
and a point source of CBOD exists at x - 0. The flow rate
Q at a distance x is:
Qf - Qo
Q = Q + — x = Q + Anx
M ^o x, o Q
372
-------
where
The BOD distribution is given by (the river cross-sectional area
is assumed constant throughout the reach):
where k, A + An
r - L 0 U
El • —^-
L = concentration of CBOD entering the river in the distributed
flow, mg/1
Case 3 can also be usea TO esTaolish the effect a purely
diluting inflow (i.e. L =0) would have on the CBOD distribution.
Case 4: For a point source at x = 0, a distributed source
with associated inflow, and a mass flux with no associated
flow (constant river cross-sectional area), the solution is
where
k A +
E = _^-- --- . , as in Case 3.
I AQ
4.2.5 Othe r__S i_mp1 i fyijig Procedures
The formulations represented by Equations IV- 22 through IV-26 offer
a range of options for examining BOD distribution in rivers.
However, there are additional methods of estimating instream concen-
trations and determining whether or not significant BOD levels exist.
373
-------
Perhaps the simplest method is assuming that BOD does not decay. An
upper limit of the instream concentration at any point can then be
determined by incorporating all known sources, and using the methods
presented in Section 4.7. If the computed instream concentrations are
below a threshold pollution level, then there is no need to apply
Equations IV-22 through IV-26 because the inclusion of a decay rate will
only lower the concentrations.
It may also be feasible, as a first estimate, to combine the CBOD
and NBOD equations into one, and use that equation to estimate the
distribution of the total oxygen-demanding material. To do this, all
source terms must include both CBOD and NBOD. One decay coeffi-
cient is used for both CBOD and NBOD decay. The larger decay coeffi-
cient of the two should be used since that will produce the larger
oxygen deficit.
In deciding which of Equations IV-22 through IV-26 to use for any
analysis, the purpose of the analysis as well as data availability
should be considered. If the main purpose is to estimate differences
in stream concentrations caused by various levels of abatement at a
sewage treatment plant, the diffuse sources of BOD need not be
considered. The resulting concentration difference can be expressed
as:
AL =
AL exp
o p
AL
•-J
L
(V +
A
(!V-27a)
(IV-27b)
where 'L is the change in BOD concentration due to a change, f-L , in
the initial concentration. Equation IV- 27A should be used for a Case 1
or Case 2 situation, and Equation IV-27B for Case 3 or Case 4. If an
estimate of the absolute level of BOD is desired, however, then the
appropriate expression including the nonpoint sources should be
utilized. It should be noted that if the diffuse sources of BOD are
374
-------
large then the improvement of instream BOD concentrations by point
source control will be relatively minor. In that case the planner
should focus on nonpoint source control.
EXAMPLE IV-5
Mixing
Zone
Ejrti mati ng BOD Distribution in a _Rvve_r
Suppose the user wants to calculate the BOD distribution in the river
shown below in Figure IV-13. There are nine point sources contributing
y0=i.ifPs
BOD= mg/l 1
Q = 300cfs 1
1 t
\
1
™\i
I ^S
1 /^
I vi>
n
/ 75 Ml. \
t
K
/2\
(4)
t
/fr
&
®
1
50 Ml.
\ Q = 200cfs
\BOD = lmg/l
tti
©®i
i/
/
r Mixing
Zone
i
'y >
» A\ /
FIGURE IV-13 HYPOTHETICAL BOD WASTE LOADINGS IN A RIVER
BOD in the stretch of river under consideration. The ninth source is
assumed to be a tributary, and contributes substantially more flow than
the other eight. Begin hy dividing the river into reaches. The
first reach (I) should include the first 75 miles in which there is
one point source of BOD at the upstream end (source (1)). Equation IV-22
375
-------
is applicable to that reach. Now, there are several choices available
regarding the division of the river between sources (2) and (8). One
choice is to divide the 50 miles into mini-reaches similar to reach I,
and reapply Equation IV-22 seven more times. A second alternative is
to group adjacent point sources into fewer and larger sources, thereby
requiring fewer applications of Equation IV-22. A third alternative is
to assume that sources (2) through (8) comprise one continuous distrib-
uted source, the total pollutant loading of this equivalent source being
equal to the sum of the individual loads. For this representation to be
valid the sources should be both evenly spread spatially and be dis-
charging comparable loads. The third alternative will be examined here,
and reach II will consist of the 50 miles following reach I. Equation
IV-25 will be used to analyze reach II. Reach III then, will begin just
downstream from the tributary (source (9)).
For reach I, Equation IV-22 is first solved. Suppose the follow-
ing characteristics of waste source (1) are known:
Q = 20 MGD = 1.55 (20) cfs = 31 cfs
W = 5000 Ib. BOD5/day
Recall that
0
W must be in Ib. BOD ultimate/day:
-5-QPl = 7353 ib. BOD, /day
. bo L
376
-------
Then
(1) (300) + 7353/5.38
300 + 31
= 5.0 mg/1
The decay coefficient is estimated from Figure IV-H as 0.4/day. No
correction will be made for temperature. Equation IV-22 can
now be expressed as (for constant cross-sectional area):
L = 5
where x is the downstream distance in feet. Note the correction
needed to convert the decay coefficient from units of I/day to I/sec
The results of the above equation for selected distances down-
stream can be expressed as follows:
x (miles)
0
30
60
75
L (mg/1)
5.0
2.6
1.3
0.9
For reach II, sources (2) through (8) are assumed to contribute
the followinc loading.
BOD = 8000 Ib/day
Q = 120 MGD = 186 cfs
The flow distribution, Q, in reach II is then:
0 X.
186
377
-------
where x i.s in miles (from 0 to 50). L , the average BOD. concentration
in the incoming flow is:
. = 8COO Ib/day _l_ms/l _ 8 Q ,,
Lr 120 MGD X 8.34 Ib/day B'U mg/'
If the average depth in reach II is assumed to be 5 feet, then:
kL = .3/day
Finally, E, is computed:
r - LQ Q A - o . 331
El ~ A ' Ao " UQ ~ 1.1
(0.3)(301)
= 2 5
186
(50)(5280)
Then, using L from the 75 mile point of Reach I as LQ:
-3,-
378
-------
In tabulated form:
x (mi )
0
20
40
50
Q (cfs)
331
405
480
517
L (mg/1)
0.9
1.8
2.3
2.5
Note that the BOD concentration is increasing within this reach.
For reach III, only enough information is given to compute the
initial concentration, utilizing weighted values for the ROD at the end
of reach II and that entering through the tributary (source (9)).
= 200JJ) + 517J2.JQ. =
o 200 +517
„
mg/l
END OF EXAMPLE IV-5
4.2.6 Interpretation of Results
The most frequent use of BOD data in river water quality
analyses involves their relationship with the dissolved oxygen
balance. This relationship will be discussed more fully in Section
4.3. At this point it is sufficient to say that it is necessary to
predict the BOD distribution in a river in order to compute dissolved
oxygen concentrations.
When a river receives a heavy load of organic matter, the normal
processes of self purification result in a series of zones of decreas-
ingly severe conditions succeeding one another downstream. Each zone
contains characteristic animals and plants (Neir.erow, 1974). A sapro-
bicity system (saprobicity is a measure of biodegradable organic
matter) has been developed that relates BOD concentrations in streams
379
-------
to the degree of pollution there. Correlations have been found, for
example, among BOD concentrations, coliform bacteria, and dissolved
oxygen in rivers (Sladecek, 1965). Sladecek (1969) has assigned 5-day
BOD values of 5 mg/1 to mildly polluted conditions and 10 mg/1 to sub-
stantial pollution.
Sources of drinking water are subject to restraints on the maximum
allowable BOD that can be contained in raw water and still qualify as
a drinking water source. Further, the degree of treatment of the raw
water is dependent on the concentrations of certain constituents, such
as BOD. One reference (NEC, 1975) has stated that water having a 5-day
BOD over 4 mg/1, in combination with high levels of other constituents,
represents a poor source of domestic water supply.
As discussed above, BOD in a river can come from a number of
sources, both point and nonpoint. Although BOD reduction from point
source might be easier to accomplish than from nonpoint sources,
there is no guarantee that BOD levels will be substantially lowered.
4.3 DISSOLVED OXYGEN
4.3.1. In t roduc_ti_on_
Historically dissolved oxygen has been and continues to be the
single most frequently used indicator of v-ater quality in streams and
rivers. Figure IV-14 shows the seasonal variability of dissolved
oxygen in 22 major waterways throughout the country (EPA, 1974) from
1968 to 1972. Invariably the levels observed from June to October
are lower than those observed in January to March. This is due
primarily to the influence of temperature on the dissolved oxygen
levels. Due to the effect of temperature, surrmer is the most critical
season in terns of organic pollutant assimilation in rivers.
380
-------
300 SEASONAL
Greater
Than
Reach
Hudson
De 1 aware
Susquehanna
Potonac
AlabaiTia
Upper Ohio
diddle Ohio
Lower Ohio
Upper Tennessee
Lower Tennessee
Upper Missouri
Middle Missouri
Lower ilissoun
Upper Mississippi
Mississippi near Minneapolis
Middle Mississippi
Lower Mississippi
Upper Arkansas
Lower Arkansas
Upper Red
Lower Ked
Brazos
Rio Grande
Upper Colorado
Lower Colorado
Sacramento
Col umbi a
Snake
Wi 1 1 amette
Yukon
Boston Harbor
Chicago Area-Tributaries
Chicago Area-Lake ,-iichigan
Detroit Area-Tributaries
Detroit Area-Rivers
Nuofer 0.00 1.75 3.50 5.25 7.00 8.75 10.50 12.50 14. (
Stations
19
17
21
Ib
10
8
59
29
6
24
?3
9
4
3
17
12
4
4
4
3
5
4
4
2
5
4
17
12
12
7
8
9
14
11
5
6
6
7
4
1
6
5
7
b
14
9
8
!
21
11
18
15
?8
23
7
16
65
11
9
7
7
b6
1 i 1
•JLf
1*
tie
•1*
%L
*T
tit
1*
_.,, %L
•P
.,£-.. Mean 15th
KEY: ^^^ Percent! le
x^ Mean
Jfin-Mnr -.- . - ^TB ... . .... ...
Mean 85th.^-'- —
Percentile
.
^r
ale
•p
1 I I
T T I
tic
1*
«
--*
•
"•
•--
*
A. i
^r
••
>b
.
"~^~~
*
•
•-
,
"*
III
FIGURE IV-14
VARIABILITY OF DISSOLVED OXYGEN BY SEASON FOR
22 MAJOR WATERWAYS, 1968-72 (EPA, 1974)
381
-------
The dissolved oxygen calculations presented below range in
complexity from a simple CBOD-DO relationship to a more general dis-
solved oxygen mass balance including CBOD, NBOD, photosynthesis,
respiration, and benthic demands. It should be stressed, however,
that the results calculated from any of the relationships provide
estimates only since each procedure incorporates various assumptions
that might not be fully met. For example, waste loading inflows are
assumed to remain constant in quality and quantity over time. In
reality loadings probably vary over time. Furthermore the choice
of system parameters involves a certain degree of judgment. However,
for any given situation, the planner can establish an envelope of
possible outcomes by different realistic choices of system parameters,
4.3.2 Dissolved Oxygen Mass-B_alance_
The general dissolved oxygen mass-balance equation that will be
utilized here is given by:
-kl-kN + k (C C] S +PR (IV-28)
5F = = - A 3x LL V a ( s Lj bb P R V '
where the new symbols introduced are:
C = dissolved oxygen concentration, mg/1
k = reaeration coefficient, I/day
a
C = saturation value cf dissolved oxygen, mg/1
Sb = benthic oxygen demand, mg/1/day
P = rate of oxygen production due to photosynthesis, nig/I/day
R = rate of oxygen consumption due to algal respiration,
mg/l/day
Stated in words, Equation IV-24 expresses the following relationship:
382
-------
At steady state, the rate of addition of dissolved oxygen to a river
due to reaeration and photosynthesis equals the depletion rate caused
by the net advective flow, carbonaceous oxidation, nitrogenous oxidation,
benthic demands, and algal respiration.
Commonly, the dissolved oxygen mass-balance equation is expressed in
terms of the deficit, D, which is the difference between the saturation
and actual concentrations.
4.3.3 Reaeration Rate
The atmosphere acts as the major source for replenishing the
dissolved oxygen resources of rivers. Reaeration tends to equili-
brate the dissolved oxygen concentration in a river with its saturation
value. Most commonly, the dissolved oxygen concentration is below
saturation and there is a net influx of oxygen into the river from the
atmosphere. On occasion, due to the production of dissolved oxygen by
algae, rivers or streams can become supersaturated, in which case there
is a net loss of oxygen to the atmosphere.
A number of expressions for the reaeration coefficient, k , have
a
been developed. Several are presented here. O'Connor's formulation
(Thomann, 1972) states that:
k = (DL U) at 20°C (IV-29)
H3/2
where
D, = oxygen diffusivity = 0.000081 ft2/hr at 20°C
H = stream depth in ft
U = stream velocity in ft/sec
383
-------
Expressed in English units,
,V2
k = 12.9 U "• at 20UC (IV-30)
— H3/2
The above formula was verified on streams and rivers ranging in average
depth from 1 foot to 30 feet with velocities ranging from 0.5 to 1.6 fps.
Its use should be limited to streams where the reaeration coefficient
is less than 12/day. Figure IV-15 illustrates how kfl changes with depth
and velocity according to this relationship.
For shallow (0.4 - 2.4 feet), fast moving streams the following
expression developed by Owens (Thomann, 1972) is preferable, as the
experimental work to develop this expression was done almost exclusively on
shallow streams:
k - 21.6 U°'67 at 20°C
a —1V85 (IV-31)
n
where U is in ft/sec and H in feet. A graphical representation of Equation
IV-31 is shown in Figure IV-16.
Covar (1976) showed that there were certain combinations of river
depths and velocities where a formula developed by Churchill
(Churchill et al_., 1962) is more accurate than either the O'Connor or Owens
formulations. The Churchill expression is:
ka = 11.6U0'969 H'1-673 per day at 20°C (IV-32)
The regions of validity, and the predicted values, for the three
formulations are shown in Figure IV-17.
Recent studies have suggested that the Owens expression overestimates
the reaeration rate for particularly shallow streams (e.g., less than a foot
in depth). Under these circumstances the Tsivoglou-Wallace method
(Tsivoglou and Wallace, 1978) is more accurate. The expression is:
384
-------
10.0
o
o
o
O
CO
!5
O
Ld
O
UJ
o
o
o
01
UJ
<
LU
cr
0.10
0.01
0.3
DEPTH (FT.)
1.0
Rapid Turbulent
1.0-2.0 FPS
Moderate
0.5-1.0 FPS
Slow Stagnant
"0.1-0.5 FPS
J_L
10.0
100.0
FIGURE IV- 15
REAERATION COEFFICIENT AS A FUNCTION OF DEPTH
(FROM HYDROSCIENCE, 1971)
385
-------
40
10
o
o
O
(M
1
V-Stream Velocity (ft/sec)
j i i i i i i
V=I.O
V»0.5
V=0.l
O.I
DEPTH (FT.)
4.0
FIGURE IV-16 R.EAERATION COEFFICIENT FOR SHALLOW STREAMS,
OWEN'S FORMULATION
386
-------
G)
3e«/l
'It.)
-~f.
o
s
s
\
c
TS
i
u
^.
s
•^
c
i
\
A.
1
D
h
^
p
o>
-<,
K
o
-------
ka (I/day)
a
-0
7776. US, @ 25°C, Q < 10 cfs (IV-33a)
4665.6 US, @ 25°C, 10 < 0 < 3000 cfs (IV-33h)
2592. US, & 25°C, Q > 3000 cfs (IV-^c.
where
S = stream slope, ft/ft
Table IV-13 compares predictions of Tsivoglou-Wallace with observed values
for several small streams in Wisconsin. The agreement is good.
EXAMPLE IV-6
Prediction of Reaeration Rates
In September, 1969, a study was conducted to determine the reaeration
rate coefficients on the Patuxent River in Maryland during the low flow
period. The study was carried out on a seven mile stretch of the river
below Laurel, Maryland. The stream was divided into seven segments, and the
reaeration rate determined for each segment. A portion of the results are
shown in the Table IV-14. Using the hydraulic data in the table predict the
reaeration rates using the methods of Tsivoglou-Wallace and of Covar.
Since the method of calculating the reaeration for each reach is the
same, an example calculation will be shown for the first reach only. Based
on a velocity of 0.39 ft/sec and a slope of 0.0013 ft/ft, the
Tsivoglou-Wallace method predicts a reaeration rate of
k = 7776 x 0.39 x 0.0013 - 3.9/day at 25°C
a
Equation IV-33a is used since Q<10 cfs.
Using Figure IV-17 and a river depth of 0.8 feet reveals that the Owens
formula is applicable. Applying Equation IV-31 shows that
n ^qo .ev
k = 21.6 ^^ = 17.4/day at 20°C
a 0.81'85
388
-------
TABLE IV-13
COMPARISON OF PREDICTED AND OBSERVED
REAERATION RATES ON SMALL STREAMS IN WISCONSIN*
Stream
Black Earth
Creek
Mud Creek
tributary
Dodge Branch
Isabel! e Creek
Madison effluent
channel
Mill Creek
Honey Creek
West Branch
Sugar River
Koshkonong Creek
Badger Mill
Creek
Observed k
(I/day at 25°C)
8.46
10.7
33.1
14.
2.06
3.31
18.4
42.5
6.09
7.98
Predicted
k Using
Tsivoglou's
Method
(1/da.y at 25°C)
7.8
4.2
34.6
-
4.1
2.2
27.4
36.4
4.8
9.1
*Grant, R.S., 1976. Reaeration-Coefficient Measurements of 10 Small
Streams in Wisconsin Using Radioactive Tracers... with a Section on
the Energy-Dissipation Model. U.S. Geological Survey. Water Resources
Investigations, 76-96.
389
-------
TABLE IV-14
TYPICAL HYDRAULIC PROPERTIES
PATUXENT RIVER (SEPTEMBER, 1969)
Reach
1-2
2-3
3-4
4-5
5-6
6-7
Flow
cfs
9.8
9.8
9.8
19.5
19.5
19.5
Length
ft
5,400
4,200
7,200
8,400
6,600
4,800
Velocity
ft/sec
0.39
0.22
0.35
0.35
0.25
0.37
Depth
ft
0.80
1.00
1.00
1.10
1.10
1.00
Slope
ft/ft
.0013
.0011
.0014
.0018
.0013
.0013
Reaerution Rate (I/day)
Observed
(25°C)
3.9
2.7
3.3
3.5
2.4
4.8
Tsi vogl ou-VJal lace
(25°C)
Covd r
(20JC)
GO
UD
O
-------
The results for all the reaches are tabulated below.
Reaeration Rate (I/day)
Reach
1-2
2-3
3-4
4-5
5-6
6-7
The predictions using the Tsivoglou-Wallace method are good for all reaches,
while the method of Owens predict values two to three times too large, and
provides evidence that Owens method probably should not be applied to
extremely shallow rivers.
Observed
(25°C)
3.9
2.7
3.3
3.5
2.4
4.8
Tsivoglou -Wai lace
(25°C)
3.9
1.9
3.8
2.9
1.5
2.2
END OF EXAMPLE IV-6
Temperature changes affect the reaeration rate, and the relationship
can be approximated by:
(kJ = (k ) 1.024(T-20) (IV-34)
3 T a 20
where
(k ) is the reaeration coefficient at T °C.
a T
In addition to temperature, substantial suspended sediment concentrations
can appreciably alter the reaeration rate in streams (Alonso ejt _al_., 1975).
As an approximation, k decreases by 9 percent per 1,000 ppm increase in
3
suspended sediment up to a 4,000 ppm load. Beyond that, concentration data
are not available to assess the response of k . It is suggested that a 40
a
percent decrease be used for higher suspended sediment loads. Rivers with
high suspended sediment loads are generally found in the western central
391
-------
states. Measured values of k for various streams and rivers are included
a
in Appendix C.
4.3.4 Effect of Dams on Reaeration
Many rivers or streams have small to moderate sized dams crossing them
in one or more places. Reaeration occurs as the water flows over the dam.
Based on experimental data (Gameson eit aJL , 1958), and later verified with
field data (Barrett et al_ 1960), the following relationship for reaeration
over dams has been developed:
b(1 *0.046T)H]°
a
where
D = dissolved oxygen deficit above dam, mg/1
D. = dissolved oxygen deficit below dam, mg/1
T = temperature, °C
H = height through which the water falls, ft
a = 1.25 in clear to slightly polluted water: 1.00 in polluted
water
b = 1.00 for weir with free fall: 1.3 for step weirs or
cascades
An alternate equation developed from data on the Mohawk River and Barge
Canal in New York State (Mastropietro, 1968) is as follows:
Da - Db = 0.037H Da (IV-36)
Equation IV-36 is valid for dams up to fifteen feet high and for
temperatures in the range of 20° to 25 C.
In handling the problem of a dam, a new reach can be started just below
the dam. Da can be calculated as the value that occurs at the end of the
upstream reach. The new deficit Db, which will become the deficit at the
beginning of the next reach, is calculated using either of the above two
formulas.
392
-------
4.3.5 Dissolved Oxygen Saturation
The rate at which atmospheric reaeration occurs depends not only on k^,
but also on the difference between the saturation concentration GS and the
actual concentration C. The saturation value of dissolved oxygen is a
function of temperature, salinity, and barometric pressure. The effect of
salinity becomes important in esuarine systems, and to a lesser degree in
rivers where high irrigation return flow can lead to substantial salinity
values. Table IV-15 depicts the relationship between oxygen saturation and
chlorinity. The expression relating salinity and chlorinity concentration
is:
Salinity (°/ ) = 0.03 + 0.001805 chlorinity (mg/1) (IV-37)
where
°/oo represents parts per thousand.
The temperature dependence (at zero salinity) can be expressed as:
C - 14.65 - 0.41022T + 0.00791T2 - 0.00007774T3 (IV-38)
s
where T is in °C. This relationship is also found in Table IV-15 for zero
chloride concentration.
Barometric pressure affects C$ as follows:
cs' = cs V-TGO^P;/ dv-39)
027E\
/ .
V "
where
C =
s
water, mg/1
C = saturation value at sea level, at the temperature of the
s
C ' = corrected value at the altitude of the river, mg/1
P. = barometric pressure at altitude, mm Hg
393
-------
TABLE IV-15
SOLUBILITY OF OXYGEN IN WATER (STANDARD METHODS, 1971)
Temp.
in
°C
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Chloride Concentration
0
5,000
10,000
in Water - mg/1
15,000
20,000
Di fference
per 100 mg
Chloride
Dissolved Oxygen - mg/1
14.6
14.2
13.8
13.5
13.1
12.8
12.5
12.2
11.9
11.6
11 .3
11.1
10.8
10.6
10.4
10.2
10.0
9.7
9.5
9.4
9.2
9.0
8.8
8.7
8.5
8.4
8.2
8.1
7.9
7.8
7.6
7.5
7.4
7.3
7.2
7.1
13.8
13.4
13.1
12.7
12.4
12.1
11.8
11.5
11.2
11.0
10.7
10.5
10.3
10.1
9.9
9.7
9.5
9.3
9.1
8.9
8.7
8.6
8.4
8.3
8.1
8.0
7.8
7.7
7.5
7.4
7.3
13.0
12.6
12.3
12.0
11.7
11.4
11.1
10.9
10.6
10.4
10.1
9.9
9.7
9.5
9.3
9.1
9.0
8.8
8.6
8.5
8.3
8.1
8.0
7.9
7.7
7.6
7.4
7.3
7.1
7.0
6.9
12.1
11.8
11.5
11.2
11.0
10.7
10.5
10.2
10.0
9.8
9.6
9.4
9.2
9.0
8.8
8.6
8.5
8.3
8.2
8.0
7.9
7.7
7.6
7.4
7.3
7.2
7.0
6.9
6.8
6.6
6.5
11.3
11 .0
10.8
10.5
10.3
10.0
9.8
9.6
9.4
9.2
9.0
8.8
8.6
8.5
8.3
8.1
8.0
7.8
7.7
7.6
7.4
7.3
7.1
7.0
6.9
6.7
6.6
6.5
6.4
6.3
6.1
0.017
0.016
0.015
0.015
0.014
0.014
0.014
0.013
0.013
0.012
0.012
0.011
0.011
o.on
0.010
0.010
0.010
0.010
0.009
0.009
0.009
0.009
0.008
0.008
0.008
0.008
0.008
0.008
0 . 008
0.008
0.008
394
-------
P = saturation vapor pressure of water at the river
temperature, mm Hg
E = elevation, feet
Table IV-16 illustrates the variability of dissolved oxygen saturation with
altitude and temperature. The significant effect of altitude is apparent
and should not be neglected. For example, at a temperature of 20 C, the
saturation value decreases from 9.2 mg/1 to 7.2 mg/1 as the altitude
increases from sea level to 6000 feet, the approximate elevation of Lake
Tahoe and the Truckee River in California and Nevada.
4.3.6 DO-BOD Interactions
A widely used dissolved oxygen predictive equation is the
Streeter-Phelps relationship which predicts the dissolved oxygen
concentration downstream from a point source of BOD. Assuming a constant
river cross-sectional area, the dissolved oxygen deficit (C -C) can be
expressed as:
D -
D exp
-k
. u
J ka-kL
where
D
L
exp
reaeration coefficient, I/day
initial deficit (at x = 0), mg/1
deficit at x, mg/1
initial BOD (at x = 0), mg/1
BOD decay coefficient, I/day
- exp
(IV-40)
L and D are found by proportioning BOD and DO deficit concentrations just
upstream of the waste discharge with the influx from the discharge itself.
As presented earlier in the BOD section, LQ is given by:
W/5.38 + L Q
u u
Q,., + Q,,
(IV-41)
395
-------
TABLE IV-16
DISSOLVED OXYGEN SATURATION
VERSUS TEMPERATURE AND ALTITUDE
Temperature
0
5
10
15
20
25
30
35
ALTITUDE (ft)
0
14.6
12.8
11.3
10.2
9.2
8.4
7.6
7.1
2,000
13.6
11.9
10.5
9.5
8.5
7.8
7.1
6.6
4,000
12.5
11.0
9.7
8.8
7.9
7.2
6.5
6.1
6,000
11.5
10.1
8.9
8.0
7.2
6.6
6.0
5.6
8,000
10.5
9.2
8.1
7.3
6.6
6.0
5.4
5.1
396
-------
where
W - discharge rate of BOD, Ib/day
L - concentration of BOD in the river upstream of the waste
u
discharge, mg/1
Qu = river flow rate upstream of discharge, cfs
Qw = flow rate of waste discharge, cfs
Qw + Qu = flow rate of river in the reach under consideration,
cfs
W in Equation IV-41 should be expressed in terms of ultimate BOD, and not
5-day BOD.
The initial deficit is found from:
D =C
0 s
Q + Q
y 4
IV-42)
where
w
concentration of dissolved oxygen in the waste, mg/1
concentration of dissolved oxygen upstream of the waste
discharge, mg/1
dissolved oxygen deficit in waste, mg/1
dissolved oxygen deficit upstream, mg/1
In cases where information is lacking, D can normally be assumed to be in
the range 1-2 mg/1.
If NBOD is to be considered as well as CBOD, Equation IV-40 can be
modified as follows:
L_K| / ~~ KI X \ / ~ K
- 4-
U
D = DQ exp
-k x
a
__-.
k -k.
a L
exp
N
r / -kN x \
[exp ( --jj-j -
exp
IV-43)
If the decay coefficient of NBOD is approximately equal to that of CBOD,
Equation IV-40 can be utilized instead of the more complicated Equation
397
-------
IV-43. In this case, L in Equation IV-40 is replaced by the sum of L and
4.3.7 Dissolved Oxygen Calculations
Calculation of dissolved oxygen in rivers can proceed as shown in
Figure IV-19. The planner needs to estimate the waste loading scheme for
the prototype, whether it be for a 20 year projection or for current
conditions. The river system can then be divided into reaches and by
repeated use of Equation IV-40, dissolved oxygen calculations can be
performed for each reach, starting from a known boundary condition and
proceeding downstream. All data and calculations should be succinctly and
clearly recorded to minimize errors.
The dissolved oxygen profile downstream from a waste discharge
characteristically has a shape shown in Figure IV-18. If the reach is
o
q
ci
-Waste Inlet
0 TC(XC)
TIME (DISTANCE)
FIGURE IV-18
CHARACTERISTIC DISSOLVED OXYGEN
PROFILE DOWNSTREAM FROM A POINT
SOURCE OF POLLUTION
398
-------
Determine projected waste loading
scenario (source/sink distribution)
Criteria Met for
Hand Calculations
Divide river into reaches
Nj,
/
Determine temperature independent
parameters for each reach:
u, Q, d, A (as needed)
\,
/
Determine reaction rates at 20° C:
k , k, , k , SD (as needed)
a I n D
for each reach
\
/
\,
NO
/
Use Computer
Model
| Incorporate temperature corrections |
Determine C for each reach
Calculate conditions at x=o
(upstream end of present reach)
\
Perform and record
desired calculations
FIGURE IV-19 FLOW PROCESS OF SOLUTION TO DISSOLVED OXYGEN
PROBLEM IN RIVERS
399
-------
long enough, the dissolved oxygen deficit will increase to some maximum
value, D , at a distance x (termed the critical distance). D is called
L. L* L-
the critical deficit. Within any reach there will always be a minimum
dissolved oxygen value that occurs, but it may not be the critical deficit,
which is defined as the minimum point on a dissolved oxygen sag. The
difference between the minimum and critical values should be kept in mind.
As one example of the difference between the values, a reach may have a
dissolved oxygen profile where concentrations are monotonically decreasing
throughout the reach. The minimum DO will then occur at the downstream end
of the reach, but this will NOT be the critical DO value, since DO is still
decreasing in the downstream direction.
The travel time to the critical deficit is given by:
k
1
k -k.
a L
/ D (k -k, )
a /-, o a L
k, I1 ~ k. L
L \ L o
(IV-44)
The distance downstream can be computed by knowing the travel time and flow
velocity:
= U • t
(IV-45)
The critical deficit can be found from:
L k. \
o L \
-k
(IV-46)
The formulas for the critical dissolved oxygen deficit are not really
applicable in the special case when ka = kL. However, these special cases
can readily be handled in one of two ways. First a small change can be made
in either k or k, so that k, and k, are approximately equal. Or second,
a L d L
the following expression can be used to predict critical travel time:
400
-------
T - I I —
1. _ ~ 7—- I 1 — i
(IV-47)
uo /
Then, the critical deficit is given by
D = exp
IV-48)
Equation IV-48 is valid for all k /k. values, and is not limited to cases
a L
where k /k. = 1.
a L
Solutions to both Equations IV-46 and IV-44 are presented in Tables
IV-17 and IV-18, respectively. There exist practical limitations to the
solutions of both equations, governed by the conditions that the solutions
be both positive and real. If in solving Equation IV-44 t is negative, the
L*
minimum dissolved oxygen concentration actually occurs at the point of
discharge, and concentrations increase immediately below the discharge.
Tables IV-17 and IV-18 are particularly useful for computing the waste
assimilative capacity of a river. Waste assimilative capacity (WAC), as
defined here, is the amount of BOD that can be discharged into a river
without causing the minimum dissolved oxygen level to fall below a specified
value. In constructing Tables IV-17 and IV-18 extra detail was incorporated
for DO/LQ values between 0.0 and 0.5. This is necessary because most
practical problems fall within this range.
The following steps show how to use Table IV-17.
1. Find the reaeration rate (ka) and the BOD decay rate (kL) for
the river being investigated.
2. Find the BOD concentration in the river just below the point
of mixing (L0).
3. Find the dissolved oxygen deficit at this location
(D0 = Cs - C).
401
-------
UJ CO
_i a:
ca uu
(X)
UJ
o
_1
•»_
(_>
j><:
-^.
(V
-*a
o — —
to r-- ;o CO O"1 Q O •—
GO ON 01 o ~- ^-
„- tO <^1
O — f" O
to r- oo -3- -— co
.— O CC
o o cr
o o o — —
O (— — i—
CM
CD
o o
o c
-------
CQ t>O
<: ID
t— oo
o:
LU
J*
n:
j*:
r— o o en co
,— i— oocnccccr---^
O i-O O
<— o o
cn TT cc
CO LT! '— CO ^T
— .— ^- o o
o ^c cxi co KJ- o
o en en c.: cc cc
O -- -=3-
— en ic
ooocn
CO CO CO
.M O
CTl f^- ufl CM C~>
- O CO
o en cc
o en cc
= oc5c3o----r-^-
oooooooooc
CO
o
o o
o o
-------
4. Compute k /k. and D /L .
5. Using the ratios ka/kL and D0/l_0, find DC/L0 where Dc is the
critical deficit.
6. Finally, calculate DC = (DC/LQ) LQ, and Cmin = Cs - DC.
To use Table IV-18 complete these steps:
1.-4. Repeat steps 1 through 3 above.
5. Using the ratios k /k, and D /L , find k t .
a L U U a L
6. Calculate t = (k t )/ka.
C a C a
4.3.8 General Dissolved Oxygen Deficit Equation
The most general dissolved oxygen mass-balance formulation to be
presented in this chapter is as follows:
where
/ Lrd) [ /A
M rd ) \Fvn |
n !c I exP '
\ N / [
+ D0 exP (
°
^ Ao
^-ja
^ AM
f(x)) -
- exp
)
^1 f (x))
(IV'49)
P
R
S
oxygen production rate due to photosynthesis, mg/l/day
oxygen utilization rate due to respiration, mg/l/day
benthic demand of oxygen, mg/l/day
404
-------
The distance function f(x) expresses the cross-sectional area relationship
throughout the reach. The area can increase or decrease linearly or remain
constant. The general form of the relationship is:
f(x) = AQx + AA x2/2 , AA = A^A^
"
XL
where
Af = area at x = XL
A = area at x = 0
x. = length of reach
For a reach of constant cross-sectional area, Aft = 0.
In developing Equation IV-49 the following relationship for CBOD was
used (as originally presented in the BOD section):
(IV-22)
An analogous expression for NBOD was also used.
In Equation IV-49, the distributed sources and sinks (P, R, Sg, L .,
Nrd) are all mass fluxes, and no volumetric flow rate is associated with any
of these sources and sinks of dissolved oxygen.
4.3.9 Photosynthesis _and Resp i rat ion
The difficulty of accurately assessing the impact of photosynthesis and
respiration on the dissolved oxygen resources of streams is not readily
apparent from the single terms appearing in Equation IV-49. Of concern are
both free floating and attached algae, as well as aquatic plants. The
extent to which algae impact the dissolved oxygen resources of a river is
dependent on many factors, such as turbidity, which can decrease light
transmittance through the water column. Additionally, the photosynthetic
rate constantly changes in response to variations in sunlight intensity and
is not truly constant as implied by Equation IV-49. Hence if algal activity
405
-------
is known to be a significant factor affecting the dissolved oxygen balance,
the use of a computer model is recommended in order to accurately assess
such influences. For example, in the Truckee River in California and
Nevada, the diurnal variation of dissolved oxygen has exhibited a range of
from 150 percent saturation during the daylight hours to 50 percent
saturation at night due to algal photosynthesis and respiration,
respectively. At the most, hand calculations can give estimates of net
dissolved oxygen production rates that then can be compared to the other
source/sink terms in Equation IV-28. From this comparison the significance
of each can be estimated.
Table IV-19 presents some observed values of photosynthetic oxygen
production rates. As shown in the table dissolved oxygen production is
expressed in units of rate per unit area (gm/m -day). To convert to units
of concentration per unit time, the algal production rate must be divided by
river depth:
P = {-J- (IV-50)
where
P = production rate of dissolved oxygen, gm/m2-day
H = average river depth, meters
P = production rate of dissolved oxygen, mg/l-day
P can now be directly compared to other terms in Equation IV-28.
By using a regression equation developed by Erdmann (1979a, 1979b), the
production rate of dissolved oxygen, P, can be determined directly if the
diurnal variation of dissolved oxygen is known. When water temperature is
fairly constant throughout the day, the photosynthetic oxygen production
rate becomes
P = 2ADO (IV-51)
where
ADO = difference between the daily maximum dissolved oxygen
concentration and the daily minimum dissolved oxygen
concentration, mg/1
406
-------
TABLE IV-19
SOME AVERAGE VALUES OF GROSS PHOTOSYNTHETIC PRODUCTION OF
DISSOLVED OXYGEN (AFTER THOMANN, 1972 AND THOMAS AND O'CONNELL, 1966)
Water Type
Aver.Gross Production
(qrams/m -day)
Average Respiration
(gm/m -day)
Truckee River - Bottom
attached algae
Tidal Creek - Diatom Bloom
(62-109-106 diatoms/1)
Delaware Estuary - summer
Duwamish River estuary -
Seattle, Washington
Neuse River System -
North Carolina
River Ivel
North Carolina Streams
Laboratory Streams
3-7
0.5-2.0
0.3-2.4
3.2-17.6
9.8
3.4-4.0
11.4
6.7-15.4
21.5
2.4-2.9
Since Equation IV-51 is based on regression analysis, the units are not
consistent.
The importance of a constant water temperature is illustrated by Figure
IV-20. This figure shows the hourly variation of dissolved oxygen over a 24
407
-------
80*
rn
<
i
CD
CO
co
O
m
a
X
m
2:
<
>
>
—\
I—I
O
m
;o
CO
O
O
O
O
8
0
Z _,.
O oo
§8
CO
ro
6
O
O
0
O
co
O
O
O
oo
DISSOLVED OXYGEN MG/L
co O it
\
\
/
O
C
33
m
CD
\
1 \
CO 33
O >
- §
\
'
\
O
"\i
1 m
\
m
-------
hour period for Wyman Creek in California and for the Ivel River in England.
Both exhibit large diurnal dissolved oxygen variations, although the reasons
differ. In curve A (Ivel River) the dissolved oxygen level gradually
increases from 0600 hr to 1800 hr, and then decreases over the next 12
hours. The cause of the changing dissolved oxygen levels is a net
photosynthetic oxygen production during the daylight hours, and a net
comsumption during evening and night. Curve B is almost a mirror image of
curve A since the minimum dissolved oxygen levels occur during daylight
hours and the maximum during nighttime. The variations exhibited by curve B
are principally caused by a changing water temperature. During the day this
creek absorbs considerable solar radiation causing the water temperature to
rise and the dissolved oxygen saturation level to decrease. At night the
creek cools off and the dissolved oxygen saturation level increases. Curve
B then is free from the influence of photosynthetic effects, so it would be
erroneous to apply Equation IV-51. Erdmann (1979a, 1797b) and
Kelly et jf[. (1975) provide more sophisticated methods to predict P when
both photosynthetic and temperature effects occur concurrently. Example
IV-7 illustrates the utility of Equation IV-51.
EXAMPLE IV-7
Prediction of Photosynthetic Oxygen Production Rate
On Mechums River near Charlottesville, Virginia, Kelly ejt al_. (1975'
collected the following data:
Time of Day Stream Dissolved
(hours after midnight^ Temperature, °C Oxvaen (mq/£)
0-0 23.3 7.6
0-5 23.3 7.6
1-0 23.4 7 6
1.5 23.4 7.5
2-0 23.5 7 4
2-5 23.5 7 2
3.0 23.5 7.3
3-5 23.5 7 3
4-0 23.4 7 3
409
-------
Time ot Uay Stream Dissolved
(hours after midnight) Temperature, "C: Oxy_2_en_l!53L?J-
4.5 23.4 7.3
5 0 23.3 7.3
5.5 23.2 7.3
6.0 23.1 7.3
6.5 23.0 7.3
7.0 22.9 7.4
7.5 22.8 7.4
8.0 22.7 7.5
8.5 22.7 7.6
9.0 22.7 7.7
9.5 22.7 7.8
10.0 22.8 8.0
10.5 23.0 8.1
11.0 23.2 8.4
11.5 23.5 8.5
12.0 23.6 8.7
12.5 24.3 8.9
13.0 24.8 9.0
13.5 25.3 9.1
14.0 25.5 9.2
14.5 25.5 9.3
15.0 25.9 9.2
15.5 26.1 9.2
16.0 26.1 9.2
16.5 26.1 9.1
17.0 26.1 9.0
17.5 25.8 8.9
18.0 25.8 8.8
18.5 25.5 8.6
19.0 25.3 8.5
19.5 25.1 8.3
20.0 24.8 .8.2
20.5 24.5 8.0
21.0 24.2 8.0
21.5 24.0 7.9
22.0 23.8 7.6
22.5 23.7 7.7
23.0 23.6 7.7
23.5 23.6 7.6
24.0 23.5 7.5
Using a sophisticated analysis, Kelly et a]_. found the daily mean
photosynthetic oxygen production to be 4.40 mg/1. Using the data shown
above and Equation IV-51 estimate the daily photosynthetic oxygen
production, P (mg/l/day).
The minimum dissolved oxygen is 7.2 mg/1, which occurs at 0230. The
maximum dissolved oxygen is 9.3 mg/1 which occurs at 1430. Hence:
410
-------
P = 2ADO = 2(9.3-7.2) = 4.2 mg/l/day
This compares very well with the value found by Kelly et aJL using a more
sophisticated analysis, even though the stream temperature varies by a few
degrees during the day. Probably one reason for the good agreement is that
the maximum and minimum values occur about 12 hours apart, which the method
assumes they do.
END OF EXAMPLE IV-7
Values of photosynthetic respiration vary widely, ranging from 0.5
gm/m2/day to greater than 20 gm/m2/day. One suggested relationship between
respiration and chlorophyll a is given as (Thomann, 1972):
R(mg/l/day) = 0.024 (chlorophyll aj (yg/1) (IV-52)
where
1 pg/1 = 10"3 mg/1
Chlorophyl a. concentration is most commonly expressed in terms of ug/1.
4.3.10 Benthic Demand
In addition to oxygen utilization by respiration of attached algae,
benthic deposits of organic material and attached bacterial growth can
utilize dissolved oxygen. Table IV-20 illustrates some uptake rates. As
with photosynthesis, the uptake rates are expressed in gm/m2-day. To use
these values in Equations IV-28 or IV-49, division by stream depth (in
meters) is necessary. Temperature effects can be approximated by
T_?n
= (S ) 1.065'-^ (IV_53)
T b 20
411
-------
TABLE IV-20
AVERAGE VALUES OF OXYGEN UPTAKE RATES OF
RIVER BOTTOMS (AFTER THOMANN, 1972)
Bottom Type and
Sphaerotilus - (
Municipal Sewage
Outfall Vicinity
Location
10 gm dry wt/M2)
Sludge -
Municipal Sewage Sludge -
"Aged" Downstream of Outfall
Cellulosic Fiber
Estuarine mud
Sandy bottom
Mineral soils
SI udge
2
Uptake (gms 02/m -day)
@ 20°C
Range
-
2-10.0
1-2
4-10
1-2
0.2-1.0
0.05-0.1
Approximate
Average
7
4
1 .5
7
1.5
0.5
0.07
The areal extent of significant oxygen demanding benthic materials is
often limited to the region just below the outfall vicinity. Although the
oxygen demand may be great over a short distance, it may be insignificant
over larger distances. The response of rivers to areally limited benthic
deposits is generally to move the critical deficit upstream, but not to
lower its value significantly.
Zison e^t jil_ (1978) contains significantly more data and further
discussion of benthic oxygen demand in rivers. Additionally Butts and Evans
(1978) conducted extensive studies of sediment oxygen demand on 20 streams
in Illinois. They found that benthic oxygen demand could be predicted as:
412
-------
= 0.15T + 0.3D + 0.11 logN - 0.56 (IV-54)
where
Sg = benthic oxygen demand, g/m2-day
T = water temperature, °C
Ds = depth of sediment, inches
N = number of macroinvertebrates per m
2
They found that N typically ranged from 10,000 to 1,000,000. Within this
range the sum of the last two terms is between ±0.1, and is negligible
compared to the first two terms. Under these conditions Equation IV-54
simplifies to
SB = 0.15T + 0.3DC (IV-55)
D b
The depths of sediment found during the study of Butts and Evans (1978)
ranged from 1 to 17 inches. Consequently Equation IV-55 is applicable to
streams which have fairly significant benthic oxygen demands. For cleaner
streams Equation IV-55 probably overestimates the benthic oxygen demand.
4.3.11 Simplifying Procedures in Dissolved Oxygen Calculations
Using Equation IV-49 might be untenable for several reasons, such as
lack of available data, or because of the voluminous calculations required
to apply it to a large number of reaches. Several suggestions are offered
here that should simplify analysis of dissolved oxygen problems.
Since the general scope of this section is to facilitate the
determination of existing or potential problem areas, the analysis should
proceed from the simple to the more complicated approach. It may be
adequate to analyze the dissolved oxygen response to the most severe
loadings first, neglecting those of secondary importance. If such an
analysis clearly indicates dissolved oxygen problems, then the inclusion of
any other pollutant discharges would only reinforce that conclusion. More
rigorous procedures (e.g. a computer model) could then be employed to
perform a detailed analysis.
413
-------
Suppose the improvement of dissolved oxygen levels due to decreased
loading from a point source is of interest. This is a common situation
since it relates to the design of waste loading abatement schemes. Such
improvement can be estimated by:
AD =
exp
k x
a
exp
where
Al_
AD
- exp -/ f(x)
V Ao >
the change in the initial BOD, mg/1
change in deficit in response to ALQ
(IV-56)
Equation IV-56 was formulated from Equation IV-49 assuming that LQ and DQ
are the only changes of significance.
Many rivers have a large number of point sources. Although this is not
necessarily a complicating factor, a detailed analysis might be too time
consuming for hand calculations. There are several possible alternatives to
deal with this situation in order to reduce the number of reaches to be
analyzed. The first, already mentioned, is to consider only the significant
pollutant sources. Second, as was illustrated in Example IV-5, a number of
uniformly distributed point sources can be considered as a single
distributed source. Third, combining several adjacent point sources is also
possible, if the length of the reach under consideration is long relative to
the distance of separation between the point sources. Analogously, a
distributed source can be approximated as a point source, contributing the
same waste loading and located at the center of the distributed source.
It may be that the planner wants only to determine the critical
dissolved oxygen concentration in each of a series of reaches. In this case
no more than two values of dissolved oxygen per reach need be calculated.
Figure IV-21 shows the solution process to be followed.
414
-------
Go to next
reach
Determine k and k,
a L
for each reach
Begin reach
calculations!
Find D ,L
o' o
(at x = o)
Find D at
x = x.
.7
\
/
V
/
\
Find D
at xc
D =DC
at
X = 0
^YES 1
\
/YES ,
\
Find tc, xc = Utc
Find D at x.
D = D
FIGURE IV-21 FLOW PROCESS IN REACH BY REACH SOLUTION TO
CRITICAL DISSOLVED OXYGEN VALUES
415
-------
One final note on dissolved oxygen evaluations should be made here. It
may be that if the planner is interested primarily in locating dissolved
oxygen problems, he need not perform any computations. This is especially
likely where dissolved oxygen data are available at various locations on the
river. Plotting dissolved oxygen time trends may reveal when, as well as
where, annual dissolved oxygen minima occur.
EXAMPLE IV-8
Determining River Assimilative Capacity from
Tables IV-17 and IV-18
Suppose the user wants to determine waste assimilative capacity (WAC)
for a river reach that has the following characteristics:
critical dissolved oxygen concentration = 5.0 mg/1
(user establishes this)
initial deficit = 1.0 mg/1
average velocity = 0.5 fps
average depth = 4 feet
chloride concentration = 0
temperature range = 10°C to 35°C
First, kg and kL need to be found. From Figure IV-17, ka (20°) = 0.8/day,
and from Figure IV-11, k, = 0.4/day. At any other temperature then, kaand
k. can be found from the temperature relationships previously developed:
k = (k ) 1.0241-20 (IV-34)
a a 20
k, = (k, ) 1.0471"20 (IV-17)
L L 20
Using Table IV-15 the dissolved oxygen saturation concentration within the
temperature range of interest can be found. This information can be then
compiled into Table IV-21 shown below.
416
-------
TABLE IV-21
COMPILATION OF INFORMATION IN EXAMPLE IV-8
T
(°0
10
15
20
25
30
35
Cs
(mg/1)
11.3
10.2
9.2
8.4
7.6
7.1
C
(mg/1)
5.0
5.0
5.0
5.0
5.0
5.0
D
(mg/1)
6.3
5.2
4.2
3.4
2.6
2.1
D /D
o c
0.16
0.19
0.24
0.29
0.38-
0.48
VkL
2.5
2.2
2.0
1.8
1.6
1.4
Using the values of DQ/DC and ka/kL, LQ can be found, which in this case is
the WAC.
Procedure
1. Table IV-21 is entered at the appropriate ka/k,L column. This
is 2.5 at 10°C.
2. Next, the entry within the ka/kL column in Table IV-17 is
found such that
Dc/Lo
D,
0.16
Since the left-most column of Table IV-17 is DQ/L0 and the entries are
D /L , the ratio of these values is calculated until that ratio equals 0.16.
For example, try DO/LQ = 0.05. Then DC/LQ = 0.23 and
0.05 = 0.22 >0.16, too big
0.23
417
-------
try DQ/LO = 0.04. Then DC/|_O - 0.23 and
n'oo = .17, close enough
then p^ - .23, or L =
fi
o .23
= 27.4 mg/1
—
The results are tabulated below for the temperature range 10°C to 35°C.
T(°C)
10
15
20
25
30
35
WAC (mg/1)
27.4
20.0
15.0
11 .3
7.6
5.4
D /L
0 0
0.04
0.05
0.07
0.09
0.13
0.19
L is directly related to the loading rate of BOD, as expressed
earlier in Equation IV-41:
WAC = (Lo)
critical
+W
critica1/5.38
Q + Q
x M
From equation IV-41 the critical waste loading W can be found. If
desired, this procedure can be repeated for different river flow rates,
and WAC and W ._ , found for the various flows. To do this, diff-
cntical
erent average depths and velocities will be needed. Generally this
analysis is most apolicable to minimum flow conditions, as
this is the most critical situation, but higher flows may be of interest
to assess the benefits of flow augmentation decisions. Novotny and
418
-------
Krenkel (1975) have used a 20 year, 3-day low flow in analyzing the
Holston River in Tennessee. For further discussion of low flow cal-
culations refer to Section 4.4.6.
In interpreting the results of this example the user should be
looking more at trends rather than particular results. For example,
notice how the WAC decreases with increasing temperature. For every
10° increase the WAC is approximately halved. A similar relationship
between WAC and flow rate could also be determined.
Finally, using Table IV-18, the travel time t can be determined
to the point of critical deficit. The appropriate DQ/LO and ka/kL
values are used to find t . Table IV-22 illustrates these results.
TABLE IV-22
CRITICAL TRAVEL TIME RESULTS
T(°C)
10
15
20
25
30
35
ka/kL
2.5
2.2
2.0
1.8
1.6
1.4
VLo
0.04
.05
.07
.09
.13
.19
t k
c a
1.4
1.3
1.2
1.13
1.0
0.9
ka
.63
.71
.8
.9
1.0
1.1
todays)
2.2
1.8
1.5
1.2
1.0
0.8
END OF EXAMPLE IV-8
419
-------
EXAMPLE IV-9
Critical Deficit Calculations for Multiple Reaches
Suppose the critical deficit in each of the three reaches of the
river illustrated in Figure IV-22 is to be determined. The conditions
upstream of the first discharae are:
T = 27UC
Q = 600 cfs
U = 0.4 fps
depth = 5.0 feet
Du = 1 mg/1
Lu = 2 mg/1
Using these data, along with the solution process outlined in
Figure IV-21, the following procedure can be used:
1. Determine ka, kL for each reach. For this example it will
be assumed that the average depth, velocity, and temperature remain
relatively constant over the three reaches, so that k and k, are
a L
also the same.
.Mixing
Du=lmg/l
Qu=600cfs
f
Zone
I2MI.
B.O.D.L=40mg/l
Q=50MGD
Mixing
Zone
4MI.
B.O.D.L=50mg/l
Q=60MGD
B.O.D.L=20mg/l
Q = IOMGD
FIGURE IV-22
HYPOTHETICAL RIVER USED IN EXAMPLE IV-9
420
-------
k (20) = 0.5, (from Figure IV-l?)
a
kL (20) = 0.35, (from Figure IV-11)
Using the temperature correction:
k (27) = 0.60, (from Equation IV-34)
a
''KL (27) = 0.48, (from Equation IV-17)
The saturation dissolved oxygen concentration at 27°C and 0%0 salinity
is (from Table IV-15) 8.1 mg/1.
2. For the first reach, calculate L and D .
o o
L = (2)(600)+(40) (50) H.55) . _, ._
0 600 +(50)(1.55) ~ 6'35 mg/1
For lack of better information about the dissolved oxygen
characteristics of the waste, it can be assumed that D = D = 1 mg/1.
The location of the critical deficit can now be calculated
using Table IV-18, or Equation IV-45• In this example Table IV-18 will
be used. To use that table, the following ratios are needed:
DO/LQ = 1/6.35 = 0.16
and
k /k, - 0.60/0.48 - 1.3
ct L
From Table IV-18, katc - .92 or
tc = .92/0.6 =1.53 days
x _ (0.4) (1.53) (3600) (24) .. . „
xc 5280" = 10.0 miles
421
-------
Since x < 12, the critical deficit actually exists, and is located 10
miles downstream. From Table IV-17 D can be found by entering it with
v*
the same ratios used in Table IV-18. The result is:
~ = .38 + Dr = 2^_ mg/1
o C
3. Before the critical conditions in reach 2 can be calculated,
the conditions at the upstream end of that reach must be established.
The conditions at the downstream end of reach 1 are
D = 2.3 mg/1, from Equation IV-40
L = 2.6 mg/1 from Equation IV-42
The conditions at the upstream end of reach 2 are thus:
.(2^) (677)+(60) (K55) =
«.Jt> mg/i
D =2.3 can be used for lack of better information on the dissolved
o
oxygen concentration in the effluent to reach 2. For use in
Table IV-18, it is found that
VLo = "
So
k t = .76
t = .76/0.6 =1.3 days
\s
x = 8.3 miles
422
-------
Since reach 2 is only 4.0 miles long, the critical deficit is not
reached. Instead the maximum deficit will occur at the downstream
end of reach 2, where:
D = 3_._3_rnq/1_, Equation IV-40
JL_=_.6,?_2 JiS/1, Equation IV-22
4. For the beginning of reach 3, LQ and D must be found.
I - (20)00)0.55) +(770.5) (6.22)
Lo -- 770.5 --- --- -=6.5mg/l
For DQ, it can be assumed that GW = 5.0 mg/1. From Equation IV-41,
then
n - R i _(8.1 - 3.3) (770.5)+(5.0)(10) (1.55), . - ,
o " °-' ~ 770.5 + 15.5~ •3"3 mg/l
The calculations of critical conditions can now be made for
this reach, as for the previous two.
END OF EXAMPLE IV-9
4.4 TEMPERATURE
4.4.1 Introduction^
The biota comprising an established aquatic ecosystem generally
respond negatively to significant abnormal temperature fluctuations.
Anthropogenic modifications of rivers and streams can alter the thermal
regime, most often by elevating the maximum and mean water temperatures,
Repercussions of elevated temperatures are manifested through a shift
423
-------
in the ecological balance and in the water quality of rivers. For
example, there is a progression in the predominance of algal species
from diatoms to green algae to blue-green algae as water temperature
increases through a specific range. Thermal discharges can increase
the ambient temperature enough to alter the predominant species to
the undesirable blue-green algae. Increased metabolic activity of
aquatic organisms, such as fish, also accompanies elevated tempera-
ture. If the increase is high enough, the results can be lethal. Much
data are available today (e.g., Committee on Water Quality Criteria,
1972) which specify lethal threshold temperatures for aquatic organisms.
Water quality may be adversely affected through decreased
solubility of dissolved oxygen and increased biochemical reaction
rates. Adequate dissolved oxygen levels, particularly at elevated
temperatures, are critical because of the increased metabolic acti-
vity. Yet, as previously discussed the saturation concentration
of dissolved oxygen diminishes with rising temperature. Worse still,
is the concurrent low flow condition which is associated, in many
parts of the country, with the warm summer months. For example, in
a study of 30 river reaches in the U.S. (EPA, 1974), 20 had lower
flows in the summer months than in the winter. This situation
further reduces assimilative capacity and usually results in the
most critical dissolved oxygen levels over the year.
Man can alter the thermal regime of rivers by removing trees,
changing the flow regime, and by increasing thermal discharges.
Diversions of water from a river can reduce the water depth, and
increase the mean and diurnal fluctuation of stream temperature.
In Long Island, modification of the natural environment of streams
has increased average stream temperatures during the summertime by
as much as 9 to 14 F (Pluhowski, 1968';. Concurrent temperature
differences of as much as 14 to 18°F !etueen sites on the same stream
were observed on dtys of high solar radiation. A principal factor
involved in these occurrences was the removal of vegetation along the
banks of the streams, permitting significantly greater penetration of
solar radiation. Other contributing factors cited by Pluhowski
424
-------
included increased stormwater runoff, a reduction in the amount of
groundwater inflow, and the introduction of ponds and lakes.
4.4.2
If a body of water at a given initial temperature is exposed
to a set of constant meteorological conditions, it will tend to
approach some other temperature asymptotically. It may warm
by gaining heat or cool by losing heat. Theoretically, after a
long period of time the temperature will become constant and
the net heat transfer will be zero. This final temperature has been
called the equilibrium temperature, E. At equilibrium, the heat
gained by absorbing solar radiation and long-wave radiation from the
atmosphere will exactly balance the heat lost by back radiation,
evaporation, and conduction.
These heat fluxes are illustrated in Figure IV- 23 which also
shows typical ranges for the fluxes. Some of these terms (Hs>Ha>Hsr>
H ) are independent of water temperature, while the remainder
(H, ,H ,H ) are dependent upon water temperature. At equilibrium then,
H (net transfer) equals zero, or
Hs - Hsr + Ha - Har ' Hbr ' Hc ' He = ° (IV'
In actuality, the water temperature rarely equals the equilibrium
temperature because the equilibrium temperature itself is constantly
changing wit.i the "ocal meteorological conditions. The equili-
brium temperature will rise during the day when solar radiation is
greatest, and fall to a minimum at night when solar radiation is absent.
A daily average equilibrium temperature may be computed using a
number of factors deluding daily average values of radiation, temp-
erature, wind speed, and vapor pressure. The daily average value will
reach a maximum in midsummer and a minimum in midwinter. Since the
actual water temperature always tends to approach, but does not reach
the equilibrium temperature, it will usually be less than equilibrium
425
-------
H = Shortwave solar radiation (400-2800 BTU ft"2 day "*)
H = Long wave atmospheric radiation (2400-3200 BTU ft~2 day"1)
a
i
=Long wave back radiation (2400-3600 BTU ft"2 day"1)
He = Evaporative heat loss (2000-8000 BTU ft"2 day"1)
= Conductive heat loss or gain (-320-+400 BTU ft"2 day"1)
.. = Reflected solar (40-200 BTU ft"2 day"1)
? i
AHar = AtmosPheric reflection (70-120 BTU ft day" )
NET RATE AT WHICH HEAT CROSSES WATER SURFACE
["
= (H
sr
He) BTU ft"2 day"1
Absorbed Radiation (HR) Temperature Dependent Terms
Independent of Water Temperature
Hbr~ (Ts + 460)4
Hc ~ (Ts - Ta)
He ~ W(es - ea)
FIGURE IV-23
MECHANISMS OF HEAT TRANSFER ACROSS A
WATER SURFACE (PARKER AND KRENKEL, 1969)
in the spring when temperatures are rising, and greater than equilibrium
in the fall when temperatures are dropping. During a 24 hour period,
the equilibrium temperature usually rises above the actual water temperature
during the day and falls below the water temperature at night, forcing the
water temperature to follow a diurnal cycle.
426
-------
The amplitude of the actual diurnal water temperature cycle is
generally dampened significantly in comparison to the amplitude of
the equilibrium temperature cycle due to the large heat capacity of
water. A thermal discharge into a water body will usually increase
the actual daily amplitude because of the water temperature dependent
terms in Equation IV-57. This situation is illustrated in the following
example (Edinger, et al., 1968). Figure IV-24 illustrates a flow through
cooling pond into which a thermal effluent is discharged (at Station B).
Sta.
Sta.G.
, 430
Sta E / ACRES
Sta. D.
Sta.C
FIGURE IV-24
SCHEMATIC OF SITE No, 3
COOLING LAKE (FROM EDINGER,
ET AL,, 1968)
Temperature observations were recorded at Stations B through H at four-
hour periods for one week. The findings are depicted in Figure IV-25.
427
-------
120
7/18 7/19 ' 7/20 7/21 7/22 7/23 7/24
DAY (4 HOUR PERIODS)
FIGURE IV-25
OBSERVED TEMPERATURES, SITE No, 3,
JULY 18 - JULY 24, 1965 (EDINGER,
ET AL,, 1968)
The highest temperatures and largest diurnal temperature variables are
recorded at Station B. The peak temperature at Station B occurs just
after noon, corresponding to the peak loading from the plant. At Station
C the peak temperature is at 1800 hours, indicating the lag in flow time
from Stations B to C. The peak temperatures at the remaining stations are
more influenced by meteorological conditions, and less by the thermal
discharge. The relationship of the observed temperatures to the
equilibrium temperature over a 24-hour period is shown in Figure IV-26 .
Note the amplitude of the equilibrium temperature E (33°F amplitude).
The average equilibrium temperature, E, is approximately 91 F.
A progression from Station B to Station H indicates that the daily
water temperature tends to approach the average equilibrium temperature.
428
-------
vo
rt-
ro
3
-a
ro
n
Cu
r-t-
c
ro
o
-h
CU
^
ro
3
X
ro
CL
cr
o
CL
O
-h
5
(— (.
ro
-s
n
Cu
3
Q-
ro
ro
CO
— i.
3
CU
r-t-
ro
Q.
-P>
•
-£*
•
oo
oo
^--i-
c
o
Q. cu^
ro
CO
o
c
«j
•— -
rn
CL
— 1.
^
ca
ro
-1
CU
^3
CL
CD
ro
*•<
ro
-s
— i
UD
C^
Oi
^,
o
3
O
h
m
JD
c:
«!•
— J
-J«
CT
"^
^~
3
— )
ro
-^
•S |-Q
3-
ro
-s
cu a'
< r*-
ro c
CO
-s
ro
0
s
rt-
CU
r-t-
(-t-
3-
ro
ro
cr
-s
— 1.
3
O
-s
ro
CU
co
rt>
•
3
CL
CL
CU
*<
5!
3"
ro
3
m
— »•
co
^3-
— i.
ID
3"
ro
-s
^
3-
o
JT
ro
<
ro
~^
**
r-t-
3-
ro
c-t-
ro
3
-a
ro
Cu
c:
ro
co
Cu
r-t-
^
3"
ro
co
ro
£
CU
c-t-
«j.
O
3
CO
m
CO
— J
0
s:
**
r-t-
3"
ro
s:
Cu
c-t-
ro
-5
c-t-
ro
3
T3
ro
-5
Cu
ff
C
-s
ro
ex
ro
o
-s
ro
Cu
CO
ro
CO
Cu
t-t-
<-*•
3-
ro
CO
ro
co
r-t-
CU
_i.
O
3
co
a
c
-s
— 1.
^3
CO
ro
3
CL
CO
f-t-
o
— 1.
3
n
-^
ro
01
CO
ro
»— H
^
c-f.
3-
ro
ro
CU
-s
— i
*•<
o
3
— i.
3
CQ
Cu
3
Q.
CU
ro
ro
<
ro
3
— i.
3
U2
3"
o
c
-s
CO
g;
3-
ro
3
m
i.
r-t-
r-t-
ro
3
Q.
CO
f-f
O
CL
ro
o
-5
ro
CU
CO
ro
*
CU
3
a.
s
3"
ro
3
r+
3"
ro
<-+
ro
^-*
ro
-s
^ '
c
— ^
;
ro
'
cr
ro
o
^~
rn
_J>
c-t-
CU
ro
-5
g-t-
ro
3
-a
ro
— ^
CU
f-t-
c:
-s
ro
-j.
CO
CU
CT
O
<
ro
C-t"
3-
ro
3
CO
r-t-
Cu
3
r~t-
CU
3
ro
0
cr
CO
ro
~£
^
-
!
cr
"^
t~-
ro
-a
ro
Cu
_i_
C ~
— 5
ro
-5
ro
CL
0
3
3
Cu
- f-^-
_J.
3
IQ
_J.
3
-h
C
ro
3
n
ro
o
3
ro
ro
0
-s
o
o
to
o
Cu
— 1
n
0
3
O.
_j.
r*-
O
3
co
• ;
•sr
3"
ro
3
r-t-
3"
ro
Cu
3
cr
— it
ro
3
j—t-
oo
Cu
ft-
~J(
o
3
CO
CD
Cu
3
Q.
"~^
u
Cu
3
Q.
c-t-
3-
ro
Cu
3
cr
ro
3
rt
ro
3
ro
CU
c-t-
C
-s
ft)
— (
z
"
Cu
— i
-~j
ro
-h
ro
o
r-t-
3-
tn
m rr
H o
c:
2:
m
a
00
O
m
en
c:
70
m
i
ro
en
2 O
W 2
\—• c: —i
UD r~ rn
cn -< 2
CO ~a
•—' i—i m
co -x
I >
C_ H
C C
-< m
ro <;
4=r >
*• rn
t-O -H
cn •—>
cn o
m
m
m
oo
73
»—H
00
o
~D O
m ~n
o
2
Tl
m
a
-i m
IE (D
c:
CD •—
GO •—
m Cd
m
a
—• m
H >
rn z
cr
2
•z
t)
TEMPERATURE (°F)
ro
TEMPERATURE (°C)
-------
0.05E2 HR - 1801 K -
15.7 [e - C(B) + 0.26T 1 /
726TB) L 3 -I UV'58
where
E = equilibrium temperature, F
K = thermal exchange coefficient, BID/ ft2/day/°F
Hp = net incoming short (H ) and long (H ) wave radiation
i» o S D an
BTU/ftVday
T = air temperature, °F
a
e, = water vapor pressure of ambient air at air temperature,
a
B = proportionality coefficient, mmHg/°F
mmHg
prop
C(B) = value dependent on B, mmHg
The thermal exchange coefficient K is expressible as:
K = 15.7 + (0.26+B) f(u) (IV- 59)
where f(u) is a function of wind speed. Different relationships for
f(u) have been developed. For purposes of hand calculations, the
following relationship will be used:
f(u) - 11. 4u (IV-60)
where u is the daily average wind speed in mph.
To calculate E using Equation IV- 58 an iterative procedure is
needed, since K, B, and C(B) depend on E. The following steps outline
a solution procedure.
1. Data needed to start the procedure include:
T , relative humidity, wind speed, and net shortwave
d
solar radiation. Figure IV-27 illustrates daily average
solar radiation reaching the continental United States
430
-------
CD
cr
70
m
O H IS
-n rc m
m >
2 GO >
m - —•
"X3 r~
o -n -<
m o
<> 33 GO
o
\-* c_ r~
i-O cr >
en r- 33
oo -<
2 O
a —
cr —•
CO O
cr z:
CO
r~
^-> 3>
cn -z.
CO
CO
m
H
n:
m
-TO
>
;o
H
3 o
m cr
z cr>
H 1C
O
Atlantic Ocean
Pacific Ocean
8
Atlantic Ocean
-------
for the months July and August. It is during these
months that stream temperatures usually reach their
annual maxima. These values do not account for the
albedo of water (the percent of incoming solar radiation
that is reflected), but since this is small, it can be
ignored. Because of the variability caused by
topography, vegetative cover, and other factors,
local sources of information should be used when possible
for solar radiation values.
2. Calculate HD = H + H (BTU/ft2/day). If Figure IV-27
r\ sn an „
is utilized for H$n, convert from langleys/day to BTU/ft /day
by multiplying by 3.7. H can be estimated from Table IV-23
ci n
by knowing the air temperature and the cloud cover fraction
(0.1 to 1.0).
3. Determine e= from Table IV-24 by entering with T and
a a
relative humidity.
4. Choose an initial value for E. The air temperature T
a
can be the first guess.
5. Enter Table IV-25 for B and C(B) at E (°F).
6. Knowing u, f(u), and B, calculate K from Equation
IV-59.
7. From Equation IV-58 make the next estimate of E (E }
* new'
by evaluating the right hand side of that equation
(call this result F(E)).
432
-------
TABLE IV-23
NET LONG WAVE ATMOSPHERIC RADIATION, H
an
co
CO
Cloud
Cover
.1
.2
.3
.4
.5
.6
.7
.8
.9
1.0
Tempera-
ture
(°F)
35
65
35
65
35
65
35
65
35
65
35
65
35
65
35
65
35
65
35
65
Han
(BTU/Sq.
Ft/ Day)
1685
2400
1694
2412
1708
2432
1728
2461
1754
2497
1785
2542
1822
2595
1865
2656
1914
2725
1968
2803
Tempera-
ture
(°F)
40
70
40
70
40
70
40
70
40
70
40
70
40
70
40
70
40
70
40
70
H
an
(BTU/Sq.
Ft/Day)
1790
2540
1799
2553
1814
2575
1835
2605
1863
2644
1896
2691
1936
2747
1981
2812
2033
2885
2091
2967
Tempera-
ture
(°F)
45
75
45
75
45
75
45
75
45
75
45
75
45
75
45
75
45
75
45
75
Han
(BTU/Sq.
Ft/Day)
1900
2688
1910
2701
1926
2724
1949
2756
1978
2797
2013
2847
2055
2907
2103
2975
2158
3053
2220
3139
Tempera-
ture
(°F)
50
80
50
80
50
80
50
80
50
80
50
80
50
80
50
80
50
80
50
80
Han
(BTU/Sq.
Ft/ Day)
2016
2842
2026
2857
2043
2881
2067
2914
2098
2958
2136
3011
2180
3074
2232
3146
2290
3228
2355
3320
Tempera-
ture
(°F)
55
85
55
85
55
85
55
85
55
85
55
85
55
85
55
85
55
85
55
85
Han
(BTU/Sq.
Ft/Day)
2138
3004
2149
3019
2167
3045
2192
3080
2225
3126
2265
3182
2312
3249
2366
3325
2428
3412
2497
3509
Tempera-
ture
(°F)
60
90
60
90
60
90
60
90
60
90
60
90
60
90
60
90
60
90
60
90
H
an
(BTU/Sq.
Ft/ Day)
2266
3173
2277
3190
2296
3216
2323
3254
2358
3303
2400
3362
2450
3432
2508
3513
2573
3604
2646
3707
-------
TABLE IV-24
WATER VAPOR PRESSURE (mmHg) VERSUS AIR TEMPERATURE, T , AND RELATIVE HUMIDITY
a
CO
-P*
Ta
<°F)
35
40
45
50
55
60
65
70
75
80
85
90
95
100
es*
(mmHg)
5.2
6.3
7.6
9.1
11.0
13.1
15.6
18.6
22.0
26.0
30.5
35.8
41.8
48.7
0.1
0.5
0.6
0.8
0.9
1.1
1.3
1.6
1.9
2.2
2.6
3.1
3.6
4.2
4.9
0.2
1.0
1.3
1.5
1.8
2.2
2.6
3.1
3.7
4.4
5.2
6.1
7.2
8.4
9.7
0.3
1.6
1.9
2.3
2.7
3.3
3.9
4.7
5.6
6.6
7.8
9.2
10.7
12.5
14.6
R E L
0.4
2.1
2.5
3.0
3.6
4.4
5.2
6.2
7.4
8.8
10.4
12.2
14.3
16.7
19.5
A T I V E
0.5
2.6
3.2
3.8
4.6
5.5
6.6
7.8
9.3
11.0
13.0
15.3
17.9
20.9
24.4
H U M I
0.6
3.1
3.8
4.6
5.5
6.6
7.9
9.4
11.2
13.2
15.6
18.3
21.5
25.1
29.2
D I T Y
0.7
3.6
4.4
5.3
6.4
7.7
9.2
10.9
13.0
15.4
18.2
21.4
25.1
29.3
34.1
0.8
4.2
5.0
6.1
7.3
8.8
10.5
12.5
14.9
17.6
20.8
24.4
28.6
33.4
39.0
0.9
4.7
5.7
6.8
8.2
9.9
11.8
14.0
16.7
19.8
23.4
27.5
32.2
37.6
43.8
1.0
5.2
6.3
7.6
9.1
11.0
13.1
15.6
18.6
22.0
26.0
30.5
35.8
41.8
48.7
* e = saturated vapor pressure
-------
TABLE IV-25
B AND C(B) AS FUNCTIONS OF TEMPERATURE
Temperature
(°F)
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
B
(mmHq/ F)
.286
.296
.306
.317
.328
.340
.352
.365
.378
.391
.405
.419
.433
.448
.464
.479
.496
.512
.529
.547
.564
.583
.601
.620
.640
C(B)
(mmHg)
-5.5
-4.5
-4.1
-4.2
-4.6
-5.4
-6.3
-7.5
-8.7
-10.0
-11.2
-12.5
-13.6
-14.7
-15.8
-16.7
-17.6
-18.3
-19.0
-19.6
-20.1
-20.7
-21.2
-21.7
-22.3
Temperature
(°F)
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
B
(mmHg/0F)
.660
.680
.701
.722
.743
.765
.787
.810
.833
.857
.881
.905
.930
.955
.980
1 .006
1 .033
1.060
1.087
1.114
1.142
1.171
1.200
1.229
1.259
C(B)
(mmHg)
-22.9
-23.6
-24.4
-25.4
-26.5
-27.8
-29.3
-31.0
-33.0
-35.1
-37.6
-40.3
-43.2
-46.4
-49.7
-53.3
-57.1
-61 .0
-64.9
-68.9
-72.9
-76.7
-80.4
-83.8
-86.8
95
1 .289
-89.3
435
-------
8. The next estimate of E is:
Enew=0.3E+0.7F(E)
(Note: this choice of E brings about a more
rapid convergence to the answer than would use
of E alone).
9- If lE- EUlF, then E , , = E .
1 ' ' actual new
If |E - -E >1 F, return to step 5 with EnQi and
new
new
repeat the procedure until the convergence criterion
is met, namely, Eactufl1 = Enew.
Instantaneous, daily average, v.-eekly average, or even longer
term average equilibrium temperature, F, can be calculated by
usina mean meteorological conditions over the period of interest
and following the solution procedure just outlined. Cal-
culating the daily average E under the most crucial
annual meteorological conditions (usually occurring in July or
August) yields the highest temperature about which that water body
. tends to naturally oscillate. The repercussions of man's activities
in terms of altering E can thus be estimated and analyzed for potential
impact.
EXAMPLE IV-10
Calculation of Equilibrium Temperature
On Long Island, New York, studies done by Pluhowski (1968) have
indicated that shading of streams by a natural vegetative canopy
can drastically affect the shortwave solar radiation reaching those
streams. The results of some of his findings are presented in
Table IV-26. In the summer, when leaves are on the trees, the
actual solar radiation reaching the Connetquot River can be as low
as 29% of that reaching unobstructed sites at nearby Mineola or
Brookhaven.
436
-------
TABLE IV-26
SUMMARY OF SOLAR-RADIATION DATA
FOR MIMEOLA, BROQKHAVEN, AND THE CONNETQUOT RIVER SITES
GO
Mean-Daily Solar Radiation in Langleys:
for the Indicated Periods
Solar
Site
(1)
1
2
3
1
2
3
1
2
3
1
2
3
1
Jan.
Jan.
Jan.
Apr.
Apr.
Apr.
Apr.
June
June
June
Aug.
Aug.
Aug.
Nov.
Dates
(2)
30, 31
28, 29
25, 26
21-23,
16-18,
19, 20
24-26,
9-11,
7, 8,
12-14,
26-28,
22-24,
29, 30
28, 29
, 1967
, 1967
, 1967
1967
1968
, 1967
1967
1967
1967
1967
1967
1967
, 1967
, 1967
Mineola
(3)
235
148
135
466
452
436
408
600
664
527
275
277
504
204
Brookhaven
(4)
244
130
135
464
502
386
411
599
671
523
260
328
484
-
Connetquot
River
Estimated
(5)
240
137
135
465
502
429
410
599
669
525
266
308
492
204
Connetquot
River
Observed
(6)
148
96
104
343
389
384
401
254
531
443
78
162
338
86
Ratio=
Connetquot River
Observed
Connetquot
River
Unobstructed
(7)
0.62
.70
.77
.74
.77
.90
.98
.42
.79
.84
.29
.53
.69
.42
Note 1 - Radiation data in column 5 are estimated unobstructed horizon values for Connetquot River
based on data from Mineola and Brookhaven (cols. 3,4).
Note 2 - Solar site 1 is typically heavily forested, solar site 2 is moderately to heavily forested,
and solar site 3 is moderately forested.
-------
Suppose the user is interested in predicting how the removal of
the riparian vegetative cover might effect E. Consider the period
22-24 August, 1967, when the Connetquot River received 162 langleys/
day of a possible 308 langleys/day of whortwave solar radiation.
Representative meteorological conditions at this time were:
T * 65°F
a
u = 2 mph
cloud cover fraction = 0.5
relative humidity = 80%
The steps in solving for E are as follows:
1. Data have been gathered, as previously listed.
2. H$n = 162 (3.7) = 600 BTU/ft2/day. This value
assumes that the vegetative canopy blocks 47% of the
solar radiation. From Table IV-23, H is
(.5 cloud cover at 65°F) 2497 BTU/ ft /day. Thus,
HR = 2497 + 600 - 3097 BTU/ ft2/day.
3. At 80% relative humidity and an air temperature
of 65°F, e =12.5 mmHg from Table IV-24.
a
4. As an initial guess of E, assume E-, = 65°F,
the air temperature.
5. From Table IV-25, B = .56, C(B) = -20.1
6. K = 15.7 + (.26 + .56) (11.4) (2) = 34.4
438
-------
7. F(E1 ) = -0.05(65)2 3098-1801 34.4-15.7
34.4 34.4 34.41. 26+. 56)
X [l2.5 + 20.1 + .26(65)1 - -6.1 + 37.7 + 33.0 = 64.6
8. E - .3(65) + .7(64.6) = 64.7
9. Since |E2-E1 <1°F
Now suppose the user wants to find E for no reduction in H due
to shading. Steps 1 through 9 again are repeated, using H =
r\ J M
308(3.7) - 1140 BTU/ft /day, with otherwise the same meteorological
conditions. Without detailing the calculations here, it is found that
F = 73.7°, a 9°F increase.
It is evident then that altering the solar radiation penetrating
to the stream can significantly change E. Even more severe cases of
repression of shortwave radiation (as noted by the 71% reduction on
26-28 August, 1967, Table IV-26) are possible, exemplifying the
large differences which may be observed.
END OF EXAMPLE IV-10
The approach illustrated in Example IV-10 for predicting equilibrium
temperature is obviously time consuming, and has been programmed for hand
held calculators in Mills eta/U (1979) . A simplified approach is also
available for predicting equilibrium temperature (Brady £t ji^., 1969) and is
described below. The predictions are usually within 3°F or less of those
found by the more complicated approach.
The data required for the simpler approach are:
t T,, dewpoint temperature (°F)
• U, mean daily wind speed (mph), and
• Hsn, net incoming shortwave radiation (Btu/ft2/day)
439
-------
Short wave solar radiation data were previously shown in Figure IV-27. The
climatic atlas (U.S. Department of Commerce, 1968) contains compilations of
dewpoint temperature and windspeed. Figures IV-28 and IV-29 show these data
for the months of July and August. Figures IV-27 through IV-29 provide the
user with all the data needed to predict equilibrium temperature using
Brady et^ a! 's approach.
To find the equilibrium^temperature the following equations are applied
sequentially:
COMPUTE ONCE
F(U) - 70 + 0.7 U2
ITERATE OVER
THESE EQUATIONS
(IV-61)
(IV-62)
B = 0,255 - 0.0085T + 0.000204T2 (IV-63)
K = 15.7 + (B + .26) F(U) (IV-64)
Ei+l = TD + Hs ' K (IV-65)
The wind speed function f(U) is found once from Equation IV-61. The
dewpoint temperature (Td) is a convenient starting choice as an initial
guess of the equilibrium temperature. T can then be calculated from
Equation IV-62; B from Equation IV-63; K from Equation IV-64; and finally
a new equilibrium temperature (E-j+i) from Equation IV-65. If
°
and E-+1
differ by more than 1F, return to Equation IV-62 with Ei+1 and repeat the
procedure until convergence is attained (usually within 2 or 3 cycles).
440
-------
504Q
JULY
55
65
•75
AUGUST
75--75
FIGURE IV-28 MEAN DEWPOI&T TEMPERATURE (°F) THROUGHOUT THE
UNITED STATES FOR JULY AND AUGUST (U,S, DEPART-
MENT OF COMMERCE, 1968)
441
-------
JULY
AUGUST
FIGURE IV-29 MEAN DAILY WIND SPEEDS (MPH) THROUGHOUT THE
UNITED STATES FOR JULY AND AUGUST (U,S, DEPARTMENT
OF COMMERCE, 1968)
442
-------
EXAMPLE IV-11
Equilibrium Temperature using Simplified Approach
Determine the average daily surface water equilibrium temperature for
Little Rock, Arkansas during the month of August. Based on Figures IV-27
through IV-29 the following data are found:
Td = 68°F
U = 7 mph
Hsn = (525)(3.7) = 1943 Btu/ft2/day
Assume as a first guess That E = Td = 68°F
then
f(U) = 70 + .7 (7)2 = 104.
T = (Td + Td)/2 = 68°
B = .62
K = 15.7 + (.62 + .26) (104) = 107.
E = 68 + 1943/107 = 86°F
For the second interation
T = (86 + 68)/2 = 77
B = 0.81
K = 127
E = 83.3°F
At the end of a third interation E = 83.7 F, so convergence has been
attained by three interations.
443
-------
As a comparison, the equilibrium temperature will also be calculated
using the longer approach. The required data are:
Tg = 80°F
T, = 68°F
Hsn = 1943
= 7.
= 19
sky cover = 0.5 (from climatic atlas)
A summary of the procedure is:
1. Han = 2958
HR = 1943 + 2958 = 4901
2. Since Td = 68°, e = 17.4
3. Choose E = T = 80°F
a
4. B = .881
C(B) = -37.6
5. f(U) = 70 + 0.7 (7)2 = 104
K = 15.7 + (0.26 + .881) (104) = 134
6. F(E) = 79.3
7. E = .3(80)+ .7 (79.3) = 80°F, after one pass.
Since the starting guess of 80°F is virtually identical with the calculated
value at step 7, a second interation is not required. The two procedures
predict equilibrium temperatures which differ by about 4°F.
END OF EXAMPLE IV-11
To estimate the effects of shading, the incoming solar radiation should
be calculated first assuming no shading, but otherwise using existing
meteorological conditions for the time of the year of interest. The effects
444
-------
of shading should be superimposed upon this result as a percent reduction.
The following (Pluhowski, 1968) can serve as guidelines in estimating solar
radiation reduction:
• 0-25 percent reduction: shading generally restricted to early
morning and late afternoon
• 25-50 percent reduction: some sunshine penetration in morning
and evening. Considerable sunshine between 1000 and 1400
hours.
• 50-75 percent reduction: very little sunshine penetration in
morning or late afternoon. Some sunshine between 1000 and
1400 hours.
t Greater than 75 percent reduction: very little penetration
even at noon.
4.4.4 Screening of Thermal Discharges
4.4.4.1 Introduction
This section presents a set of procedures which can be used to
determine whether the thermal discharge at a proposed power plant site or
the discharge from the expansion of an existing site is likely to violate
thermal standards. Procedures are presented to test for contravention of
the following types of standards:
• The AT Criterion: The increase in temperature of water
passing through the condenser must not exceed a specified
maximum.
• The Maximum Discharge Temperature Criterion: The temperature
of the heated effluent must not exceed a specified maximum.
445
-------
• The Thermal Block Criterion: The cross-sectional area of a
river occupied by temperatures greater than a specified value
must not exceed a specified percentage of the total area.
• The Surface Area Criterion: The surface area covered by
isotherms exceeding a specified temperature increment (above
ambient) must not exceed a specified maximum.
Actual values associated with the above standards vary by political
jurisdiction. Accordingly, regulations must be consulted.
The thermal discharge screening procedures are designed to address the
following questions:
• Is the power plant, as proposed, acceptable at the candidate
location?
t What is the largest power plant that can be placed at the
candidate location? Equivalently, can an existing power plant
at the candidate location be expanded?
The methods do not analyze interactions among multiple powerplants on
the same river. Such an analysis can be rather more complex. A report by
Tetra Tech (1978) addresses that question.
The methods developed to evaluate instream thermal criteria use heat
balance equations assuming a steady-state, well mixed system at low flow.
The power plants are assumed to employ once through cooling, as shown in
Figure IV-30.
446
-------
Intake Channel — —— -»*
^
WR
X,
-^
A
t
Qo/^
Qr ^
,
Ti
1
Power _ _ A_
Plant T0=Tj+AT
t
^___ ___ Outlet
*~~~ Channel
Skimmer Wall 1
1
River |A
PLAN VIEW
FIGURE IV-30 IDEALIZATION OF A RUN-OF-THE-RIVER
POWER PLANT
The selection of well mixed conditions appears to be justified.
Studies by Stefan and Gulliver (1978) on the Mississippi and Missouri Rivers
have dealt with the lateral mixing of thermal plumes which were released at
the shoreline and were not initially well mixed across the river. The
investigators found that over a short distance, thermal losses were
negligible and that the well mixed isotherm (the isotherm that would result
were the plume initially well mixed laterally and vertically) eventually
extended across nearly the entire width of the river, albeit at some
distance downstream. This indicates that if the thermal block criterion is
not met for the well mixed case, it is not likely to be met for the
shoreline discharge either. A similar conclusion can be reached regarding
the surface area constraint. Thus, at this level of analysis, it is not
necessary to consider the consequence of incomplete lateral or vertical
mixing adjacent to the shoreline discharge.
One simplification which can be used at the option of the user for the
surface area calculation should be mentioned. Surface water that is
undisturbed by anthropogenic influences (in a thermal sense) approaches the
equilibrium temperature. This temperature is dictated by natural
meteorological conditions. Surface water temperature in rivers, especially
during steady low-flow periods, can be near equilibrium. In calculating the
447
-------
surface area occupied by isotherms exceeding a specified temperature it is
necessary to know the equilibrium temperature. However, since the procedure
for calculating equilibrium temperature is fairly complicated, considerable
savings in computational effort can be obtained by assuming the ambient
water is at its equilibrium temperature.
Some circumstances, in addition to anthropogenic influences, tend to
produce ambient temperatures different from equilibrium. For example,
• Locally, large quantities of groundwater may discharge into
the river.
• Hypolimnionic releases from large reservoirs may occur nearby.
• Snow melt may supply a substantial amount of inflow.
As a result of the first two influences, the stream water temperature may be
lower than equilibrium since the source of the water comprising the stream
flow has been shielded from the heating effect of solar radiation. Snow
melt, although not likely to influence the river's thermal regime during the
late summer, can be important through spring and into early summer in areas
where high-mountain snowpack exists over most, or all, of the year.
The screening procedure that follows assumes the river water, once it
has been heated by the thermal plume, is above equilibrium. This means that
the water temperature will then decrease in the downstream direction, which
is generally, but not always, true.
Table IV-27 shows the data needed to apply the thermal screening
methods. The symbols are defined in the table and suggested default values
are given for variables where appropriate. The variables are introduced in
the table in the order they occur in the screening procedure.
448
-------
TABLE IV-27
DATA NEEDED FOR THERMAL DISCHARGE SCREENING
Variable
Term Definition
Default Value
Mwe
AT,
max!
AT
max2
maxmin
Capacity of power plant in
megawatts electric (bus bar)
Percent of total energy pro-
duced that goes to electricity
production
Percent of total energy produced
that is dissipated through the
cooling water
River-flow rate above power
plant (nr/s)
Mass density of water
(kg/mi)
Specific heat of water
(Btu/DF-kg)
Temperature rise in the river
cross section that constitutes
a thermal block (°F)
Maximum legal allowable tempera-
ture rise across the condenser (°F)
Maximum allowable temperature
rise across the condenser such
that T < (T ) (°F)
e v e'max v '
Temperature of heated effluent (°F)
Maximum legal allowable tempera-
ture of heated effluent (°F)
The lesser of ATm , and
«»»* <°F>
The maximum allowable flow rate
through the cooling system
(m3/s)
new fossil fuel
plants:38
nuclear plants:32
new fossil fuel
plants:48
nuclear plants:68
1000
2.2
5
20
.25Q,
449
-------
TABLE IV-27 (continued)
Variable
Term Definition
Default Value
AT
V
d
E
K
sa
A
sa
W
ra
The isotherm defining the boundary
of the surface area for which legal
limits have been established (°F)
Mean velocity of the river
(ra/s)
Mean hydraulic depth of river in
reach under consideration (m)
Equilibrium temperature l°F)
Surface thermal transfer coeffi-
cient (Btu/d • °F • m2)
Surface area of river down to AT
isotherm (m2)
Legal maximum surface area limit
which can be covered by the AT
and greater isotherms (m2)
Average surface width of river
down to AT,, isotherm (m)
Sa
River temperature just above where
a tributary joins the mainstern
Temperature of tributary (°F)
Flow rate of tributary (m^/s)
Air Temperature (°F)
Relative
humidity
sn
an
Wind speed at 7 meters above
surface (m/s)
Net shortwave solar radiation
(Btu/m2 • d)
Net long wave solar radiation
(Btu/m2 • d)
450
-------
4.4.4.2 Evaluating the Thermal Block Criterion
The initial temperature elevation that results when the thermal plume
becomes well mixed with the river water is given as:
T (IV-66)
= J_ . !c MW . J_ . 3.414 x 106 (IV-67)
Qr epMWe PCp —mt
where
ATwm = temperature elevation of the initially well mixed
isotherm ( F)
Q = flowrate of cooling water (m3/s)
AT = Te - Tr (°F)
T£ = temperature of heated effluent (°F)
Tr = temperature of river water upstream of power plant (°F)
All other terms are defined in Table IV-27. To find AT , Equation IV-67 is
wm
solved. If ATwm is less than the thermal block temperature increment
(ATtb), the thermal block criterion is not contravened. Otherwise, it is.
4.4.4.3 Acceptability of the Temperature Rise Across the Condenser
and of the Temperature of the Heated Effluent
Whether these criteria are met or not depends on a number of factors,
such as the cooling water flow rate. Since the cooling water flow rate can
be designed to be within a specified range, it is determined here whether a
feasible range exists such that the two above mentioned criteria are met.
The minimum acceptable flow rate such that both temperature criteria do
not exceed their standards is as follows:
m \ 6c MU« 1 3.414 x 106
-------
where
(Q ) . = minimum flow rate such that the two temperature
p mm
criteria are not exceeded (m3/s)
By evaluating Equation IV-68 the minimum cooling water flow can be
determined.
As an example of how AT . is chosen, suppose the following
maxnnn
conditions exist:
• The maximum legal temperature rise across the condenser is
20°F.
• The maximum legal temperature of the heated effluent is 86°F.
t The ambient river temperature is 74 F.
From these conditions, ATmax2 (the allowable temperature increase across
the condenser such that the temperature of the effluent does not exceed the
legal maximum) = 86°F - 74°F = 12°F. So AT. = minimum (20°F, 12°F) =
11 id AIM III
12°F. 12°F must be chosen, then, as the maximal temperature rise across the
condenser.
Once Equation IV-68 has been solved, the ratio of cooling water to
river flow should be checked so that the value is within acceptable limits.
Equation IV-66 can be rewritten as:
Q« AT
JB. = wm (IV-69)
Qr AT
Since AT has been calculated from Equation IV-67 and AT has been
wm
calculated as AT . , the flow rate fraction can be calculated from
Equation IV-69. If this fraction exceeds a certain percent (e.g. 25 percent
or some user defined value), then the cooling water flow rate is too large
to be acceptable. If the flow rate fraction is not excessive, the actual
flow rate can be chosen so that:
452
-------
(Q )m.n ±Q ± (Qp)max (IV-70)
where
(Q )max = maximum allowable cooling water flow rate (m3/s)
4.4.4.4 Evaluating the Surface Area Constraint
The evaluation of this criterion may require the user to perform
considerably more calculations than for any of the other prescreening
criteria. The two major complicating factors that are encountered are:
1. determining the river equilibrium temperature, and 2. evaluating the
effects of tributaries.
If it is the case that ATwm does not exceed ATsa the surface area
criterion will not be contravened and no calculations have to be performed.
If ATwm exceeds ATsa, the criterion might be exceeded. In this case it is
necessary to determine the distance from the location of the thermal
discharge to the downstream location of the AT$a isotherm. This distance is
given by:
-pC Vd /T - E\
x = —^__ in ( T _ E 1 24- 3600 (IV-71)
\ wm /
where
Tsa = ATsa + Tr
wm ~ wm r
Section 4.4.3 discusses procedures for predicting K and E. Once K and E are
found, xsa can be determined from Equation IV-71. If one or more tributaries
exist with the distance x,. then x „ should be recalculated as discussed
sa sa
in Section 4.4.4.5.
The surface area included within this reach is:
A = xsa ' W (IV-72)
453
-------
where
A = surface area of the river from the point of thermal
discharge to x (m2)
W = average river width in this reach (m)
If A < A then the surface area criterion is not contravened. Otherwise,
sa
it is.
4.4.4.5 Evaluating the Effects of a Tributary in Mitigating Temperature
Within a Thermal Plume
Tributaries, when they join a river subjected to the influences of a
thermal plume, generally act to reduce the elevated river temperature. They
may therefore prevent the surface area constraint from being exceeded when
it otherwise would.
Equation IV-71 assumes no tributaries exist throughout the reach
defined by x . If it is found that x > x, (x . is defined below under
S cl S a L L
Equation IV-73) then it is necessary to examine the impact of the tributary
flow on the surface area constraint. This is done by computing the water
temperature (°F) just above the location where the tributary joins the
mainstream using the following equation:
= (Twm " E) exP (^C"Vd ~*ll • 360o)
Tra
where
T = river temperature just upstream of tributary ( F)
1T3
x = distance from power plant discharge to tributary (m)
After the river has mixed with the tributary the new river temperature ( F)
is given by:
(r\ - TraQr + TtQt (IV-74)
V rl new 0+0,.
454
-------
where
T. = temperature of the tributary ( F)
Qt = flow rate of tributary (m3/s)
If
I < AT + T (IV-7b)
•i new — sa ra v
then this location marks the downstream location of the ATga isotherm and
the surface area A can be calculated using the distance x$a as the distance
down to the tributary, xt. Otherwise the ATsa isotherm is located further
downstream. In this case Equation IV-71 is reapplied (first making
appropriate adjustments to V and d) where the initial temperature is
(Tr)new (which was Twm in Equation IV-71) and the final temperature is still
T ,. The distance xpa is determined by adding this additional distance to
S a S a
xt-
4.4.4.6 Determining Whether the Thermal Block or the Surface Area
Constraint is the More Limiting
One of these two constraints may cause a greater limitation on power
plant size than the other. If ATtb < AT$a the thermal block constraint will
be more limiting, and there is no need to continue with the analysis in this
part. If, however, ATtb > ATsa, the surface area constraint may be more
limiting. To determine if it is, find ATwm (call it ATwmsa) using the
following equation:
I ' \ I "KXsa \
ATwmsa = E + Tsa - E exP ,X-Vd^-24T3600 " Tr (IV-76)
\ / \ P /
where
T$a - AT + T (IV-77)
and
A
x = -ii (IV-78)
455
-------
If a tributary exists in the reach delineated by x , recompute x as
S3 S d
outlined in Section 4.4.4.5.
If ATwmsg < ATtb, the surface area constraint is more restrictive, so
set ATwm = ATwmsa- Otherwise set ATwm = ATtb.
4.4.4.7 Determining the Maximum Plant Capacity
The maximum power plant capacity can be determined based upon the
maximum well mixed temperature elevation and the river flow rate. It is
given by:
)max = e ' pCo ' (VT)max - — (IV'79)
max ec p p max ^414 x 1Q6
= -£ • pC -AT Q • ---- 360° (IV-80)
6c P ™r 3.4.4 x 10*
By using Equation IV-80 and the maximum allowable AT , the maximum capacity
Will
can be found.
4.4.4.8 Readjusting the Maximum Cooling Water Flow Rate
If the minimum acceptable flow rate is greater than the maximum
allowable, the power plant size must be reduced. To do this, set
where
Qp = actual cooling water flow rate (m3/s)
(Q )max = maximum allowable cooling water flow rate (m3/s)
AT is recalculated by:
WIN
(Note: the surface area and thermal block constraints are still met and
need not be recomputed.)
456
-------
AT = AT . Rmax (iv-82)
wm maxmin
where
AT . is the same AT calculated earlier.
maxim n
EXAMPLE IV-12
Estimating AT Across a Power Plant Heat Exchange Unit
Suppose the user wants to determine AT for the Hartford Electric Light
Company's South Meadow Steam Electric Power Plant (a fossil fuel plant)
located on the Connecticut River. Data available are (Jones et aj_. , 1975):
capacity ................... 217 MW
cooling water flow rate ........... 341 ftVsec
waste heat discharged to cooling water .... 422 MW
Since the waste heat being dissipated through the cooling water is known, AT
can be calculated directly using that value in conjunction with the known
flow rate. Assume, however, that the waste heat being discharged is not
known. It can be estimated from the plant capacity as follows. First,
assume the plant efficiency is 33 percent. The rate at which fuel is burned
when at capacity is then:
= 658 MW
If 10 percent of the total energy is lost up the stacks, then approximately
58 percent is dissipated through the cooling water, or
658 (.58) = 382 MW
Compared with the known 422 MW of heat discharged to the cooling water, the
above calculation would underestimate AT.
457
-------
AT is calculated by
AT thermal loading rate to coolin£j^aj^-jjjrg3awa.tts.
ft 4i/n/in6 BTU Vl hr
hr73600
where
YC =62.4 BTU/ft3/°F
Q0 = flow rate, ftVsec
Substituting the appropriate values into the above equation, it is found
that (using the known thermal loading to the cooling water):
> (3.414) (106)/3600 _
L - . -
AT -
AT
Equation IV-83 is not feasible to use when the thermal loading rate to the
cooling water is unknown. As an alternative approach, the following
expression can be employed:
AT = -_L . _^. Mue . ' . J.tiH x ID (TV-84}
Qo ep PCp 3600- (IV84)
where
e = percent of total energy produced that is transmitted as
electricity. For new fossil fuel plants: 38 percent: for
nuclear plants: 32 percent.
e_ = percent of total energy produced that is dissipated through
\f
cooling water. For new fossil fuel plants: 48 percent;
for nuclear plants: 68 percent
MWe = capacity of power plant in megawatts electric
Equation IV-84 predicts that AT is
1 . 58 .017 ' . 3 • 414 10 _ -17 cor
341 32 62.4 3600
458
-------
AT is only about 1°F less than predicted by Equation IV-83.
END OF EXAMPLE IV-12
4.4.5 Longitudinal Temperature Variation
If the temperature at a particular location in a river is known, the
steady-state temperature distribution downstream from that point can be
estimated by:
T - E /-.061 • Kx \ /TW oc^
=! V = exo I r^-rn (Iv-85)
where
T = temperature at x = 0, °F
m .
T = stream temperature at a distance x, where x is measured in
miles
E = equilibrium temperature, F
K = thermal transfer coefficient, BTU/ft2/day/°F
U = stream velocity, ft/sec
d = stream depth, feet
p = water density, lb/ft3
Cp = heat capacity of water, BTU/lb/°F (pCp = 62.4 BTU/ftV°F)
An important fact is revealed upon inspection of Equation IV-85.
Suppose that a thermal discharge heats the ambient water to a temperature
Tm, but Tm is less than the instantaneous equilibrium temperature E. In
that instance the stream temperature will continue to rise exponentially
downstream, approaching E. The rate at which T approaches E is dependent on
the thermal transfer coefficient, as well as stream velocity and depth.
Equation IV-66 is graphically illustrated in Figure IV-31.
459
-------
r -
Kx
pcpdU
FIGURE IV-31 DOWNSTREAM TEMPERATURE PROFILE FOR COMPLETELY
MIXED STREAM., T-E/T -E vs, r (FROM EDINGER,
1965)
460
-------
EXAMPLE IV-13
Use of Figure IV-31
Suppose an average daily thermal transfer coefficient, K, of 200
BTU/ft2/day has been calculated. The river of interest has an initial
temperature "excess" (i.e. Tm-E>0). How far downstream will that excess be
50 percent of the original? Other stream data:
U = .5 fps
d = 4 feet
pC = 62.4 BTU/ft3/°F
From Figure IV-31, r is to be found such that
V E
The correct r equals 0.68. Solving for x in terms of r it is found:
roC dU (0.68) (62.4) (4) (.5)(24)(3600)
x =
200
= 3.6 x 104 feet = 6.9 miles
The associated travel time is T = U>-XlQ_ x I hr = 2o.4 hours
.b jouU
END OF EXAMPLE IV-13
4.4.6 Diurnal Temperature Variation
Although it is beyond the scope of this report to analyze diurnal
stream temperature variations, a few brief statements should be made.
Diurnal stream temperature variations on Long Island, New York, were
mentioned in Section 4.4.1. Documentation of large diurnal temperature
variations is not limited to New York. For example, studies in Oregon
(Brown, 1969), Hawaii (Hathaway, 1978) and California (Mills, 1979) have
revealed that solar radiation entering shallow streams and rivers produces a
461
-------
significant difference between maximum and minimum daily temperatures.
Figure IV-32 shows one such example on the Santa Ana River near Mentone,
California (Mills, 1979). The water temperature varied by 17°F over a
period of 24 hours. One significant effect of the temperature variation is
its effect on dissolved oxygen levels. Figure IV-33 shows the measured
dissolved oxygen concentrations and predicted saturation levels over the
same time period at the same location on the Santa Ana River. The dissolved
oxygen concentrations ranged from a high of 9.2 mg/1 to a low of 8.0 mg/1.
The variations were caused predominantly by the temperature changes. This
illustrates several points.
• Temperature data concomitant with dissolved oxygen data might
be needed to properly interpret the cause of dissolved oxygen
variations in shallow rivers receiving large amounts of solar
radiation.
t Removing riparian vegetation around shallow rivers tends to
increase the daily maximum temperature and decrease the daily
minimum temperature.
• Impacts on the dissolved oxygen levels and indigenous biota
can be significant.
4.4.7 Low Flow and Temperature
Evidence has previously been cited in this chapter to show that in many
parts of the country high temperature conditions are concomitant with low
flow. The planner needs to be able to quantify better the nebulous term
"low flow" to fruitfully use this concept as a planning tool. For example,
suppose a decision is made based on the low flow condition of this year.
What are the chances that this low flow will be exceeded in the future? If
they are high, then any decision (e.g. at particular level of waste
abatement at a sewage treatment plant) based on the observed conditions
could have unexpected deleterious results at a future time. It is paramount
then, to predict how often flow will fall below a specified rate.
462
-------
o
Ul
CC
cc
Ui
a.
5
HI
h-
90
85
80
75
70
65
60
55
50
1
— I I 1 I
KEY
Air Temperature
Water Temperature
1040
6/19
_L
_L
1440
1840
2240
0240
6/20
TIME OF DAY (Military Time)
0640
1040
FIGURE IV-32
MEASURED AIR AND WATER TEMPERATURES FOR
THE SANTA ANA RIVER NEAR MENTONE, CALIFORNIA,
IN JUNE 1979,
463
-------
10.0
9.8
9.6
9.4
9.2
KEY
Observed DO
Saturation, Cs
1040 1440 1840 2240 0240
6/1f 6/20
TIME OF DAY (Military Time)
0640
1040
FIGURE IV-33
MEASURED DISSOLVED OXYGEN CONCENTRATION
AND PREDICTED SATURATION CONCENTRATION FOR
THE SANTA ANA RIVER NEAR MENTONE, CALIFORNIA,
IN JUNE 1979,
464
-------
Two measures or indices of low flow that have been found useful are
flow duration and low-flow frequency. Although it is beyond the scope of
this report to explain in detail how to develop these measures, examples of
each will be presented that explain their utility. The majority of the
material in this section is from Cragwall (1966) who provides a discussion
on low flow, and cites additional references. Many texts on engineering
hydrology (e.g. Linsley et al_., 1958) also discuss low flow. Figure IV-34
shows a flow duration curve for the Hatchie River at Bolivar, Tennessee.
The vertical axis is the daily discharge and the horizontal is the percent
of time a flow is equaled or exceeded. For example, 95 percent of the time
from 1930-58 the flow exceeded 177 cfs. It can also be assumed that this
flow (177 cfs) will probably be exceeded 95 percent of the time in other
years. Thus this concept offers one means by which to quantify "low flow".
A second concept is the low flow frequency curve, illustrated in Figure
IV-35. This depicts the relationship between discharge and recurrence
interval of different duration flows. For example the 7 day mean flow of
100 cfs can be expected to occur once each 19 years. Stated another way,
since probability is the reciprocal of recurrence interval, in any one year
there is about a 5 percent probability that a seven day mean flow of less
than 100 cfs will occur. A commonly used flow for analyses is the 7 day
mean flow at a recurrence interval of 10 years, or 7Qio.
4.4.8 Interrelationships Between Temperature Prediction Tools
The three major temperature prediction tools presented in Section 4.4
are:
* water temperature alterations caused by a power plant
• equilibrium temperature
• longitudinal river temperature profile
Figure IV-36 shows three river temperature profiles which illustrate how
these tools can be used jointly. Curve A represents a temperature profile
of a river where a power plant is located a distance D below some reference
point. The temperature on the river above the power plant is T2 which is
465
-------
DAILY DISCHARGE (cfs)
cn
CD
cr
73
m
m
<-> z
?3
> (—)
cn cr
2: 73
> <
r~ m
h-1 >
73
-------
10000
120 Day
60 Day
30 Day
a
100
RECURRENCE INTERVAL (YEARS)
FIGURE IV-35
FREQUENCY OF LOWEST MEAN DISCHARGES OF
INDICATED DURATION, HATCHIE RIVER AT
BOLIVAR, TENN, (FROM CRAGWALL, 1966)
467
-------
Stream
Temperature
A (Power plant present)
B (Release from hypolimmon)
D
DISTANCE
FIGURE IV-36 THREE RIVER TEMPERATURE PROFILES
468
-------
slightly below the equilibrium temperature. Due to the thermal discharge
from the power plant, the river's temperature is increased to "!\, above the
equilibrium temperature. Below the mixing zone area, the water temperature
gradually decreases toward equilibrium, as the excess heat is dissipated
into the atmosphere.
Curve B illustrates the temperature profile of a river whose water
comes predominantly from the hypolimnion of a reservoir. While in the
reservoir the water is insolated from the solar radiation, so the
temperature is below the equilibrium temperature. As the water is withdrawn
from the reservoir and begins to flow downstream, its temperature increases
due to solar radiation and atmospheric heating. The temperature tends to
approach the same equilibrium temperature (the two rivers are assumed to be
in the same geographic area).
Curve C shows the temperature profile of river B which now has a power
plant, similar to the one on river A, discharging into it. If the flow
rates of the two rivers are the same, so is the initial temperature increase
(i.e. T3 - Ti = TV T2). However, the temperature of the river continues
to increase, in contrast to profile A, because T3 is less than E. This
illustrates an unusual, but entirely possible, situation where river
temperature continues to increase below a thermal discharge.
4.5 NUTRIENTS AND EUTROPHICATION POTENTIAL
4.5.1 Introduction
Within the past decade the elements most often responsible for
accelerating eutrophication - nitrogen and phosphorus - have shown generally
increasing levels in rivers (EPA, 1974). Median concentrations increased in
the period from 1968 to 1972 over the period from 1963 to 1967 in 82 percent
of the reaches sampled for total phosphorus, 74 percent for nitrate, and 56
percent for total phosphate.
469
-------
These increasing concentrations afford more favorable conditions for
eutrophication, although many rivers with high nutrient levels do not have
algal blooms. Algal growth can be inhibited in numerous ways. For example,
turbidity can decrease light transmittance through water and effectively
stop growth. Decreasing turbidity could, however, have a deleterious side
effect of promoting excessive algal growth, unless stream nutrient levels
are concurrently decreased. High water velocity can also prevent algae from
reaching bloom proportions before they are carried out of the river system.
The eutrophication problem, then, is transferred to the water body into
which the river empties.
4.5.2 Basic Theory
Stumm and Morgan (1970) have proposed a representation for the
stoichiometry of algal growth:
106CO + 16NO" + HPO 2" + 122H 0 + 18H+( + trace
2 3 ** 2
elements; energy)
(IV-86)
K H 0 N P I + 138 0
'106263110161' 2
algal protoplasm
where P and R represent photosynthesis and respiration, respectively.
Observe that in the algal protoplasm the ratio of C:N:P is
C:N:P = 106:16:1, by atomic ratios (IV-87)
C:N:P = 41:7:1, by weight ratios (IV-88)
470
-------
From the above two equations it can be inferred that only small
amounts of phosphorus are needed to support algal growth in relation
to the amounts of carbon and nitrogen required. If phosphorus is not
present in the amount required for algal growth then algal production
will be curtailed, regardless of how much of the other nutrients is
available. Phosphorus is then termed growth limiting. It is possible
for other elements, particularly nitrogen, and occasionally carbon or
trace metals, to be growth limiting as well (Stumm and Sturnm-Zol linger,
1972).
Nitrogen uptake by algae is generally in the nitrate form if
nitrate is available. However, different types of fresh water algae
can utilize either organic nitrogen or inorganic nitrogen in the form of
ammonia, depending on what is available (Stumrn and Stumm-Zoll inger,
1972). Algae typically require phosphorus in an inorganic form,
usually as orthophosphate ion (Kormondy, 1969).
Some indication of whether nitrogen or phosphorus is growth
limiting may be made by determining the weight ratio of the appro-
priate forms of nitrogen and phosphorus found in a river, and
comparing that with the stoichiometric ratio required for growth.
This gives an idea regarding the nutrient on which control efforts
should focus. Specifically, let
] (IV-89)
where
[TN] = concentration of total nitrogen in river, mg-N/1
[OPCL-P] = concentration of orthophosphate, mg-P/1
If R>10, phosphorus is more likely to limit than N.
If R<5, nitrogen is more likely limiting than P.
If 5
-------
Since the N:P ratio in algal biomass can vary from species to species,
this makes the determination of the limiting nutrient somewhat uncer-
tain, and leads to the indeterminate range of 5
-------
For example, algae utilize nutrients, die, and settle to the bottom.
Although there is a recycling of algal cell-bound nutrients, the
settling rate may surpass the rate of recycling. Assuming total
nitrogen and total phosphorus to be conservative should give an
estimate of the upper limit of the instream concentrations of these
nutrients.
The instream concentration of total nitrogen (TN) or total
phosphorus (TP) resulting from a point discharge is (formulas will be
presented for TN only; those for TP are exactly analogous):
(IV_90a)
or
S. 38 (I
Qu + Qw
where
TNu = instream TN upstream of discharge. mg-N/1
TNw = concentration of TN in point discharge, mg-N/1
Qu = flow in river upstream of point discharge, cfs
Qw = flow rate of point discharge, cfs
TNQ - resulting instream TN concentration, mg-N/1
wp = loading rate of point source, Ib/day
The expression for TNQ is given by either Equation IV-9iA or IV-91B
The appropriate form to use will depend on the form of the available
data.
473
-------
To determine the instream concentration of total nitrogen due to a
distributed discharge, use:
j X
TN = TNQ + -5- (TNr - TNQ) (IV-91a)
or
TN Q
(IV-915)
Q 5.38 Q
where
TN = TN entering with the distributed flow, mg-N/1
TNQ = instream TN at x = 0, mg-N/1
x = distance downstream from the point source discharge
Q = stream flow rate at x, cfs
Q = stream flow rate at x = 0, cfs
Ag = incremental flow increase per unit distance, cfs/mile
w = mass flux of TN entering the stream through the
distributed source, Ib/day/mile
The choice of whether to use Equation IV-91a of IV-91b depends on the
available data. Based on the approach detailed in Chapter III, the mass
flux of nutrient entering the stream (in units of Ib/day/mile) can be
generated. When this approach is used, then Equation IV-91b is applicable.
To use Equation IV-91a the concentration of pollutant from the nonpoint
source has to be known. This can be accomplished using the approach of
Omernik (1977). Nonpoint source nitrogen and phosphorus concentrations are
predicted as fractions of land use type or based on color coded maps if land
use categories are not known. The data used to predict nitrogen and
phosphorus concentrations were generated in a National Eutrophication Survey
(NES) program wherein a nationwide network of 928 nonpoint-source watersheds
were monitored. This method accounts for only the nonpoint source
contribution. Consequently, if point source exist within the watershed,
their contributions must be included as well in order to accurately predict
instream concentrations.
474
-------
Table IV-29 summarizes the predictive formulas developed by Omernik for
total phosphorus, orthophosphorus, total nitrogen, and inorganic nitrogen.
The formulas are regionalized by eastern, central, and western United
States. Agricultural, urban, and forested lands comprise the independent
variables in the formulas.
Omernik's analysis of the NES data indicates that:
1. Streams draining agricultural watersheds had considerably
higher nutrient concentrations than those draining forested
watersheds.
2. Nutrient concentrations were generally directly proportional
to the percent of the land in agriculture and inversely
proportional to the percent of land in forest.
3. Mean concentrations of total phosphorus and total nitrogen
were nearly nine times greater in streams draining
agricultural lands than in streams draining forested lands.
4. Mean phosphorus concentrations in streams draining forested
watersheds in the west were generally twice as high as those
in the east.
5. Total and inorganic nitrogen in streams draining agricultural
watersheds were considerably higher in the heart of the corn
belt than elsewhere.
As an alternative to the equations shown in Table IV-29, Omernik
provides three colored maps of nonpoint source related concentrations of
nutrients in streams. They can be used where detailed information necessary
for more accurate prediction is unavailable.
475
-------
TABLE IV-29
REGIONAL STREAM NUTRIENT CONCENTRATION PREDICTIVE MODELS
Nutrient Form Model, Correlation Coefficient and Multiplicative Standard
Region Error
Total phosphorus
East Log1Q (PCONC) = -1.8364 + 0.00971 (% agric + % urb)
r = 0.74, f = 1.85
Central Log1£) (PCONC) =-1.5697 + 0.00811 (% agric * % urb) -0.002312 (X for)
r = 0.70, f = 2.05
West Log^ (PCONC) =-1.1504 + 0.00460 (%agric + %urb) -0.00632 (% for)
r = 0.70, f = 1.91
Orthophosphorus
East L°9io (°PCONC) = -2-2219 + 0.00934 (% agric + % urb)
r = 0.73, f = 1.86
Central Log]Q (OPCONC) = -2.0815 + 0.00868 (% agric + % urb)
r = 0.63, f = 2.05
West Log10 (OPCONC) = -1.5513 + 0.00510 (% agric + % urb) -0.00476 (% for)
r = 0.64, f = 1.91
Total nitrogen
East Log]0 (NCONC) = -0.08557 + 0.00716 (% agric + % urb) -0.00227 (% for)
r = 0.85, f = 1.51
Central Log]Q (NCONC) = -0.01609 + 0.00399 (% agric + % urb) -0.00306 (% for)
r = 0.77, f = 1.50
West L°9io (NCONC) = -0.03665 + 0.00425 (% agric + X urb) -0.00376 (% for)
r = 0.61, f = 1.75
East Log1Q (1IICONC) =-0.3479 + 0.00858 (% agric + % urb) -0.00584 (% for)
r = 0.84, f = 1.93
Central Log.. (INCONC) = -0.5219 + 0.00482 (~ agric + % urb) -0.00572 (% for)
r = 0.71,f = 2.06
West Lo9in UNCONC> = -°-6339 ^ 0.00789 (% agric + X urb) -0.00657 (% for)
r = 0.65, f = 2.45
From: Omernik (1977)
476
-------
4.5.4 Nutrient Accounting System
It may be desirable to determine the impact of each nutrient source on
the total instream concentration in order to distinguish among the major
sources. An accounting procedure utilizing Equations IV-90 and IV-91 can be
developed to do this. The following steps outline the procedure.
1. Segment River. Divide the river into major segments. These
segment divisions may reflect waste loading distributions or
another convenient division scheme chosen at the discretion of
the planner. The segments are not necessarily the same as the
reaches that have previously been discussed (see Section 4.1).
The delineation of reaches as described earlier is based upon
lengths of river having uniform hydraulic conditions.
Segments, as used here, are purely a convenient subdivision of
the river.
2. Quantify and Locate Sources of Nutrients. The quantification
of point, nonpoint, and natural sources on the mainstem and
tributaries should be accomplished using1 the best available
data. Tabulation can be performed for each different season
to reflect the discharge pattern characteristic of each
season. The quantification should include total nitrogen and
total phosphorus. Tabulate data in terms of average daily
input (Ib/day). Characterize the location of the nutrient
sources by river mile. For nonpoint sources characterize by
river mile at both the beginning and end of the source.
3. Perform Mass-Balance. Sum the known sources to determine the
total nutrient loading to each segment. Then make the
following comparisons:
a. Compare the total loading with the nutrient input from the
mainstem at the upstream end of the segment. This direct
comparison permits an assessment of the collective impact
of the nutrient sources entering a segment and the
upstream contribution of the mainstem.
477
-------
b. Perform an intersource comparison to ascertain the
relative impact of each nutrient source. Express the
results for each source as a percent of the total loading.
When a tributary has a high percent contribution steps 1 through 3
can be repeated for the tributary itself to track the sources of the
nutrients.
Apply Equations IV-90 and IV-91 to each reach within the segment
to determine the instream nutrient concentration throughout the
segment. Once this is done this step can be repeated for the next
reach.
By applying this analysis one can determine the relative impact of
any discharge, determined jointly by the flux of the nutrient and the
discharge location. Section 4.1.10 provided a detailed example problem
which illustrates the procedure. A brief example also follows.
EXAMPLE IV-14
Computing Total Nitrogen Distribution
This example illustrates the use of Equations IV-90b and IV-91b in
calculating the total nitrogen distribution in a river. Suppose the
user has been able to estimate the point and nonpoint loading of total
nitrogen in a river as shown in Table IV-30.
478
-------
TABLE IV-30
TOTAL NITROGEN DISTRIBUTION IN A RIVER IN
RESPONSE TO POINT AND NON-POINT SOURCE LOADING
Reach
Number
1
2
3
4
River
Mile-
Point
0
9.99
10.0
14.99
15.0
20.99
21.0
26.0
TN
Added*
(Ibs/day)
400 L
500 D
0
700 D
800 L
650 D
0
900 D
TN
Cumul ati ve
(Ibs/day)
400
900
900
1,600
2,400
3,050
3,050
3,950
Q
Cumulative
(cfs)
300
400
400
600
700
900
900
1 ,000
TN Concen-
tration
(mg-N/1)
0.25
0.42
0.42
0.50
0.64
0.62
0.62
0.73
*"L" indicates a localized or point source. "D" indicates a diffuse
or non-point source whose range of input is over the entire reach.
If these loading rates are estimated over a year, then the flow rates
used should also be average annual flows. To compute the concentration
at mile 0, Equation IV-90b can be used:
(0)(Qu) , 400
TN
300_
1.55
0.25 mg-N/1
where the following conversions were used:
1 MGD = 1.55 cfs
1 mg/l = 8.34 Ib/MG
To determine the concentration at milepoint 9.99, use Equation IV-91b:
500
TN = (0.25) 300
__
8". 34
400
1.55
- 0.42 mg-N/1
479
-------
Note that wx in Equation IV-91 is the 500 Ib/day shown in Table IV-30. By
reapplying these two basic equations for each reach the user can work
downstream through the four reaches. Also note that the total nitrogen
concentration has decreased slightly through reach 3, even though more TN
has been added. This is because the incoming flow has served to lower the
concentration by dilution.
END OF EXAMPLE IV-14
4.6 TOTAL COLIFORM BACTERIA
4.6.1 Introduction
Total coliform bacteria are considered an indicator of the presence of
pathogenic organisms, and as such relate to the potential for public health
problems. Allowable levels of total coliform bacteria in rivers vary from
state to state and according to the water use description characterizing the
particular river segment. For example, in Montana (Montana State Dept. of
Health and Environmental Sciences, 1973) the raw water supply may not have
more than an average of 50 MPN/100 ml* total coliforms if it is to be used
as a potable water supply following simple disinfection. In water suitable
for bathing, swimming and recreation, as well as growth and marginal
propagation of salmonid fishes, an average of 1,000 MPN/100 ml is allowable.
Concentrations of total coliforms vary with the season of the year.
Often the heaviest loadings occur during the summer months, but this impact
is somewhat offset due to the more rapid die-off at higher temperatures and
more intense solar radiation. In the Willamette River (Figure IV-37), for
example, the highest counts of 1971-72 were actually observed from November
through May (EPA, 1974).
Treated municipal sewage comprises a major source of coliform
pollution. Urban stormwater runoff can also be significant, especially
*MPN means "Most Probable Number". Coliform organisms are not counted
individually, but their densities are statistically determined and the
results stated as MPN/100 ml.
480
-------
SEASONAL RIVER PROFILES
WILLAMETTE RIVER
Total Conforms
100,000
I
§ 10,000
L.
O)
•»
o 100
«*-
"o
o
_ 10
K
'(
100,000
E
o
2 10,000
0.
CO
E
o
±: 100
o
o 10
1
JUN. TO OCT. 1972
\ A / /\ ^Xl OREGON
A/\/ / \ / SIANUAHU?
A/I / /^^\V ^ A
vr^
V \
1 1 1 1 I 1 1 1 1 J
) 20 40 60 80 100 120 140 160 180 200
River Miles
NOV. TO MAY 1971 /I972
/\
-- x" — '" \ A
1 \ \ \/ ^
OREGON \ ^ ~-\ ^«^
STANDARD -* ^-^
1 1 1 1 1 1 1 1 1 1
LEGEND:
MEDIAN
85%
0 20 40 60 80 100 120 140 160 180 200
River Miles
FIGURE IV-37 TOTAL COLIFORM PROFILES FOR
THE WILLAMETTE RIVER (EPA, 1974)
481
-------
through combined sewer outflows. Rural storm water runoff transports
significant fecal contamination from livestock pastures, poultry and pig
feeding pens, and feedlots. Wildlife both within refuges and in the wilds
can contribute as well. For guidance in the interpretation of preliminary
coliform analyses, Table IV-31 can be used.
TABLE IV-31
TOTAL COLIFORM ANALYSIS (EPA, 1976)
If the Calculated
Concentration is:
Less than 100/100 ml
Less than 1 ,000/100 ml
More than 1,000/100 ml
More than 10,000/100 ml
Probabil ity of
a Col iform Problem
Improbable
Possible
Probable
Highly Probable
4.6.2 Mass Balance for Total Col iforms
The mass balance equations applicable to total coliform organisms are
exactly analogous to Equations IV-18, IV-21, and IV-23A and IV-23B, since
first order decay is used for both. For purposes of hand computations, the
following decay coefficient is acceptable:
kt = 1.0 + 0.02 (T-20) (IV-92)
where
ktc = decay coefficient for total col iforms, I/day
T = water temperature, °C
Those equations with the widest applicability are listed below. For a point
source of col iforms:
482
-------
TC - TCQ exp
-J'
tc
A
(IV-93)
For both point and distributed sources of coliforms:
TC
_ r— + ITC -
•tc
For a change in coliform concentration due to a point source
modification:
(IV-94)
ATC -
ATC exp
ATC
"J tc /ft
V (*•'
•tc
(IV-95a)
(IV-95b)
where
TC = total coliform concentration, MPN/100 ml
TC = initial total coliform concentration, MPN/100 ml
'tc " u:
TC = total coliform level in distributed flow, MPN/100 ml
'tc
Because of the potential variability in coliform loadings, seasonal
analyses may be warranted. Typically the summer months are of primary
concern because loadings often increase during this time period and water
contact recreation is at its maximum. Major storm events may also be of
interest, because of the large coliform loading that may be associated with
them.
483
-------
EXAMPLE IV-15
Estimating the Change in Total Coliform Levels
in Response to a Waste Loading Change
Compare the change in total coliform levels, ATC, produced by a
change ATCQ at a given location in a river. Further, determine how this
change is affected by a distributed flow entering the river. Relevant
data for the river are as follows:
UQ = Ifps
T = 20°C
Q = 500 cfs
Q = 800 cfs
x = 10 miles
kt = 1.0/day at 20°C
First the computations will be performed assuming no distributed flow.
Equation IV-95A is then applicable. Computing the exponent jt x (at a flow
distance of 10 miles):
J* x (1.0) (10) (5280) _ n ,n
tc = (24) (3600) (1) ' U-b"
So
ATC
ATCQ
= exp (-.611) = 0.54
or
ATC =0.54 ATCQ
For example if ATC = -1,000 MPN/100 ml then ATC = -540 MPN/100 ml (negative
ATC indicates that the coliform level has decreased from what it previously
was).
484
-------
Now suppose the distributed flow of 300 cfs is included in the
computation. Then,
F - ktc Ao * AQ
tC " AQ
Ao = VUo = 500/1 = 50° ft2
AQ = TOTWO-) = °-0057 ft2/sec
F _ (1.0) (500)
tc I24j (3600 P(0. 0057)
= 2.02
Then
_ 2'02 . 0 39
O " \8ooy ~ u-jy
or
ATC = 0.39 ^TCQ
For ATCQ = 1,000 MPN/100 ml, ATC = -390 MPN/1 00 ml.
Note that this decrease is 150 MPN/100 ml less than if no distributed flow
existed.
To determine the absolute total coliform level, simply add to the
original level the resulting change caused by the waste loading
modification.
ATC_ _ SOO
ATC "
END OF EXAMPLE IV-15
485
-------
4.7 CONSERVATIVE CONSTITUENTS
4.7.1 Introduction
Conservative constituents are those which are not reactive and remain
either in solution or in suspension. They are advected through the water
column at the velocity of the river with no loss of mass. The analysis of
nutrients, already discussed in this report, was performed assuming they
acted conservatively. Other substances, such as salinity, can also be
considered as conservative. Chapter 3 contains information on salinity in
irrigation return flow for many rivers with salinity problems.
4.7.2 Mass Balance for Conservative Constituents
Two simple mass balance equations are sufficient for analyzing
conservative constituents. The first relates the instream concentration due
to a point source loading:
SO + W/5.38
S = u " + ' (IV-96)
where
S - resulting pollutant concentration, mg/1
S = upstream concentration, mg/1
Q = upstream flow rate, cfs
Q = point source flow rate, cfs
W = loading rate of pollutants, Ib/day
When a distributed flow is present along some length of the river, then the
distribution of the conservative pollutant is given by:
S 0
(IV-97)
5.38C)
where
w = distributed loading rate, Ib/day/mi
x = distance downstream, miles
486
-------
S = initial concentration (at x = 0), mg/1
S in Equation IV-97 is identical with S in Equation IV-93.
•EXAMPLE IV-16-
Calculating Salinity Distribution in a River
Salinity problems are receiving increased attention in the western
United States, particularly relating to the economic issues in the Colorado
River Basin and international compacts with Mexico. In the Colorado River
high salinity levels in the lower reaches adversely affect nearly twelve
million people and approximately one million acres of fertile irrigated
farmland (Bessler and Maletic, 1975). The salinity now averages
approximately 865 mg/1 at Imperial Dam and is projected to be 1,160 mg/1 or
more by the year 2000, unless firm control actions are taken.
Consider the river shown in Figure IV-38. Predict the salinity
distribution based on the inflows and withdrawals shown. Assume the data
are averaged over a period of a year. These data, along with the salinity
concentrations at different river mileposts are shown in Table IV-32.
To calculate S (salinity at milepoint 100) use Equation IV-96:
6
S = ^ J^xT0)n. 55/8. 34) _
2000 ido Tig/l
At milepost 199.9, Equation IV-97 is appropriate and S is given by:
6)(2pO_0) x (4xl06)(1.55
50~0(T~ 5000
. (186)(2pO_0) x (4xl06)(1.55/8.34)
~~~ ~~ = ^J mg' '
487
-------
Q = 3000cfs Q = 5000cfs
W=4xl06lb/day W = 25xl06lb/day 0 - 25OOcfs
\
-— "— ^— •"^•-•—••^— - — •> ---- *— J— J-J- J -- j --- -*"»---»-T»««|f lp1l»«M J
^ \ / Tiff!
Q = l500cfs * 0-SOOOcfs I » 1 I I
W-PvlfAh/rtm/ ^ U-3UUUCTS Q=2000cfs
W-2xlO°lb/day Q=,000cfs W=8xloP|b/day W = 20xlo6|b/day
0 100 200 300 400 500 600 700 750
i i i r ~i— i~~ i ii
RIVER MILES
FIGURE IV-38 SALINITY DISTRIBUTION IN A HYPOTHETICAL RIVER
At milepoint 280, 1,000 cfs of flow leaves the mainstem (perhaps for
irrigation purposes). The concentration of salinity in this flow is the
same as that in the mainstem. So the mass rate of withdrawal is
W - ~Y~ (223 x 1000) = -1.2xl06 Ib/day
A negative sign is used to signify a withdrawal. Completing the
remainder of the table is solely a matter of reapplying these basic concepts,
4.8 SEDIMENTATION
4.8.1 Introduction
END OF EXAMPLE IV-16
One of the more difficult classes of hydraulic engineering problems
associated with rivers involves the erosion, transportation, and deposition
of sediment. Sedimentation is important economically, particularly relating
to filling of reservoirs and harbors, and to maintaining channel
488
-------
TABLE IV-32
SALINITY DISTRIBUTION IN A HYPOTHETICAL RIVER
Reach
Number
1
2
3
4
5
6
7
8
9
10
River
Mile
Point
0
99.9
100
199.9
200
279.9
280
359.9
360
449.9
450
499.9
500
524.9
525
599.9
600
649.9
650
750
Sal inity
Added*
(Ibs/day)
0
0 c
2x10^
4x10°
0
0
-1.2x10°
0
0
25x10°
0
0 c
8 106
0
-7.9xlOb
0
-4.7x10°
0
0 ,
20x10°
L
D
L
D
L
L
L
D
Sal inity
Cumulative
(Ibs/day)
0
0 c
2x10^
6x10°
6x10°
6x10° ,
4.8x10°
4.8x10°
4.8x10°
29.8x10°
29.8x10°
29.8x10°
37.8x10°
37.8x10°
29.9x10°
29.9x10°
25.2x10°
25.2x10°
25.2x10°
45.2x10°
Q
Cumulative
(cfs)
500
500
2000
5000
5000
5000
4000
4000
4000
9000
9000
9000
12000
12000
9500
9500
8000
8000
8000
10000
Salinity
Concentration
(mg/1 )
0
0
186
223
223
223
223
223
223
615
615
615
585
585
585
585
585
585
585
840
*'L' indicates a localized or point source at the milepoint shown in
the same row.
'D' indicates a diffuse or non-point source ending at the milepoint
shown in the same row and beginning at the milepoint in the above row.
489
-------
navigability and stability. Table IV-2, located in Section 4.1, documents
some suspended solids problems encountered in eight major U.S. waterways.
The sediment load carried in a river can be divided into two
components: the bed material load and the wash load. The bed material load
is composed of those solid particles represented in the bed. The transport
of this material is accomplished both along the bed (bed load) and suspended
within the water (suspended load). Although there is no sharp demarcation
delineating bed load from suspended load, many researchers have developed
individual expressions for each transport component. The total bed material
load is the sum of the bed load and the suspended load. Other researchers
have developed a unified theory from which the total bed material load can
be predicted from a single expression.
The wash load is usually produced through land erosion, rather than
channel scour. Wash load is composed of grain sizes finer than found in the
bed material. It readily remains in suspension and is washed out of the
river without being deposited. A definite relationship between the
hydraulic properties of a river and the wash load capacity apparently does
not exist, making it difficult to advance an analytical method for washload
prediction (Graf, 1971). Not all the erodible material entering a stream is
transported as wash load, but a large portion may become part of the bed
material and be transported as bed material load.
Figure IV-39 provides a graphical illustration of the difference
between wash load and bed material load. For a particular flow condition in
a particular river, the river has the capacity to transport a certain
quantity of sediment (q ) which generally decreases as particle size
increases. At some large particle size the river cannot exert enough force
to transport particles of that size or larger. This situation would occur
at some point to the right of point D on curve COD. This same river might
be supplied with sediment at a rate AOB, which is unrelated to transport
capacity.
To the left of point 0 the river is transporting all the material of
that size range being supplied to it. Sediment having diameters less than
d* are classified as wash load, because the amount being transported is
490
-------
CD
d
73
m
UD
O *-<
-n o
73 Z
o
2 dd
• - m
SEDIMENT TRANSPORT CAPABILITY OF FLOW
o
r~
o
m
m
O CO
o a:
oo i —
-i o
> >
-i a
m
z a
t—4
< CO
m m
73 O
H >
•< H
>. m
CO >
*--J |—
(-D
N
m
O)
rn
g
^
m
o
r~
m
SEDIMENT SUPPLY RATE
-------
supply limited, and not transport limited. To the right of point 0,
supply exceeds transport capacity. The amount given by the curve OD is
transported, and the difference in OB and OD is deposited in the stream bed.
The methods to be presented in the following sections are generally
concerned with predicting curve OD (i.e. the bed material load), although
Section 4.8.2 does provide a brief description of how to estimate long-term
sediment supply rates.
As a guide in evaluating whether a river is carrying a significant
quantity of suspended sediment, Table IV-33 can be consulted. 100 mg/1 is
the delineation between a potential and probable problem. In a table
previously introduced (Table IV-1), a reference level of 80 mg/1 was set for
protection of aquatic life.
TABLE IV-33
RELATIONSHIP OF TOTAL SUSPENDED SEDIMENT CONCENTRATION
TO PROBLEM POTENTIAL (AFTER EPA, 1976)
If Calculated Probability of
Concentration is: a Problem
Less than 10 mg/1 Improbable
Less than 100 mg/1 Potential
More than 100 mg/1 Probable
4.8.2 Long-Term Sediment Loading from Runoff
The procedures outlined in Chapter 3 will permit an assessment
of the sediment loading to a river on a long-term basis. When using
those procedures care should be taken to incorporate the entire
drainage area of the watershed. As an estimate, the loading can be
assumed conservative (i.e. all sediment that comes into the river will
be washed out of the river over an extended time period). Under that
assumption the procedure outlined in Section 4.7 can be utilized for an
estimate of average yearly suspended solids concentrations at locations
throughout the river system. This result should be interpreted as
492
-------
an indicator of the impact of the runoff on sediment loads within a
river and not as actual suspended solids concentrations. Not all of
the incoming sediment will be transported as suspended load since a
large fraction can be transported as bed load. The transport process
is generally of an intermittent nature with higher concentrations
occurring during periods of high flow.
Care should be taken not to apply the conservative assumption at
points on a river where that assumption is clearly violated, such as
at reservoirs which can be efficient sediment traps. An example for
the computation of sediment loading to rivers has been considered in
Chapter 3.
4.8.3 Bed Material Load
As previously mentioned, the estimation of bed material transport
poses a difficult problem, and is an area where there is no consensus
regarding the best predictive relationship to use. Numerous bed
material load relationships (Task Committee on Preparation of Sedimen-
tation Manual, 1971) have been developed over the past century, some
requiring considerably more input data than others. In this report
the DuBoys relationship (Task Committee on Preparation of Sedimentation
Manual, 1971) will be used in part because of its simplicity. The
relationship,which is restricted to uniform flow in alluvial channels,
is:
gb = *ao (TQ - TC) (IV-98)
where
gb - bed load, Ib/sec/ft of width of river
'!' = coefficient depending on grain size, ft /lb/sec
p
T - yRu S, bed shear stress, Ib/ft
o n
Y = specific weight of water, Ib/ft
Ru = river hydraulic radius, ft
H
493
-------
S = slope of stream, ft/ft
ic = critical shear stress, lb/ft2
The values of Y and r can be expressed as functions of the median
size (by weight) of the bed sediment (den)- These relationships are
expressed graphically in Figure IV-40. To aid in determining d™
Table IV -34 is presented to show the size range of sediment and each
associated class name. If the class name of the predominant sediment
type comprising a stream bed is known, then the sediment size (in mm)
can be estimated. Appendix C also contains
rivers and streams.
™
data for numerous
Once d5Q is estimated, then
and TC can easily be evaluated,
leaving only TQ to compute. A summary of hydraulic radii (the ratio
of cross-sectional area to wetted perimeter) for different channel
200
.2 .4 .6 .8 I 2 4
MEDIAN SIZE OF BED SEDIMENT, d.
(MM)
FIGURE IV-40
* AND TC FOR DUBOYS RELATIONSHIP
AS FUNCTIONS OF MEDIAN SIZE OF
BED SEDIMENT (TASK COMMITTEE ON
PREPARATION OF SEDIMENTATION
MANUAL, 1971)
494
-------
TABLE IV-34
SEDIMENT GRADE SCALE (TASK COMMITTEE ON PREPARATION
OF SEDIMENTATION MANUAL, 1971)
Class Name
Very large boulders
Large boulders
Medium boulders
Small boulders
Large cobbles
Small cobbles
Very coarse gravel
Coarse gravel
Medium gravel
Fine gravel
Very fine gravel
Very coarse sand
Coarse sand
Medium sand
Find sand
Very fine sand
Coarse silt
Medium silt
Fine si 1 1
Very fine silt
Coarse clay
Medium clay
Fine clay
Very fine clay
Size
Mil 1 imeters
2-1
1-1/2
1/2-1/4
1/4-1/8
1/8-1/16
1/16-1/32
1/32-1/64
1/64-1/128
1/128-1/256
1/256-1/512
1/512-1/024
i /1 024-1/2048
1/2048-1/4096
4096-2048
2048-1024
1024-512
512-256
246-128
128-64
64-32
32-16
16-8
8-4
4-2
2.000-1 .000
1.000-0.500
0.500-0.250
0.250-0.125
0.125-0.062
0.062-0.031
0.031-0.016
0.016-0.008
0.008-0.004
0.004-0.0020
0.0020-0.0010
0.0010-0.0005
0.0005-0.00024
P Approximate Sieve Mesh
ange Openings Per Inch
United States
Microns Inches Tyler Standard
160-80
80-40
40-20
20-10
10-5
5-2.5
2.5 -1.3
1.3 -0.6
0.6 -0.3 2-1/2
0.3 -0.16 5 5
0.16-0.08 9 10
2000-1000 16 18
1000-500 32 35
500-250 60 60
250-125 115 120
125-62 250 230
62-31
31-16
16-8
8-4
4-2
2-1
1-0.5
0.5-0.24
-------
geometries is shown in Figure IV-41. For very wide, shallow channels,
the hydraulic radius approximately equals the depth of flow. Many
river cross-sections can be approximated by a parabolic section. To
calculate "c" in the relationship for hydraulic radius of a parabolic
section, refer to Table IV-35.
If the bed slope is unknown it can be estimated by using a
topographic map and finding contour lines approximately five hundred
feet above and below the point on the river where the measurement is
to be made. Dividing this elevation difference by the horizontal
distance over which the difference is measured, produces the slope.
TABLE IV-35
COMPUTING D/T FOR DETERMINING THE HYDRAULIC RADIUS OF
A PARABOLIC SECTION (FROM KING, 1954)
*
D
T
.0
.1
.2
.3
.4
.5
.6
.7
.8
.9
.00
.667
.650
.607
.554
.500
.451
.408
.370
.338
.311
*D
T C
.01
.667
.646
.602
.548
.495
.446
.404
.367
.335
.308
.02
.666
.643
.597
.543
.490
.442
.400
.364
.333
.306
.03
.665
.639
.592
.537
.485
.437
.396
.360
.330
.303
.04
.664
.635
.586
.532
.480
.433
.392
.357
.327
.301
.05
.662
.631
.581
.526
.475
.428
.388
.354
.324
.298
.06
.660
.626
.575
.521
.470
.424
.385
.351
.321
.296
.07
.658
.622
.570
.516
.465
.420
.381
.348
.319
.294
.08
.656
.617
.564
.510
.460
.416
.377
.344
.316
.291
.09
.653
.612
.559
.505
.455
.412
.374
.341
.313
.289
496
-------
CHANNEL SLOPE
HYDRAULIC RADIUS
f
1
Trapezoidal
3
e
Q-±_zxLJL__ x = D/b, z = e/D
1 + 2 x Vl + x2
T \
f
Rectangular
bD
b + 2D
Triangular
z = e/D
T
Parabolic
n
cu
(for c, see
Table IV-29)
FIGURE IV-^1 HYDRAULIC RADII FOR DIFFERENT CHANNEL SHAPES
(FROM KING, 1954)
497
-------
Adequate methods that are within the scope of this report and
which would provide a straightforward estimation of suspended sediment
discharge presently do not exist. Most relationships require a known
reference level concentration at some depth within the river to predict
the concentration at another depth (Morris and Wiggert, 1972). To
determine the suspended sediment load, then, a summation of contributions
at each depth must be made. Since these formulas apply to one grain
size this procedure should be repeated for all grain sizes present.
Einstein (Graf, 1971) has developed a method for computing suspended
sediment discharge that does not require knowledge of a reference concen-
tration, but it is an advanced approach. For this report the contri-
bution of the suspended load will be estimated from the bed material
load by the relationship given in Table IV-36. The relationship in
Table IV-36 is valid for graded channels (by graded is meant that the
slope is stable over time, being neither steepened nor flattened by
flow or other influence).
TABLE IV-36
RELATIONSHIP BETWEEN WIDTH TO DEPTH RATIO
OF A GRADED STREAM AND THE SUSPENDED AND
BED LOAD DISCHARGE (AFTER FENWICK, 1969)
Suspended
Load % of Total
Bed Material Load
85-100
65-85
30-65
Bed Load % of
Total Bed
Material Load
0-15
15-35
35-70
Width-
Depth
Ratio
7
7-25
25
Once the width to depth ratio for the stream in question is determined,
the suspended load can then be approximated after first computing the
bed material load, and then using Table IV-36.
498
-------
Once the suspended load discharge is estimated the average
concentration at a section can be computed by:
4il.6xl04 (IV-99a)
or g
Css - -^1.6 x 104 (IV-99b)
where
C = average suspended solids concentration, mg/1
G = suspended solids discharge, Ib/sec
Q = flow rate, cfs
g -- suspended solids discharge per unit width, Ib/sec/ft
q = flow rate per unit width, cfs/ft
The procedures discussed in this section can be summarized as follows
1. Determine the bed load discharge g (Ib/sec/ft) using Equation
IV-98. The required input data are channel slope, hydraulic
radius (see Figure IV-41), and the median sediment size, d™
(see Appendix C). Once d has been estimated the unknown
bu
parameters T and ¥ can be found from Figure IV-40.
2. Multiply g, by the river width to find the total bed load
discharge.
3. Determine the width/depth ratio.
4. Use Table IV-36 to determine the suspended load.
5. To determine the suspended sediment concentration use Equation
IV-99.
499
-------
6. Compare the suspended sediment concentration against the data
in Table IV-33 to find out if a problem potentially exists.
7. The total bed material load is sum of the total bed load
(step 2) and the total suspended load (step 4).
The user may be primarily concerned with the total bed material load
rather than either bed load or suspended load individually. Total bed
material load can be directly calculated using a number of predictive
formulas. The method of Yang (1976) based on unit stream power is presented
here. Yang's method has been verified for the following parameter ranges:
t median bed size from 0.16 mm to 1.0 mm
• channel depth 0.2 ft to 49.9 ft
• water temperature from 0°C to 29.4°C
• stream velocity from 1.23 fps to 7.82 fps
• flow rate from 2.7 cfs to 470,000 cfs
• slope from 0.0000428 to 0.00188
t total sediment concentration (excluding wash load) from 2.8
ppm to 2,440 ppm
The input data are the same as for the DuBoy's method, with the
addition of water temperature. The predictive formula, however, is
considerably more complicated, so the method has been programmed on a hand
held calculator and the program is included. The predictive expression is:
log Ct = 5.435 - 0.286 log ^ - 0.457 log ^
+ (1.799 - 0.409 log ^ - 0.314 log ^) log (— - -^-) (IV-100)
where
C. = total sediment concentration in parts per million by weight
D = median sieve diameter
S = water surface slope or energy slope
U* = shear velocity
U = average water velocity
Ucr = Critica1 average water velocity at incipient motion
500
-------
v = kinematic viscosity
w = terminal fall velocity
The term -^- can be calculated as
w
Ucr 2 5 U D
+ 0.66 when 1.2 < ^ < 70 (IV-101)
. .
) - 0.06 v
and
cr
w ' " — v
= 2.05 when 70 < (IV-102)
Figure IV-42 shows the required user instructions to execute the
program on a TI-59. Figure IV-43 contains the program listing and a sample
input/output. This program was written by Colorado State University (1979).
EXAMPLE IV-17
Estimation of Bed Material Load
Table IV-37 shows characteristics of the Colorado River at Taylor's
Ferry, California, and of the Niobrara River near Cody, Nebraska. Suppose
one desires to calculate the bed load for the Colorado River at this
location for flow ranges of 8-35 cfs/ft. The following data will be used:
d^Q = 0.33 mm
Y = 62.4 lb/ft3, at 60°F
S - 0.000217 ft/ft
Using Figure IV-40 one finds
V - 64
T - 0.019
501
-------
TITLE.
.PAGE i_OF.
PROGRAMMER.
.DATE.
Partitioning (Op 17) I 4, 6, 0. 6, Ol Library Module.
.Printer Optional Cards 1
PROGRAM DESCRIPTION
Program: Yang's Sediment Transport Equation
USER INSTRUCTIONS
STEP
1
2
3
4
5
6
7
PROCEDURE
Ift2\
Enter kinematic viscosity, vlj=-l
Enter slope SQ (ft/ft)
Enter median sediment diameter, d (ft)
Enter flow velocity, U (r^r )
sec
Enter flow depth, Y (ft)
Compute sediment concentration (ppm)
To input new data, repeat steps 1
through 6.
ENTER
V
So
ds
U
Y
PRESS
A
B
C
0
E
2nd
A'
DISPLAY
V
So
ds
U
Y
Ct
FIGURE IV-42
USER INSTRUCTIONS FOR YANG'S SEDIMENT TRANSPORT
EQUATION,
502
-------
Procram Listina:
000
001
002
003
004
005
006
007
003
009
0 1 0
Oil
012
013
014
015
016
01?
01 S
019
020
021
022
023.
024
025
026
02^
0 2 3
029
030
03 1
032
0 3 3
034
035
036
037
0 3 S
o r '-•
0 4 0
IJ4. i
0-2
0 4 '•
Q44
04f
C4r
'J- ~*
C'4 E
0-l'r-
:-' ~ '.'
76
77
93
00
00
00
02
32
43
02
77
i->i-i
•j'j
33
65
02
93
09
05
01
07
55
43
06
95
42
00
'~"2
76
83
53
53
53
43
H2
45
03
65
03
06
•3:-:
00
06
IJ4
35
03
06
65
43
06
O 21
sr 1
LBL
GE
•
0
0
0
"J
X;T
RCL
02
GE
DMS
X*
X
2
•
C(
5
4
7
-f-
RCL
06
S
STD
00
RTN
LBL
DMS
<
<
(
RCL
02
i i •!. .•
•( -•••
o
X
3
£,
•
o
6
4
"*
+
O
K
x
RCL
Ot
X^
'>
051
052
053
054
055
056
057
053
059
060
061
Of2
063
064
065
066
067
063
069
070
071
072
073
074
075
076
077
078
079
080
081
082
083
034
085
086
087
083
039
090
091
092
093
094
095
096
097
093
099
100
34
75
06
65
43
06
54
55
43
02
54
42
00
92
76
11
42
06
22
52
92
76
12
42
01
92
76
13
42
02
92
76
14
42
03
92
76
15
42
04
92
76
16
71
( <
c-i
53
03
02
93
rx
-
6
X
RCL
06
>
-i.
RCL
02
>
STD
00
RTN
LBL
R
STG
06
INV
EE
RTN
LBL
B
STD
01
RTN
LBL
C
STD
02
RTN
LBL
D
STD
03
RTN
LBL
E
STD
04 '
RTN
LBL
ft'
SBR
GE
<
(
3
•
101
102
103
• 104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
133
139
140
141
142
143
144
145
146
147
143
149
150
02
65
43
04
65
43
01
54
34
42
05
A=;
43
02
55
M f,
to
06
54
42
07
32
07
00
s~, --
•J _
77
89
53
53
02
93
05
55
53
43
07
28
75
93
00
06
54
54
S5
93
06
06
54
42
OS
61
^-,
X
RCL
04
X
RCL
01
\
*
rx
STD
05
X
RCL
02
•r
RCL
06
)
STD
07
X:T
7
0
X;T
GE
n
<
<
2
•
5
+
f
RCL
07
LOG
-
•
0
6
>
•.
*r
•
6
£
>
STD
08
GTD
151
152
153
154
155
156
157
153
159
160
161
162
163
164
165
166
16T
163
169
170
171
172
173
17^
175
176
177
173
179
ISO
181
132
183
184
135
186
187
188
139
190
191
192
193
194
195
196
19^
193
19?
200
70
76
89
02
93
00
05
42
03
76
70
53
53
05
93
04
03
fi5
75
93
02
03
06
65
r--i
43
00
65
43
02
CT
•Jw
43
06
cr <
•Jt
2S
A •',
»^ ^
09
75
'-4 X
04
05
07
65
cr--.
--' -^
.< •— i
** ."t
05
55
A •-.
tO
00
tr .<
vJ T
PfiD
LBL
rf
2
•
o
5
STD
08
LBL
RRD
(
(.
5
•
4
3
5
-
•
-j
1-1
•j
6
X
(
RCL
00
\s
.r%
RCL
02
-i-
RCL
06
)
LOG
STC
09
-
•
4
1^
cr
"7
S-.
<
PCL
05
—
PCL
C1 G
'i
S
FIGURE IV-43
PROGRAM LISTING AND SAMPLE INPUT/OUTPUT FOR
YANG'S SEDIMENT TRANSPORT EQUATION
503
-------
Program Listing (continued):
201
202
203
20-1
205
206
207
208
209
210
211
212
213
214
215
216
217
213
219
220
221
2^2
223
224
225
226
227
22'-'
229
230
231
232
233
23-i
235
236
237
233
239
240
241
2^i
243
24*
245
"i .* •"
24"
24 5
2^?
250
28
42
10
54
35
53
53
01
93
07
09
09
75
93
04
00
09
65
43
09
75
93
03
01
04
65
43
10
54
65
53
43
03
65
43
01
55
43
00
75
43
08
65
43
01
54
23
54
54
22
LDG
STD
10
>
+
C
<;
1
m
7
9
9
-
9
4
0
9
X
RCL
09
-
.
3
1
4
X
RCL
10
>
X
<
RCL
03
X
RCL
01
-r
RCL
00
-
RCL
08
X
RCL
01
')
LDG
>
>
INV
251
252
253
254
28 LDG
98 flDV
99 PRT
92 RTN
Sample Input:
V
sn
0
d
s
U
Y
» .0000111
- .0017
- .000623
= 2.89
= 0.51
Output.-
Ct « 2117.066395
FIGURE IV-43 (CONTINUED)
504
-------
TABLE IV-37
CHARACTERISTICS OF THE COLORADO AND NIOBRARA RIVERS
(TASK COMMITTEE ON PREPARATION OF SEDIMENTATION MANUAL, 1971)
Data
Depth range, ft
Range in q, in cubic feet per
second per foot of width
Mean width, in feet
Stream
Colorado
River
(Taylor's Ferryj
4-12
8-35
350
Niobrara
River
(Cody, Neb.)
0.7-1.3
1.7-5
110
Slope, in feet per foot
Minimum value
Maximum value
Value used in calculations
0.000147
0.000333
0.000217
0.00116
0.00126
0.00129
Water temperature, in degrees
Fahrenheit
Minimum value
Maximum value
Value used in calculations
Geometric mean*
in mil 1
d
d
d
d
Mean
35'
50'
f I~ 3
65
r\ t~\ 9
90
si
imeters
in mill
in mil 1
in mill
in mill
ze, d ,
m
sediment size,
imeters
imeters
imeters
imeters
in millimeters
48
81
60
0.
0.
0.
0.
0.
0.
320
287
330
378
530
396
33
86
60
0.
0.
0.
0.
0.
0.
283
233
277
335
530
342
The geometric mean of a set of values X,
ihus the
geometric mean of the values 1, 2, 3, and 4 is (1x2x3x4) ' = 2.213.
Compare with arithmetic mean of 2.5.
505
-------
All that remains is the computation of the hydraulic radius. Since the
width is much greater than the depth, assume R = D.
H
4 ft at q - 8 cfs/ft
12 ft at q = 35 cfs/ft
Using Equation IV-98 it is found that the bed load is
0.12 Ib/sec/ft at q = 8 cfs/ft
gb
' 1.5 Ib/sec/ft at q = 35 cfs/ft
The actual bed material load observed at Taylor's Ferry has been compared
with the DuBoys prediction for a range of flow rates (Task Committee on
Preparation of Sedimentation Manual, 1971). This relationship is shown in
Figure IV-44 (The DuBoys curve in Figure IV-44 does not quite match the
calculations in this example because slightly different data were used).
Observe that the DuBoys relationship overpredicts the bed material load for
nearly all flow ranges. This pattern is repeated for the Niobrara River
(Figure IV-45). This suggests that the bed material load estimated by the
DuBoys relationship will in general exceed the actual bed material load.
This is further substantiated by other work (Stall et _al_., 1958). The more
accurate predictions of bed material load occur under high flow conditions,
which is generally when the prediction of bed material load is most
important.
506
-------
COLORADO RIVER
AT TAYLORS FERRY
o i.uu —
g 0.80-
- 0.60-
£ 0.40-
LJ
g 0.20-
X
On in
U. 1 \J
(S) 0.08—
Q 006-
z °-04~~
UJ
1 002-
Q
QJ
co n ni
Duboys/
/-
o
1 1
1
s£
/025. Referring to Table IV-36, the suspended load
should be between 30 and 65 percent of the bed material load. Assume
it is on the lower end of the scale, about 40%. Then the suspended
load is
3ss
0.08 Ib/sec/ft at q = 8 cfs/ft
Ib/scc/ft at q - 35 cfs/ft
i n
507
-------
or
'ss
160 mg/1 at q - 8 cfs/ft
440 mg/1 at q - 35 cfs/ft
NIOBRARA RIVER
04 0.8 2 468 10
' 0.6 I
WATER DISCHARGE (cfs/ft)
FIGURE IV-
SEDIMENT DISCHARGE AS A FUNCTION
OF I.'ATER DISCHARGE FOR THE
NIOBRARA RIVER AT CODY,
NEBRASKA (TASK COMMITTEE ON
PREPARATION OF SEDIMENTATION
MANUAL, 1971)
from Equation IV-99. These concentrations indicate that suspended sediment
concentrations are excessively high throughout the range of flows normally
encountered at Taylor's Ferry. Data on suspended sediment concentrations
have been gathered at Taylor's Ferry (U.S. Bureau of Reclamation, 1958).
The averages of 30 measurements taken there are as follows:
508
-------
Q = 7350 cfs (or q - 21 cfs/ft)
Css = 132 mg/1
Observed range of suspended sediment
concentration: 40-277 mg/1
The method of Yang predicts total concentrations of 40 to 80 mg/1,
which is within but toward the lower end of the observated data. The method
of DuBoy's predicts concentration between 160 and 440 mg/1 which is toward,
and beyond, the upper end of observation. These results illustrate the
possible variability of predictions between different approaches, and are
not necessarily atypical.
END OF EXAMPLE IV-17
4.9 TOXIC SUBSTANCES
4.9.1 Methods of Entry of Toxic Pollutants into Rivers
Although Chapter 3 discussed both point and nonpoint sources of
pollutants, the major pollutant source categories are summarized in Table
IV-38 to indicate how these scenarios differentially govern a pollutant's
fate. For simplicity, fate is analyzed in terms of volatilization and
sorption since these processes are important for a wide number of toxic
organic chemicals. These processes govern whether a pollutant remains in
the water column and whether the pollutant is transported as solute or
sorbate. If the effects of these processes are known, even if only
qualitatively, then the influence of processes such as photolysis and
biodegradation can be better predicted. For example, if a pollutant is
sorbed to suspended and bedded sediments, it is more protected from
photolytic reactions than when it is dissolved in the water column.
A common mode of pollutant entry is by a continuous discharge, either
from a municipal or industrial source. As mixing of the effluent and river
water occurs, partitioning begins. The sorbate is transported with the
509
-------
TABLE IV-38
METHODS OF INTRODUCTION OF TOXIC ORGANIC COMPOUNDS INTO RIVERS,
AND FATE IN TERMS OF VOLATILIZATION AND SORPTION
Pathway
Fate
Continuous input
- solute transported and volatilized
- sorbate transported with suspended solids
and with bed load
- sorbed to immobile sediments
- buried by sorption and net deposition
Cessation of continuous
input
desorbed from immobile sediments
solute transported and volatilized
resorbed to suspended sediments
contaminated sediments resuspended
portion remains buried
Washoff from land
application
Accidental releases
(e.g. spills)
Leaching
- transport of a major portion of pollutant
may be governed by first large storm event
- transported as solute and sorbate
- settles and accumulates on bed
- buried
- subsequently resuspended
- If s.g. >1, pollutant settles on streambed
- Volatilization may be unimportant
--reentrained back into stream and sorbed
on suspended solids
--pollutant can be slowly transported along
bottom
--diffused into bedded sediments
- If s.g. <1, pollutant tends to remain on
surface and be transported at speed of
surface current
--volatilization can be important
--gradually dissolved and sorbed
--dispersion attenuates peak concentrations
—wind speed and direction influential
- slow movement (years) of solute from dump
or disposal site to stream
- continues for years after cleanup of
dump
510
-------
suspended sediments, and can interact with the bed load and immobile bedded
sediments. Depending on the rate of exchange of the sorbate with the bedded
sediments and on the net sediment deposition rate, some of the sorbate can
gradually become buried in the bedded sediments.
If a continuous input ceases, the water column initially tends to clean
itself of the pollutant as uncontaminated upstream water replaces
contaminated river water downstream from the former source of pollution.
However, pollutant from the contaminated bottom sediments can desorb back
into the water column at low concentrations and the river bed becomes an
internal source of pollutant. The desorption period can last a long period
of time, depending on the amount of pollutant contained in the bottom
sediments. Section 4.9.3.4 discusses this phenomenon in detail.
Periodic nonpoint sources, such as washoff after an agricultural
application, is another pathway of pollutant entry into rivers. The mass of
pollutant transported tends to be governed by the timing of the first storm
event following application together with the degradation and volatilization
processes operative during the interim period.
Accidental releases of pollutants, even through infrequent events, can
be important. Exceptionally high concentrations of pollutants can result
from spills and the total mass supplied almost instantaneously can be the
equivalent of a continuous release lasting for many days. For example, in
1973 a chloroform spill on the Mississippi River resulted in about
800,000 kg (1,750,000 Ibs) of chloroform being released over a period of
several hours (Thibodeaux, 1977). Based on the background concentration of
chloroform in the river (5 ppb), the release was equivalent to a continuous
supply of chloroform released at background rates for a period of 300 days.
Many chemicals in their pure or nearly pure form have specific
gravities significantly different from unity. Because of this, and their
often limited solubility in water, it is a mistake to believe that all
spilled pollutants travel with the speed of the river, have infinite
dissolution capability, and disperse accordingly. High density pollutants
can sink to the river bed and become slowly reentrained back into the water
column while simultaneously diffusing and sorbing into the bedded
511
-------
sediments. Depending on the rate of dissolution of the spilled pollutant,
as well as the significance of the sorption and diffusion processes, the
spilled pollutant may remain in the riverine system for either an extended
or brief period of time.
In contrast to high density pollutants, pollutants with specific
gravities less than unity tend to at least partially remain on or near the
water's surface while undergoing dissolution. For these pollutants,
volatilization and photolysis can be extremely important. As the pollutant
is dissolved in the water column and moves downstream, dispersion becomes
important in attenuating the peak concentration.
Pollutants which leach from a surface or subsurface disposal site may
eventually reach a river. Although the mass input rate may be low, the
source can be continuous and last for years, even after cleanup of the site.
The sequence of instream events following the initiation and then the
cessation of point sources of toxicants further illustrates the role that
sorption plays in governing fate of sorbates. Figure IV-46 illustrates the
two situations. Figure IV-46a shows the pollutant distributions below a
point source at two distinct times (ti and t2 where t2> ti) following
initiation of the point source. As the toxicant is discharged the water
column concentration (the sum of the dissolved and sorbed phases) abruptly
increases at the mixing location. As the pollutant travels downstream, the
sorbate tends to partially desorb onto the formerly uncontaminated bottom
sediments. Additionally there may be a net exchange between the bedded
sediments and water column sediments, even if there is no net deposition.
As a result of these processes, the water column concentration tends to
decrease in the downstream direction. It may take a period of time greater
than t\ for the effects of the discharge to reach a distance D*. Depending
on the distance, and on the rate of accumulation of the toxicant in the
bottom sediments, as well as on other factors, the time required for the
water column concentration to be noticeably elevated at D* could greatly
exceed the travel time of the river over the distance.
512
-------
cn
t—1
oo
O
c:
70
m
i
Jr-
CTi
00 —< —I
O 2 O
C •— X
3D —I •—
O •— O
m > >
o
m
m ^3
co >
co H
> --
-H O
H- Z
O CO
O O
~n |—
POLLUTANT CONCENTRATION
POLLUTANT CONCENTRATION
fD
ro
-------
After the discharge of the toxicant has continued for a period of time,
the net exchange with the bedded sediment may diminish, so that the toxicant
concentration becomes constant over some distance both in the water column
and in the sediments. This situation is illustrated by the solid curve in
Figure IV-46b. Suppose at this time the input of the pollutant ceases. The
water column concentration just below the point source tends to abruptly
approach zero. As this happens, desorption of the toxicant from the bedded
sediment can occur, tending to replenish pollutant levels in the water
column, but to a lower level. Gradually, the pollutant can be desorbed from
the bedded sediments at a given location so that the bottom sediments are
naturally cleansed, from the upstream to the downstream direction. This
process can take many years and low levels of pollutant in the water column
can be detected throughout this period. More discussion of this phenomenon
is provided later in Section 4.9.3 and Example IV-18.
4.9.2 Vertical Distributionof Sorbate within Rivers
Even though most of the analytical tools presented later in Section
4.9.3 assume that, for simplicity, suspended solids concentrations are
uniformly distributed throughout the water column, in reality this is not
true. The vertical distribution of solids depends both on particle and
river characteristics. Heavier particles (those with the greater settling
velocities) are transported closer to the stream bottom while the lighter
particles are more uniformly distributed. This observation is significant
because pollutants which sorb to the particles also exhibit a non-uniform
vertical distribution. Pollutants which do not sorb tend to become
uniformly distributed vertically, regardless of the sediment distribution.
By understanding this, the user can better interpret instream pollutant
data, particularly if the pollutant tends to reside as sorbate. It may be
that a single pollutant sample is not sufficient to accurately characterize
the pollutant distribution, and in fact could be misleading in terms of the
total burden of the pollutant carried within the water column. Depth
integrated samples might be necessary to gain an accurate knowledge of the
pollutant's distribution.
514
-------
Figure IV-47 shows the vertical distribution of suspended solids in an
equilibrium condition. The parameter shown in the figure is defined:
Vs (IV-103)
where
V = settling velocity of suspended solids
K = von Karman's constant (^0.4)
U* = shear velocity = (g Ru S)°-5 , ft/sec
n
g = acceleration due to gravity, 32.? ft/sec2
Ru = hydraulic radius of river, ft
n
S = slope, dimensionless.
Very small values of z represent clay-sized particles, while larger values
represent first silt, and then sand. Figure IV-47 illustrates that clay
particles tend to be uniformly distributed vertically (50 percent in the top
half of the water column). About 75 percent of silt and over 95 percent of
the sand particles (typically) reside in the bottom half of the water
column. This suggests that in rivers where the suspended sediments are silt
and sand, the sorbed pollutant distribution will be vertically skewed. If
the suspended material is predominantly clay the sorbed pollutant
distribution will be uniform. Since pollutants tend to sorb to sand to a
lesser degree than to silt and clay, the vertical distribution of sorbed
pollutant will not be as skewed as the suspended sediment distribution.
Figures IV-48 through IV-49 show the fraction of pollutant present as
solute (C/Ct) versus relative depth for families of z values and K Sa
values. Sa is the suspended sediment concentration a small distance above
the bottom. For K Sa values less than 0.1, the sorbate concentration is
generally negligible compared to the solute concentration regardless of the
depth or the nature of the suspended material. For larger K Sa values, the
sorbate level can be important, depending of the nature of the suspended
material. For extremely large K Sa values, the sorbate concentration will
greatly exceed the solute concentration, at least near the river bed.
515
-------
INCREASED SETTLING
VELOCITY
0.001
RELATIVE SEDIMENT CONCENTRATION S/Sa
FIGURE IV-47 VERTICAL EQUILIBRIUM DISTRIBUTION OF
SUSPENDED SOLIDS IN A RIVER
516
-------
cn
o
m
i—i
<
oo
c~> <
o m
O H
m —i
z o
H >
;o r~
o co
^ Z*D
i—i
7"s Cd
00 H
> i-
O
II 2
I—1 O
CD TI
m
>
H
1-^
m
CV5
O
XI
m
m
TD
O
O
O
z
o
m
z
H
13
O
Z
O
o
o
o01
b'
o
p
b'
RELATIVE DEPTH
p
ro
-4-
p
CO
p
cn
p
O)
4-
o
-vl
p
00
p
CD
CO
03
0
m
-------
RELATIVE DEPTH
en
i—«
00
en
m
<
UD
o m
o -H
m >—>
2 o
H >
;o r~
H C3
O CO
Z H
x*x 03
\r> H
c> •—-•
o
I—' O
CD -n
m
>
H
I—<
m
oo
o
3J
m
m
o
O
z
o
rn
H
I]
^
O
o
~-
o
-------
Based on the hydraulic characteristics of the river, characteristics of
the material being transported in suspension, and the partition coefficient
of the pollutant, predictions can be made of the pollutant's distribution in
the water column. To use Figures IV-48 and IV-49 requires knowledge of Sa,
the suspended solids concentration at a distance n = a above the bottom
(where typically a = 0.05, or 5 percent of the river's depth). The
equilibrium expression for suspended sediments, which is found in numerous
sediment transport texts (e.g. Graf, 1971) can be rearranged to express Sa
as:
Sa = S("' (¥ I??)' (IV-104)
where
n = relative depth above bottom
To use this equation the suspended solids concentration must be known at one
depth in the water column. Typically, a depth averaged suspended solids
concentration might be readily available. Under these circumstances Sa can
be estimated as:
s
Sa = —1 ' ^a / (IV-105)
/ l~~) c'n
where a
S = depth average suspended sediment concentration.
The denominator of Equation IV-105 can be integrated numerically by one of
many available solution techniques (e.g. see Carnahan et_ _a]_., 1969). For
the case when a = 0.05 the relationship between Sa and S is given below:
z = 0.2—»-Sa = 1.8 S
z - 0.6—*-Sa = 4.4 S
z = 1.0—*-Sa = 8.2 S (IV-106)
z = 2.0—*-Sa = 17 S
z = 5.0—»-Sa = 20 S
519
-------
Based on a knowledge of S, Sa can be estimated from Equation IV-106, and in
turn can be used in Figures IV-48 and IV-49.
Typically there is a segregation of particle sizes found in suspension
compared with these found in the bed load, and in the immobile bed
materials. Based on these differences, the following can be hypothesized:
Xs > Xbl
where
X = sorbed pollutant concentration on suspended materials, mass
pollutant/mass sediment
Xbi = sorbed pollutant concentration on bed load, mass
pollutant/mass sediment
X,-m = sorbed pollutant concentration on immovable sediment, mass
pollutant/mass sediment.
Investigations carried out by Miles (1976) appear to support this
relationship. Miles collected insecticide residues on stream sediments and
in the water column. Results of the DDT analysis of Big Creek, Norfolk
County, Ontario, 1973 (DDT was banned in 1970) are as follows:
Concentration of DDT on Sediments
(rna_ss_ of pollutants/mass of sediments
suspended sediments 110 ppb = X
bed load 76 ppb = X, ,
immovable bed 26 ppb = X.
im
Miles (1976) also found that DDT transported in the dissolved phase ranged
from 10 to 92 percent of the total transported in the water column. This
finding is consistent with the results in Table 11-14 which shows that the
percent of pollutant transported in the dissolved phase can be high even for
pollutants such as DDT as long as the suspended solids concentration is not
extremely high.
520
-------
Contaminant data collected in bedded sediments can be very
illuminating. Although in a screening approach it is not anticipated the
user will go to the field to collect sediment core samples, some data might
be available. Depending on the quantity of data available the following
types of information might be determinable:
• The spatial extent of contaminated sediments, and pollutant
concentrations in the sediments
• The depth of contaminated sediment
• The quantity of toxicant contained in the sediment
0 A time history of pollution levels to determine whether they
are increasing or decreasing
• The probable sources of the pollutant, based on the location
and quantity of contaminated sediments.
Although extensive sampling is required to accurately determine all of the
above items, such programs have been successfully accomplished. For
example, an extensive sediment sampling program was conducted in the Hudson
River in New York to determine the sources of PCBs in the contaminated
sediments, and the degree of contamination (Turk, 1979).
4.9.3 Transport and Transformation Expressions for Toxicants in Rivers
The tools presented in this section can be used to predict instream
concentrations of toxicants for a variety of different situations.
Specifically, the following scenarios are addressed:
• mixing zone analysis,
• continuous point source discharges,
521
-------
t continuous nonpoint source discharges,
• desorption from bedded sediments,
• spills and instantaneous release of soluble chemicals, and
• spills of high density chemicals which sink to the river bed.
In contrast to many conventional organic pollutants which degrade into
innocuous substances, many toxicants are transformed to other chemicals
which can be as harmful or more harmful than the original. Consequently,
when toxicants are continuously discharged into a river, in addition to
predicting the concentration profile, it is useful to also determine:
• the pollutant's advection rate past a specified location,
• the pollutant's volatilization rate over a specified reach,
and
• the pollutant's rate of transformation to other species over a
specified reach.
The toxicant's fate is thus segregated into the processes of advection,
volatilization, and transformation.
In the following three sections on mixing zones, point sources, and
nonpoint sources, the user will find there are different methods of
approaching the problems. One way to simplify the analysis is to first
assume toxicants act conservatively. The user can then perform a first
level analysis to find out whether criteria are violated. If they are not,
then a detailed analysis is really not required if the objective is to
determine criteria compliance. If violations are predicted, a more detailed
analysis of these "hot spots" can be performed by considering the various
processes affecting the toxicant in the river. This approach requires more
work, but by judiciously applying the tools available, the analyses can be
expedited.
522
-------
4.9.3.1 Mixing Zone Expressions
Section IV-4.1.9 presented earlier delineated one- and two-dimensional
mixing zone expressions for conventional pollutants. The one-dimensional
expressions need to be extended in order to differentiate between solute and
sorbate. To do this, the following expressions for pollutant concentration
and the suspended solids concentrations are needed:
S = U " W "- (IV-107)
r = Cut Qu + Cwt Qw (IV-108)
to Du + QW
where
S , C f = concentration of suspended solids and concentration of
sum of solute and sorbate in the river above the
location of mixing, respectively
S , C 4. = concentration of suspended solids and concentration of
w wt
sum of solute and sorbate in the wastewater,
respectively.
S, C = concentration of suspended solids and concentration of
sum of solute and sorbate in the river following mixing,
respectively.
The dissolved phase concentration, C, of the pollutant at the completion of
mixing is given by:
r - to (IV-109)
L " 1 + k S
P
where
523
-------
C and S are found from the two previous expressions.
to
The concentration of the solute following mixing depends on characteristics
of the waste source, the river's flow rate, and the suspended solids
concentration in the river and waste source. The solute concentration might
also change after mixing with a tributary of very high suspended solids
concentration (high SJ, even if it contains no additional pollutant
(Cwt=0).
Equation IV-108 is particularly useful because it predicts the total
instream concentration of toxicant following initial mixing. This is often
the critical test in establishing whether or not water quality standards are
violated by a point source.
In cases where initial mixing is incomplete (that is the waste is
initially diluted with a fraction of the total river flow), the
two-dimensional mixing equation shown earlier as Equation IV-4 will more
accurately predict C tQ. Then Equation IV-109 can be used to find the solute
concentration.
When there are numerous discharges of the same toxicant, analysis
becomes more complicated. The most straightforward method of handling this
situation is to sequentially apply Equation IV-108 to the series of
discharges to find the concentration as a function of distance downstream.
If the solute concentration is needed, then sequential application of
Equations IV-108 and IV-109 is required.
The analysis of multiple point sources can be simplified in one of two
ways. One, the sources can be transformed to an equivalent nonpoint source
by assuming that the toxicant input is uniformly distributed between the
series of point sources. This approach is discussed in Section 4.9.3.3.
Two, a series of closely grouped point sources can be handled as an
equivalent point source. The equivalent point source has a flow rate equal
to the sum of the flow rates from the individual plants, or:
524
-------
Q = £ Q . (IV-110)
w /r', xwi
where
Q . = flow rate from i;th treatment plant
W1
n = number of treatment plants being grouped
The total pollutant load can be expressed in one of two ways. If the
concentrations in the wastewater are known then the total loading is:
C Q = £ C . Q . (IV-111)
u xi.i *—' un xun * '
W ^W .~, Wl
where
C . = concentration of toxicant in effluent of ith plant
wi —
If the mass emission rates are known instead then:
C Q = £ M. (IV-112)
w w £j^
5.38
where
M- = mass emission rate of toxicant from ith plant is Ibs/day.
The conversion factor 5.38 converts mass emission rate in Ibs/day to flow
units in cfs and concentration units in mg/1 (ppm).
The grouping procedure described above has been applied by the
U.S. Environmental Protection Agency (1981) to a case study in Indiana to
evaluate the economic impact of toxicant standards. Numbers of point
sources were grouped together using a procedure called cluster analysis.
The cluster analysis added the loadings of major and minor industrial
dischargers within a ten-mile radius of each other. Ten clusters were
identified and few violations occurred within them once the best available
technology was attained.
525
-------
For certain applications the object of using a mixing zone equation is
to directly find the maximum allowable concentration in the discharge so
that the receiving water criteria are not violated. Under these
circumstances Equation IV-108 can be rewritten as:
(Q + Q )
CTC ~^Q ' when Cut
w
where
(C .) = maximum allowable concentration of the toxicant in the
wt max
waste discharge so that the water quality criterion is
met under critical conditions.
C. = water quality criterion for the toxicant
Q = critical river flow rate (e.g. 7Qio)
Equation IV-113b is applicable when the concentration of the toxicant is
zero upstream of the discharge point.
4.9.3.2 Point Source Discharges
For point sources of toxicants, the pollutant interactions depicted in
Figure IV-50 are simulated. While transformation of toxicants is generally
more complex than this, in many instances these interactions are sufficient
to analyze the instream processes affecting not only point source discharges
but also nonpoint source discharges, and instantaneous releases of soluble
pollutants. Figure IV-50 reveals that:
• The solute only is assumed to volatilize.
526
-------
V
FIGURE IV-50 INSTREAM TRANSFORMATION PROCESSES
ANALYZED FOR TOXICANTS,
527
-------
• First order transformation processes degrade only the solute.
• Adsorption and desorption are assumed to occur at rates much
faster than other processes.
• No interactions with the bottom sediments occur (this is
analyzed in later sections).
Based on these interactions, the concentration profile below a point source
of toxicant is expressible as:
where
C = concentration of dissolved phase of the toxicant at a
distance x below the point source
C = concentration of the dissolved phase of the toxicant at
x = 0 (after the point source discharge has mixed with the
river water)
k ' = k /D
v v
D = water depth
k. = individual first order decay rates which are transforming
the toxicant (other than volatilization)
P = partial pressure of the toxicant in the atmosphere above the
river,
and the remaining variables have previously been defined. Typically the
partial pressure is zero, so that Equation IV-114 simplifies to:
528
-------
C = CQ exp
The initial dissolved phase concentration is given by:
where
C =
1 + V
(IV-115)
(IV-116)
C was defined by Equation IV-108
to
The total pollutant concentration, C. , at any location is:
Ct = C (1 + K S) ,
(IV-117)
The sorbed phase concentration expressed as mass per unit volume of water
is:
cs = ct - c .
(IV-118)
and the sorbed phase concentration expressed as mass per unit mass of
sediment is:
X - K C
P
(IV-119)
The most direct application of Equation IV-114 or IV-115, plus
Equations IV-117 through IV-119 is to find the instream concentration as a
function of distance below the point source. There are, however, other uses
of the expressions. Consider Equation IV-115, for example. The ratio C/CQ
can be directly calculated as a function of distance. Thus the fractional
dissolved phase concentration can be calculated without ever knowing the
initial concentration C . This approach has the advantage of requiring less
data. Similarly, the fractional concentration can be calculated for any
specified distance, such as the end of a reach. Or, the distance x can be
found so that the fractional concentration is some specified number, which
may relate to an acceptable level of toxicant. The length of river
subjected to unacceptable levels can then be found.
529
-------
The user might additionally want to know the distribution of pollutant
fluxes in terms of advection (Mg), volatilization (My), and transformation
(NL). Expressions for these are presented for the case of P = 0. These
formulae allow the user to predict the fluxes associated with the point
source discharge where volatilization is not altered by a background
concentration in the atmosphere. Under these conditions:
ft = Ma + MV + ftt (IV-120)
Equation IV-120 states that the rate of entry of the toxicant into the river
(M) equals the rate of advection of that toxicant past some location x ,
plus the rate of volatilization across the water surface between the
discharge location and some other specified location plus the rate of
transformation of the toxicant to other substances within the water column
between the same two locations. By knowing expressions for each of M , M ,
and M. the user knows the major processes controlling the toxicant's fate
within any reach of river.
The mass flux advected past a location x is given by:
Ma = (Qu + Qw) C + (Qu + Qj Cs (IV-121)
where the concentrations C and C are evaluated at x = x . The
volatilization mass flux is given by:
Mv = Ac kv Cc
U (1 + K S)
P
where
l~e*p rOlTTTr
< xi
S) Xs I
(IV-122)
A = cross-sectional area of river
c
and all other terms have previously been defined. In some cases the user
might have an estimate of the average dissolved phase concentration, C,
within the reach under consideration. Under these circumstances the
volatilization flux is simply:
530
-------
ky C
(IV-123)
where
A = surface area of the reach under investigation.
o
The transformation mass flux is expressible as:
Mt • Ac
U (1 +
1-exp
k;
U (1 + KpS) s
(IV-124)
Since the sum of Equations IV-121, IV-122, and IV-124 equals the mass
emission rate of the toxicant, Equation IV-120 can be used to double check
the fluxes calculated.
4.9.3.3 Nonpoint Source Discharge
This section parallels the previous section on point source discharges
by presenting expressions for the steady-state concentration profile, and
for mass fluxes. In addition to applying this methodology to a nonpoint
source, another and possibly more useful application is to express a series
of point sources as an equivalent nonpoint source. The equivalent nonpoint
source discharge rate is simply the sum of the discharge rates of the
pollutant from all the point sources. This approach is not as accurate as
analyzing each point source individually but is much faster depending on the
number of point sources. For example, suppose a river segment has ten
separate point sources located within 50 miles of each other. The most
rigorous analysis would consist of considering each point source
individually, where mixing zone and point source equations are applied
sequentially ten times each. This obviously is a great deal of work for a
hand calculation approach. By considering these point sources as a single
equivalent nonpoint source, a single equation application is sufficient to
analyze the problem. Example IV-5 shown earlier in the BOD section
illustrates this procedure.
531
-------
The solute concentration in a river resulting from a steady nonpoint
source of toxicant is:
'•**(°.*fc)(^)*
where
1 + K S
k = —J—P— m
= k2 + k;
= total concentration of toxicant in nonpoint source
m
xi
Qf = river flow rate at end of nonpoint source
Q0 = river flow rate at beginning of nonpoint source
x-| = length of nonpoint source.
Equations IV-117 through IV-119 can be used to find Ct, Cs, and X,
respectively.
In a manner similar to point source discharges, Equation IV-120 which
expresses the mass balance between toxicant inflow rate to the river and
loss rate by advection, volatilization, and transformation, is valid. The
appropriate expressions are (when P = 0):
= (Q0+mx)C + (Q0+mx) CkpS , at = x$
••—•*** - !_,. •** ... ,— ^ '— —
solute sorbate
transport transport
532
-------
for the advective flux. For the volatilization flux
(IV-127)
For the transformation flux:
dV-128)
As a first cut analysis, the user might want to assume that the toxicants
act conservatively. If criteria are not violated under these circumstances,
then criteria will not be violated if decay or transformation processes are
included.
4.9.3.4 Desorption of Toxicant from a River Bed
Because many toxicants are transported as sorbate rather than as
solute, a significant fraction of the pollutant which enters a riverine
system can ultimately be deposited in the bedded sediments. If the toxicant
is resistant to degradation processes it can remain in the sediments for
extended periods of time. During this time, the toxicant can slowly be
desorbed back into the water column or scoured into suspension.
Figure IV-46 shown earlier illustrated an idealization of the process
of desorption of a toxicant from bedded sediments. The process can be
described as follows. Supposed the average concentration of the pollutant
in the bedded sediment is X when the analysis begins (called t = 0). The
concentration X, at any later time is estimated from mass balance
considerations as:
X =
for x > k
(IV-129)
otherwise
533
-------
where
X = concentration of pollutant in bed at some time t = 0
M = mass of contaminated sediment per unit area of river bed,
g/cmz
U = stream velocity, cm/sec
6 = equivalent depth of water in sediment M , cm
K = partition coefficient
Equation IV-129 reveals that desorption can be interpreted as a frontal
phenomenon where desorption is completed at one location before progressing
downstream. Based on this interpretation, an effective removal velocity of
the front is:
Ue - {Li- (iv-130)
S P
The time T. required to desorb the toxicant over any specified distance is:
Td = XL/Ue (IV
where
x = length of contaminated river segment
During the period of desorption the average concentration in the water
column is:
C =
-2- for x> U t (IV-132)
KpD
0 , otherwise (IV-133)
To use Equations IV-129 through IV-133, estimates for XQ, M$, and 5 are
required. If both the mass of contaminated sediment per unit area of river
bed (M ) and the mass of toxicant in the sediments are known, then XQ can be
determined. Conversely, if both XQ and the total mass of toxicant in the
sediments are known, then M can be calculated.
534
-------
In lieu of having data on M and 6, these quantities can be estimated
based on the depth of contaminated sediments by using Table IV-39. In
addition to the depth, the percent solids by weight must be estimated. This
parameter generally increases with depths and can be chosen as 50 percent,
unless better data are available. The data in Table IV-39 were derived from
the following two equations:
Dc
-^ (IV-134)
and
6 - ~S C}^n( (IV-135)
where
MS = mass of contaminated sediment, g/m2
6 = equivalent water depth, mm
S = specific gravity of solids
D = depth of contamination, mm
In cases where the depth of contamination exceeds 100 mm the equations can
be used in lieu of Table IV-39.
The Hudson River in New York State provides an illustration of an
extreme case of PCB contamination (Turk, 1980). Between 1951 and 1977 PCBs
were discharged from point sources near Fort Edward and Hudson Falls, about
80 km (50 mi) above Albany, New York. Figure IV-51 shows the general
vicinity.
During this time period the mass emission rate of PCBs decreased from
15 kg/day (33 Ibs/day) to less than 1 g/day (0.002 Ibs/day). PCB
concentration in the bottom materials range from about 200 yg/g near Fort
Edward to about 4 yg/g near Waterford, about 70 km (43 mi) downstream. In
1975 the New York State Department of Environmental Conservation began a
study to determine the source of contamination. At that time they estimated
that the total mass of PCBs in the bottom sediments was 225,000 kg
(500,000 Ibs).
535
-------
TABLE IV-39
MASS OF CONTAMINATED SEDIMENTS AND EQUIVALENT WATER
DEPTH AS A FUNCTION OF DEPTH OF CONTAMINATION
Depth (mm) Percent Solids by Weight
1 20
50
80
5 20
50
80
10 20
50
80
20 20
50
80
50 20
50
80
100 20
50
80
Ms (g/cm2)
0.02
0.06
0.11
0.11
0.30
0.55
0.23
0.60
1.1
0.45
1.2
2.2
1.1
3.0
5.5
2.3
6.0
11.0
6 (mm)
0.9
0.6
0.3
4.5
3.0
1.4
9.1
6.0
2.7
18.
12.
5.5
45.
30.
14.
91.
60.
27.
536
-------
Hudson
i. Falls
Fort
Edward
N
0 5 10 15 KILOMETERS
FIGURE IV-51 LOCATION MAP OF HUDSON RIVER,
NEW YORK,
537
-------
It has been found that PCBs are being naturally desorbed from the river
bed under moderate and low flow conditions. The estimated transport rates
are:
• at Glen Falls = 0.0 kg/day (above discharge)
• at Schuylersville = 4.0 kg/day
• at Stillwater = 5.0 kg/day
• at Waterford = 4.0 kg/day (70 km downstream)
It is interesting to note that these transport rates are approximately 30
percent as high as the original point source mass emission rates. At a
desorption rate of about 4 kg/day, the river between Glen Falls and
Waterford would be rid of PCBs in about 150 years.
Turk (1980) found that flood events transport large quantities of PCBs,
although this transport mechanism is only operative periodically. Turk
estimated that due to the combined removal rates of PCBs during high flow
periods (by scour) and during low flow periods (by desorption), the
residence time of PCBs above Waterford would be about one century.
•EXAMPLE IV-18
For discharges of 600 m3/sec or less, it has been found that the Hudson
River bed provides 4 kg/day of PCBs to the water column at locations between
Schuylersville and Waterford, New York. Determine the PCB concentration in
the water column at the following two flow rates:
a. 600 mVsec
b. 50 m3/sec.
Compare these concentrations to the freshwater criterion of 0.001 yg/1
promulgated in the "Red Book".
Since the mass emission rate and river flow rate are known,
Equation IV-11 can be rearranged to yield the total instream concentration:
M
Ct = 86".~4Q
538
-------
where
M = mass loading, kg/day
C. = concentration of pollutant, ppm
L
Q = flow rate, m3/sec
For the problem at hand,
M = 4 kg/day
Q = 50 and 600 m3/sec
For Q = 600 m3/sec,
4
= °-08 x 10"3
- 0.08 iig/1, or 80 times the Red Book criterion
For Q = 50 rnVsec,
CT s 86.Tx~50 = °'9 X 10"3 Ppm
= 0.9 yg/1, or 900 times the criterion
As a second part to the problem estimate the time required to remove
the PCBs in the sediment by desorption (ignoring scour), assuming the
desorption rate of 4 kg/day is not known. Base the calculations on Table
IV-39 or Equations IV-130 and IV-131. Use the following data:
• depth of contaminated sediment = 600 mm
• river velocity = 1 fps
• partition coefficient : 103 to 10"
Because the depth of contamination exceeds the maximum value tabulated in
Table IV-39, Equations IV-134 and IV-135 are used instead. Assuming
S = 1.5 and P = 80,
539
-------
1 1.5
\
- 65 9/cm2
10 1 1.5 80
\
- 1.5 x 600
---
/ \
(100-80)
V 80 / _ 1Cn mm _ 1C
100-80\ " 16° m ~ IB cm
1 + 1
/100-80\
.5 ( 80 j
The effective transport velocity is:
Ue = eH-lf* = '25 X ^ U f°r KP ' 10'
and
- -25x10- u forKp=103
The time required for desorption over the 70 km (43 mi) reach is:
T = .253x"lO->0x 1 Sec ' 29°
and
T = 29 years for K = 103
Probably the biggest unknown in this problem is K . Based on a range of K
from 103 to 104, the time of desorption ranges from 29 to 290 years, within
the range predicted from observed desorption rates.
- END OF EXAMPLE IV-18 -
4.9.3.5 Instantaneous Releases of Low Density Toxicajrts
Many toxicants have specific gravities less than or equal to unity.
Should a toxicant less dense than water be spilled in its pure form, the
toxicant can ride atop the water body for a period of time, while (perhaps)
being rapidly volatilized and photolyzed as it becomes entrained and
dissolved in the river.
540
-------
Analysis of releases of low density pollutants is complicated and, in
many cases, beyond the scope of hand calculation analyses. Often spills of
toxicants occur over a part of the river, so the resultant movement is
three-dimensional because the toxicant spreads laterally, longitudinally,
and vertically due to turbulence and advection. Buoyant spreading and
mixing can further complicate the dispersal process.
Toxicant spills can occur in numerous ways. In one instance the
toxicant may be discharged directly onto the surface of the river, and
depending on the rate of mixing with ambient water a significant portion
could volatilize directly from the pure phase. On the other hand submerged
spills may result in the chemical becoming mixed with river water before it
reaches the water's surface. Under these circumstances volatilization
fluxes will not be as great.
When a chemical is spilled in pure form, the time required for the
chemical to mix with the river water depends, in addition to other factors,
on the solubility of the chemical. Some chemicals may be highly soluble in
water (essentially infinitely soluble) while others may have a very low
solubility. Figure IV-52 illustrates these two different situations.
Figure IV-52a shows the case of a toxicant of infinite solubility. The
toxicant maintains its pure state (mole fraction equals unity) for some
distance away from the spill site, and then the mole fraction gradually
begins to decrease as the chemical mixes with water. Figure IV-52b
illustrates the same basic situation, except that the toxicant has a finite
solubility. In this case there will be a rapid drop between the mole
fraction of the pure toxicant (unity) and the mole fraction at solubility
(much less than unity). For the pure phase toxicant shown in Figure IV-52b
to become mixed in water at concentrations at or below the solubility limit
might require a substantial amount of water.
Based on the discussions in the previous paragraphs, tools for analyses
of the following will be presented:
541
-------
1.0
MOLE
FRACTION 0.1
0.01
0.0
MOLE
FRACTION
PURE
CHEMICAL
(a)
1.0
0.1
0.01
o.o
PURE
CHEMICAL
Solubility
INFINITE SOLUBILITY
FINITE SOLUBILITY
Distance from Spill Centerline
(b)
FIGURE IV-52 HYPOTHETICAL CONCENTRATION DISTRIBUTIONS
OF FINITELY SOLUBLE AND INFINITELY
SOLUBLE TOXICANTS,
542
-------
• Volatilization mass flux from a pure toxicant contained within
a fixed area.
• The fate of a highly soluble toxicant which mixes
instantaneously with the river water.
• The fate of a low solubility toxicant which mixes with a
finite volume of river water.
Toxicants which exist in the gas phase under atmospheric pressure and
typical natural temperatures are excluded from analysis here, even though
they might be transported as a liquid under high pressure (e.g. liquified
chlorine). If a tank transporting such a chemical were to rupture under
water, the chemical would boil and most of the gas would rapidly escape into
the atmosphere. Some of the gas would however become dissolved in the river
water during ascent of the bubbles.
4.9.3.5.1 Volatilization of Toxicant in Pure State
This section will present a method to predict the volatilization mass
flux rate of a pure chemical, and the time required to volatilize a known
amount of the chemical which occupies a specific area of river surface. The
volatilization flux rate is given by:
MW k • P • MW
IV-136
_ _ __ _
RUT 82.136 T
where
F = volatilization mass flux, g/cm2/hr
P = saturation vapor pressure of
MW = molecular weight of toxicant
P = saturation vapor pressure of pure liquid toxicant, atm
T = temperature of ambient water, K
k = gas phase transfer coefficient, cm/hr.
543
-------
The expression for the gas phase transfer coefficient was shown earlier in
Equation 11-46. A suggested default value is 3000 cm/hr. The saturation
vapor pressure of a number of toxicants were shown earlier in Tables II-5
through II-9. Weast (1969) also contains extensive data.
Based on the amount of pure toxicant (M) contained within a spill area
(A), the time required to volatilize the chemical is:
* - icnrnnr {IV-137)
where
t = time, hr
M = mass, kilograms
A = area, m2
This expression is limited to situations where the spill area A is of
constant size over the period of volatilization so it is not applicable to
unconfined spills where the area could change rapidly with time.
4.9.3.5.2 Analysis of Chemicals Which Instantaneously_M ix
Depending on the mass of spilled toxicant and its solubility, the
spilled toxicant may rapidly attain concentrations in the water column below
solubility. Under these circumstance, Equation IV-138 presented below can
be used. The analysis below assumes that Henry's Law is valid (e.g. the
mole fraction of dissolved chemical is much less than 1.0) which is
reasonable for many toxicants only moments after a spill.
It is worthwhile to calculate the volume of water required for a mass M
of spilled chemical to be diluted to its solubility limit. This can provide
a rough idea as to whether mixing is likely to be "instantaneous" or not.
Suppose that a mass M of spilled chemical has a solubility C$. The volume
of water needed to be mixed with the pure chemical so that the solubility
limit is achieved is:
544
-------
V = -A- (IV-138)
0 Cs
where
M = mass of spill , kg
C = solubility, mg/1
V = volume of water, m3
The concentration profile resulting from an instantaneous spill (and
assuming concentrations at or below the solubility limit are rapidly
attained) is expressed by:
( IV-139)
where
C = dissolved phase concentration
k =
e 1 + KpS
M
M = total mass released
and the remaining variables have been previously defined. In most instances
the user would like to predict the maximum concentrations remaining in the
river for different elapsed times following the spill, given by the peaks in
Figure IV-53. Under such conditions, and assuming P = 0, Equation IV-139
simplifies to:
L) f I 4- \
(IV-140)
The various components of the mass balance at time t follow (for
P = 0).
Mass of dissolved pollutants Mp (t = t )_
MD(t = ts) = MD exp (-kets)
545
-------
tn
CD
c:
33
m
o -<
O -D
> o
H H
•— X
o m
O
-n >
o r-
o --
s: v>
— H
z ^3
CD •— •
&
> cr
oo .-
-D O
o
X
I—«
o
o
c:
C/)
o
en
o
CD
CD
CQ
CQ
CO
•o
Cn
°
8
Concentration^ (PPB)
Cn O oi O
8
|
co
8
ro
o
o
CD
en
-------
Mass of sorbed po Hut ants Mg (t = tg ) :
Ms = KpS MD exp(-kets) (IV-142)
Mass of pollutant which has volatil ized M . (t = t^ ) :
„ - , . ---- — _ .
Mv
-------
4.9.3.6 Spill Analysis of High Density Toxicants
Spills of hazardous chemicals have been of concern for quite a number
of years, and interest will increase as the quantity and variety of
toxicants transported increase. In past years the primary emphasis has been
on analysis and containment of oil spills. This has probably been for a
number of reasons:
• Large quantities of oil are transported, and are therefore
subject to more frequent spills.
• The environmental consequences of an oil spill can be severe
and visually offensive.
• Oil floats, so oil spills are easy to detect and monitor.
In contrast to oil, many hazardous chemicals have specific gravities
greater than one, so that in their pure form, they tend to sink in water.
Table IV-40 lists some such chemicals. Chlorine, although it may be
transported under pressure as a liquid, is a gas under atmospheric
conditions. Even so, if a liquid chlorine barge were involved in an
accident on a river some of the chlorine could become dissolved in the water
since the solubility of chlorine in water is 50,000 mg/1, although most
would probably gasify and form a toxic cloud.
The chemicals shown in Table IV-40 are generally either slightly
soluble (10 to 10,000 ppm) or soluble (10,000 to 1,000,000 ppm). In any
case the solubility levels generally exceed or greatly exceed proposed water
quality criteria. Thus if a mass of chemical were spilled into a river, it
is to be expected that concentrations near the chemical's solubility limit
could be detected in the immediate vicinity of the spill. As the chemical
is dissolved and travels downstream, it could eventually become mixed over
the channel cross-section and expose all organisms living within the water
column (and perhaps those living in the bedded sediments as well) to its
effects. With increasing distance the concentrations of the toxicant will
decrease to reflect the additional mixing afforded by the flow of the entire
river, plus dispersion, degradation, and volatilization processes.
548
-------
TABLE IV-40
WATER-SOLUBLE, HIGH DENSITY (p>l), IMMISCIBLE CHEMICALS
Species
Acetic acid
Acetic anhydride
Acetophenone
Aniline
Benzaldehyde
Benzyl alcohol
Bromine
Carbon disulfide
Carbon tetrachloride
Chlorine (liquid)
Chloroform
Chloropthalene
Dichloroethane
Ethyl bromide
Ethyl ene bromide
Furfural
Glycerol
Hydrogen peroxide
Mercury
Naphthalene
Nitrobenzene
Phenol
Phenylhydrazine
Phosphorus trichloride
Trichloroethane
N-Propyl bromide
Quinol ine
Tetrachloroethane
Waterb
Density
in air
(g/cm3)
1.06
1.087
1.03
1.022
1.04
1.043
2.93
1.26
1.595
3.2
1.5
1.256
1.431
2.18
1.159
1.26
1.46
13.54
1.15
1.205
1.071
1.097
1.5
1.325
1.353
1.095
1.60
1.00
Solubility a
in water Interfacial Tension (dynes/cm)
(mg/1) Air Water Vapor
50,000 68.0,no ,, o
JU " C 1 . o^rjo
500,000 - - 32.720c
5,550 - - 39.8200
34,000 44.0 - 42.92QO
1,000 40.04 15.51200
46,000 39.02QO 4.7522 50 39.02C)0
41.700 41.520<> - 41.52QO
2,200 - 48.362Qo
500 - 45200 26.952[)0
50,000 - - 18-420°
5,000 27.142QO 32.82QO
40.74200
9,000 23-4350
10,600 - 31.2200 24.1520o
4,300 - 36.542[)0 38.3720o
83,100 43.5,no - 43.5,no
C\J C\J
63.4180
50,000 - - 76-liQ 2°
.0005 470 37520°
"5H ?P C 9Q Q
JU tti.tt^jo ~ ^o.B.|270
1900 43.92QO - 43-920°
67,000 40.92()0 - 40.02()0
46.12QO
50,000 - - 29.12QO
10 22H40
2,500 - - 19-652()0
60,000 45.02QO
3,000 36.322 50 -
N.A. 73.05]go N.A. 72
In air, water, and its own vapor. Temperature is °C.
Under pressure.
Mercury and v/ater data included for reference.
From: Thibodcau* (1979)
549
-------
A technique is presented here to estimate the concentration which can
exist in the water column and the duration of the elevated levels following
a spill. In particular tools are presented to predict:
t The concentration of toxicant in the water column at the
downstream end of the spill.
• The concentration of the toxicant after it has become
completely mixed with the entire river.
• The time required to dissolve the spilled toxicant.
• The amount of toxicant remaining sorbed to the bottom
sediments and in the pore water following dissolution.
It is, of course, more accurate but more costly to measure concentrations
directly rather than predicting them. However, since the toxicant is
"somewhere" on the river bottom, and might not be immobile, detecting the
location of the toxicant will take time. By estimating the dissolution time
of the spill, it can be determined if it is feasible to even set up and
carry out a sampling program.
The tools delineated above are useful not only to analyze spills which
have occurred, but also for answering hypothetical questions which relate to
the consequence of spills based on river traffic, sizes of containers, kinds
of toxicants being transported, and characteristics of the rivers. Based on
this information the user can evaluate possible "spill scenarios" to predict
impacts before they occur. Such information would be useful to formulate
post-spill responses. In situations where a spill of a toxicant would
produce extreme consequences, provisions could be made to mitigate the
consequences before they occur.
550
-------
4.9.3.6.1 Description of Spill Process
Spills which contaminate rivers can be the result of a variety of
accidents: leaking barges, broken pipelines, highway accidents, and
clandestine dumping. The scope here is limited to those situations where
the toxicant has been deposited on the bottom of the river. This situation
is most likely to result from an accident on or under the water's surface.
Figure IV-54 conceptualizes what might happen when a barge carrying a high
density pollutant ruptures.
Depending on the volume of contaminant, the size of the hole, among
other factors, the toxicant might issue from the barge as a continuous jet.
However, because the volumetric flow rate of the jet is probably small, and
perhaps even intermittent, the toxicant probably breaks up into drops of
various sizes as it falls through the water column. Some of the finest
drops might never reach the stream bed, but rather be transported in
suspension within the water column, and gradually dissolve. The majority of
the toxicant may settle on the river bed and form drops, globs, or pools
(using the terminology of Thibodeaux, 1979). The drop size depends on the
intrafacial tension and density differences between the toxicant and the
water (Hu and Kintner, 1955). Pools tend to form in the valleys of sand
waves, and occur when large drops or globs coalese. Thibodeaux (1980)
provides techniques to estimate the residence time of drops, globs, and
pools. For the simplified analyses here the spill is assumed to be in the
shape of a continuous pool.
551
-------
Barge
River bottom.
sand dune-
"Water column
-velocity profile
Medium drops
.Fine drops
/i me UIU^ION .... -.
I * ".'.'".
V ."-• •-••.-••• •••.-.• •
v . •-. - —.•—..• •
A .-.•-....•;•: . .-• •••
,-•••••:•••••:.' r».**
crest and valleys. ——
Large drops
W//////77fr/7//77W/7/7v7777////////y/7777Zw
Envelope of zone of contamination/ Droplet Droplet-glob Pool
zone
zone
zone
FIGURE IV-54 ILLUSTRATION OF HYPOTHETICAL SPILL INCIDENT
(FROM THIBODEAUX, 1979),
4.9.3.6.2 Fate of Pollutant Following Settling
Once the toxicant has settled on the river bed its fate is governed by
numerous processes. Depending on the texture of the bottom materials
(e.g. sands, cobbles, boulders), the density of the toxicant, and its
interfacial tension, the toxicant could settle in deep depressions, and
dissolution would be slowed.
Many pollutants have large partition coefficients so that sorption to
bottom sediments is significant. The characteristics of the sediments
affect the partition coefficient, but in many cases sorption can compete
with dissolution as a major process controlling the pollutant's fate.
Although transformation processes other than sorption and dissolution are
operative the moment the toxicant enters the water, they are not considered
here.
552
-------
In September 1974 an electrical transformer being loaded onto a barge
fell into the Duwamish Waterway in the State of Washington (Thibodeaux,
1980). 260 gallons of Aroclor 1242, a PCB mixture of specific gravity 1.4,
were spilled into the river. Divers observed that pools of free PCB on the
bottom moved back and forth with the tide. Pools of PCBs were removed from
the bottom using suction dredges, and a second stage operation involved a
high solids dredge. Probably due to its low solubility (0.2 ppb) and high
sorption characteristics, much of the PCB was recovered (from 210 to
240 gallons).
4.9.3.6.3 Predictive Tools
It is hypothesized that a toxicant spill contaminates an area of width
W and length L , where the length is measured in the flow direction. The
toxicant which reaches the river bed is assumed to be highly concentrated,
and its dissolution is controlled by a thin layer immediately above where
molecular diffusion limits the vertical flux of the pollutant. Above this
layer the toxicant is rapidly entrained into the river. There are several
expressions available to determine the thickness of the diffusion layer
(e.g. Novotny, 1969 and Mills, 1976). The expression developed by Mills
will be used here, because the required information is easier to attain
while the two approaches appear to give comparable results. The expression
is:
11.6 • 1.49 v R, V6
6 = _ _J! (IV-147)
" vg (Jn
where
6, = thickness of diffusive sublayer
v = dynamic viscosity of water
R, = hydraulic radius of the river
U = river velocity
n = Manning's coefficient
553
-------
Just downstream from the spill zone, but before complete mixing with the
river, the concentration of the toxicant in the water column is:
CL = (CQ-CS) exp ^-^JLj + cs (IV-148)
where
C = background concentration of chemical
C = solubility of chemical in water
D = diffusion coefficient of chemical in water
cw
H = water depth
U = river velocity
The concentration at the location of complete mixing is
CWM = CLT+ co(1 -Ir1 {IV-149)
where
W = spill width
W = river width
The time T. required to dissolve the chemical is:
Td = r-Ju- (IV-150)
where
M = total amount of pollutant which is dissolved (an amount less
than or equal to the amount spilled).
As the spilled toxicant dissolves in the flowing river water, it
concurrently diffuses into the immobile bedded sediments, where a portion is
sorbed onto the sediments. Consequently, some residual toxicant will remain
in the bottom sediments following the initial dissolution phase. The
residual will then slowly diffuse and desorb back out into the river,
although diffusion deeper into the sediments can also occur because of the
concentration gradient. The time required for the residual toxicant to
554
-------
naturally desorb and diffuse back into the water column can greatly exceed
the original period of dissolution.
The quantity of toxicant which resides in the sediments following the
initial dissolution period can be predicted as follows. It is assumed that
the dissolution and downward diffusion/sorption proceed independently until
all the spilled toxicant has been removed. The time t can be found such
that this statement is true. From a practical standpoint, the user can
simply determine the time required for complete dissolution, and then find
the total mass which would have diffused/sorbed into the bottom sediments
during this period. Since this approach accounts for more toxicant than was
originally present, the time period should be decreased by the fractional
amount of toxicant created. If the amount of excess toxicant is no more
than 15 percent of the total amount spilled, then a time adjustment is not
required.
Based on the processes of sorption and diffusion the vertical profile
of dissolved chemical in the river bed at time t following the appearance of
the toxicant on the bottom is given by:
C - Cb
C -C
s L
where
C = concentration of dissolved chemical in the pore water, in
units of mass of dissolved chemical per unit volume of pore
water
C^ = background concentration of chemical in pore water
Cs = solubility of chemical in water
z = vertical distance, measured downward from the
sediment-water interface
555
-------
De = effective molcular diffusion coefficient
PS = density of sediments
Kp = partition coefficient
n = porosity of porous medium
From Equation IV-151, the total mass of pollutant found in the
sediments at time t is:
/•/ \
ndV (IV-152a)
:sjdz (IV-152b)
where
AC = spill area
concentrati
volume of pore water
C = concentration of pollutant sorbed to sediments, per unit
C can be related to C by:
Cs = CPskp (ITj (IV-153)
Combining Equations IV-151, IV-152 and IV-153 the total mass in the sediment
is:
MT = 0.563nCs l + pk AV4 (iV-154)
•EXAMPLE IV-19-
The following is an excerpt from Chemical Engineering Volume 80,
September 3, 1973, as reported in Thibodeaux (1979).
556
-------
"Approximately 1.75 x 106 Ibs of chloroform were released
from a barge that sank near Baton Rouge, Louisiana, and the
chemical began flowing down the Mississippi River toward the
Gulf of Mexico. Although state health officials did not push
the panic button, noting that they did not anticipate too much
trouble from the accident, the U.S. Coast Guard warned
downriver communities to keep a close surveillance on their
water supply systems, particularly if intakes were close to
the river bottom (chloroform is heavier than water).11
Based on the low flow conditions and the time history of the chloroform
concentration much of the chloroform (of specific gravity 1.5) was initially
deposited on the river bed. Determine the fate of the chloroform during the
first few days following the spill. The following processes are considered:
• dissolution into the main body of the water,
• diffusion and sorption into the bottom sediments,
• volatilization into the atmosphere, and
• sorption to suspended sediments.
Since chloroform is highly volatile and does not have a strong tendency
to sorb to solids, volatilization is an important process controlling its
fate, while sorption is not. The following analysis substantiates this
statement.
The data pertinent to the spill are (Thibodeaux, 1979; Neely
et al_., 1976):
• river flow rate = 7590 m3/sec (268,000 cfs)
• width of river = 1220 m = 4000 ft
557
-------
0 river velocity = 56.3 cm/sec = 1.85 ft/sec
• water depth = 11 m = 36.3 ft
• diffusion coefficient of chloroform in water = 1x10 " cmVsec
• length of spill zone = 180 m = 590 ft
• background chloroform concentration = 5 ppb
Using a Manning's n of 0.03, the diffusion layer thickness is:
.. _ 11.6 x 1.49 x .915 x 10~5 / — M/R , -2
6 ' ~ -- " X 36 7 = =
-
'8 X 10
The average concentration of chloroform in the water just below the spill
zone during the period of dissolution is:
C = (5 x 10-3 - 8200) exp— A
8200
2.8 x 10~2 x 11. x 56. 3
= 850 ppb
In order to estimate the time required to dissolve the chloroform the
average width of the spill zone is required. The width is estimated to be
256 ft (78 m) (Thibodeaux, 1981).
Based on these data the dissolution time is:
T _ .0.9 x 1.75 x_ 106 _. ,
'd ~ 5738 ~x .850 x T. 85 x" 156 T 3^3 = ^ days
The factor 0.9 is used in the above expression because about 10 percent of
the spill dissolved before ever reaching the bottom (Neely et_ aj_. , 1976).
The amount of chloroform which diffused and sorbed into the sediments
during this time period (20 days) will be estimated. The porosity of the
sandy bottom is approximately 0.35, and the partition coefficient is assumed
to be 1.0. This is a realistic value based on K = 93 (see Table II-5).
The total mass contained i-i the sediments after 20 days is:
-------
MTota, - .35 (180 x 78) (W.6S x 1 x
.35
x 10-2-3 x (5-4.437) ~6000 kg
6000 kg is less than 2 percent of the total mass which reaches the bottom
(715,000 kg). Based on this result, it is not likely that the dissolution
period is markedly affected by diffusion of the chloroform into the bottom
sediments. Because of the vertical concentration gradient that has been
established in the sediment profile, some of the chloroform will temporarily
continue to diffuse downward after the dissolution period. Hence
concentrations in the water column due to desorption of the chloroform and
upward diffusion back into the water column are not likely to be high
compared to those observed during the initial dissolution period.
Following the chloroform spill, chloroform concentrations were measured
at several locations in the Mississippi River below the spill. Figure
IV-55a shows the time history of the chloroform concentration at a location
16.3 miles below the spill for the first 60 hours following the spill. A
more compressed time scale is shown in Figure IV-55b and illustrates how the
concentrations varied for 20 days following the spill. The peak
concentration passes very rapidly (on the order of 1 day) and the maximum
observed concentration is about 365 ppb. At this location, the chloroform
is approximately well-mixed with the river at this point (Neely et _a1.,
1976).
Based on Figure IV-55b the total amount of chloroform passing the
location can be estimated as follows:
Mass = /CQdt = Q/ Cdt
=/CQdt = Q/"
The right-most integral is simply the area under the concentration-time
curve in Figure IV-555. Without showing the calculations, the total mass of
chloroform (above background) which passes the location 16.3 miles below the
spill is about 300,000 kg. Since the total amount of chloroform spilled was
about 800,000 kg, more than half of the chloroform was unaccounted for. It
is unlikely, as earlier calculations showed, that diffusion and sorption
559
-------
350
300
I 250
c
o
i 200
-------
400
300
o
5
Q.
-------
into the bottom sediment was significant. Volatilization could be important
arid will be discussed shortly.
The observed results shown in Figure IV-55a are compared against those
predicted in this example. A concentration of 850 ppb was predicted just
below the spill site; the maximum shown in Figure IV-55a is 365 ppb. It is
expected, for several reasons, that the concentrations 16.3 miles below the
spill site will be less than at the spill site. First it is probable that
additional dilution occurred as the chloroform was transported to the
sampling site. An estimate of the dilution can be attained by multiplying
the river width by the spill width, or:
4000 _ ,,
260 "
The well-mixed concentration becomes:
850
,n ,
= 60 ppb
Comparing this to Figure IV-55a, it is noted that this value approximates
the average concentration following an elapsed time of about 20 hours, but
misses the peak during the first 20 hours. There may be a number of factors
responsible for this behavior, and one of the most important will be
examined here. During the spill of chloroform, it was estimated that about
10 percent, or 80,000 kg were transported downstream directly without ever
reaching the river bottom. The travel time to the sampling site is:
16.3 mi - _ ,
= 13 hours
Figure IV-55a shows that this coincides with the arrival of the peak at mile
16.3. The peak concentration can be estimated using Equation IV-140
presented earlier. The diffusion coefficient is approximately 210 m 2/sec
(McQuivey _ejt aj_. , 1976) for the lower Mississippi River. The predicted peak
in concentration at mile 16.3 is:
80000 x 103
C = ------ — - — - — ----- = 520 ppb
2 x 4000 x 36.3 x (.3048)2 Vn-210-3600- 13
562
-------
This concentration is somewhat higher than the maximum 365 ppb observed, but
this is to be expected since Equation IV-140 assumes the mass is input
instantaneously, while in reality about 8 hours elapsed. Further if the
concentration due to the dissolved portion of the spill is calculated at 20
hours, a concentration of 15 ppb is obtained. This illustrates that the
mass due to initial dissolution has almost passed the sampling location, and
the remaining contribution to the elevated concentrations measured is due
largely to dissolution of chloroform which has settled on the river bottom.
It appears that there are two basic phenomena which account for the measured
concentration-time profile: an initial period of dissolution of chloroform
(less than 1 day) before it settles to the bottom, and a subsequent period
(10 to 15 days) of dissolution of settled chloroform.
The absence of an adequate mass balance between the amount of
chloroform which entered the river as a result of the spill and the amount
which passed a location 16.3 mi below the spill has not been addressed.
Volatilzation losses could be one reason for the imbalance.
Equation IV-123 can be used to estimate the volatilization losses.
Since the chloroform was initially deposited on the bottom of the river,
during a portion of the travel distance it was not in contact with the
atmosphere, and so volatilization could not occur. The approximate travel
time for vertical mixing to occur is (Fischer et a_l_., 1979):
L =
where
H = water depth
e = vertical diffusivity
Choosing an ez value of 50 cm2/sec, based on Fischer e_t aJL (1979) and a
depth of 11 m, the travel time required to effect vertical mixing is
hr = 2.7 hrs
563
-------
Based on a velocity of 1.85 ft/sec, the travel distance is about 3.3 miles.
Hence the pollutant is in contact with the atmosphere for about 13 miles.
Since only the dissolved phase of chloroform volatilizes, the fraction
of the total chloroform as solute will be estimated using Equation IV-109:
c-c
__
Hkp,S
The partition coefficient K was estimated as 1.0. The sediment
P
concentration is about 400 ppm. Hence:
c . i =1.0
Ct 1 + 1 x 400 + 10~6
Thus, essentially all the chloroform is dissolved and is available for
volatilization.
Henry's Law constant for chloroform can be found based on the data in
Table II-5:
• vapor pressure = 150 Torr
• solubility in water = 8200 ppm
• molecular weight = 118.
Henry's Law constant is:
= 3 v in
-3
760 x 8200 mole
From Table 11-15 a typical volatilization rate is about 17 cm/hr.
The average chloroform concentrations for the 13 miles above the data
collection point are:
200 ppb for 1 day
40 ppb for the next 9 days
10 ppb for the next 9 days
564
-------
The total amount of chloroform volatilized is (using Equation IV-109):
Ik C. A At
v i c
= 0.17 x 24 x 1200 x 21 x 103(200 + 40 x 9 + 10 x 9 -5 x 19)x 103
= 5.8 x 107 = 58000 kg
g
Hence, all of the unaccounted for chloroform (about 480,000 kg) could not
have volatilized within 13 miles.
Over 50 percent of the chloroform still remains unaccounted for. It is
possible that other transformation processes were operative. The
enviromental fate of chloroform in terms of photolysis, hydrolysis,
oxidation, and biological degradation was reviewed in Callahan et_ a]_., 1979.
It was concluded that these processes are of minor importance compared to
volatilization and so are probably not significant here.
It is possible that the samples of chloroform shown in Figure IV-55b
were not cross-sectional averages. The chloroform concentration could have
been weighted toward the stream bottom or toward one side. A dye study
performed by McQuivey (1976) on the lower Mississippi River showed that
50 miles were required before complete mixing was attained, while the
sampling was conducted 16.3 miles below the spill. Even though chloroform
does not sorb strongly, there is a possibility that the suspended solids and
bed load concentration near the bottom of the river were high enough to
cause substantial sorption. Based on the evidence there is a distinct
possibility that some of the "missing" chloroform was actually advected past
the sampling locations without being detected.
•END OF EXAMPLE IV-19'
565
-------
REFERENCES
Alonso, C.V., J.R. McHenry, and J.-C.S. Hong. 1975. "The Influence of
Suspended Sediment on the Reaeration of Uniform Streams" Water
Research. Vol. 9 pp 695-700.
American Public Health Association. 1973. Standard Methods for the
Examination of Water and Waste Water. 13th Edition. American Public
Health Assn., Washington, D.C.
Bansal, M.K. 1975. "Deoxygenation in Natural Streams", Water Resources
BuTJetjm. Vol.11, No.3, pp 491-504.
Barnes,H.H. 1967. Roughness Characteristics of Natural Channels.
Geological Survey Water-Supply Paper 1849.
Barrett, M.J., A.L. Gameson, and C.G. Ogden. 1960. "Aeration Studies of
Four Wier Systems, "Water and Water Engineering. London.
Bessler, M.B., and J.T. Maletic. 1975. "Salinity Control and Federal Water
Quality Act," American Society of Civil Engineers, Journal of the
Hydraulics Division;Vol.101, HY5, pp 581-594.
Brady, O.K., W.L. Graves, and J.C. Geyer. 1969. Surface Heat Exchange at
Power Plant Cooling Lakes. Edison Electric Institute.
Brown, G.W. 1969. Predicting Temperature of Small Streams. Water
Resources Research. Vol.5, No.l.
Butts, T.A. and R.L. Evans. 1978. Sediment Oxygen Research Studies of
Selected Northeastern Illinois Streams. Office of Water Research and
Technology.
Callahan, M.A. et al., 1979. Water Related Environmental Fate of 129
Priority Pollutants. EPA-440/4-79-029.
Carnahan, B., H.A. Luther, and J.O. Wilkes. 1969. Applied Numerical
Methods. John Wiley & Sons.
Chow, V.T. 1959. Open Channel Hydraulics. McGraw-Hill.
Committee on Water Quality Criteria, National Academy of Sciences and
National Academy of Engineering. 1973 Water Quality Criteria 1972.
U.S. Environmental Protection Agency, Washington, D.C.,
EPA-R3-73-033.
Covar, A.P. 1976. Selecting the Proper Reaeration Coefficient for Use in
Water Quality Models. Presented at the U.S.Environmental Protection
Agency Conference on Environmental Simulation and Modeling, April
19-22.
Cragwall, J.S. 1966. "Low-Flow Analysis of Streamflow Data," 5th Annual
Saitary and Water Resources Engineering Conference. Nashville, Tenn.
Edinger, J.E. 1965. Heat Exchange in the Environment. John Hopkins
University, pg 43.
566
-------
Edinger, J.E., O.K. Brady, and J.C. Graves. 1968. "The Variations of Water
Temperature Due to Electric Cooling Operations," Journal of Water
Pollution Control Federation. Vol.40, No.9, pp 1637-1639.
Edinger, J.E., and J.C. Geyer. 1965. Heat Exchange in the Environment.
Edison Electric Institute, Publication 65-902.
Erdmann, J.B. 1979a. Systematic Diurnal Curve Analysis. Journal Water
Pollution Control Federation, Vol.51,No.1.
Erdmann, J.B. 1979b. Simplified Diurnal Curve Analysis. Journal
Environmental Engineering Division, ASCE, Vol.105, No.EE6.
Fenwick, G.B. 1969. State of Knowledge of Channel Stabilization in Major
Alluvial Rivers. Corps of Engineers, U.S. Army, Technical Report
No. 7.
Fischer, H.B., E.J. List, R.C.Y. Koh, J. Imberger, N.H. Brooks. 1979.
Mixing in Inland and Coastal Waters. Academic Press, pg.483.
Gameson, A.L., K.G. Vandyke, and C.G. Oger. 1958. The Effect of
Temperature on Aeration at Wiers, Water and Water Engineering.
London.
Graf, W.H. 1971. Hydraulics of Sediment Transport. McGraw-Hill Book
Company; New York.
Hu, S. and R.C. Kintner. 1955. The Fall of Single Liquid Drops Through
Water. Amer.Inst.Chem.Eng.J., Vol.1, No.l, pg 42.
Hydrologic Engineering Center, Corps of Engineers. 1975. Water Quality
Modeling of Rivers and Reservoirs. U.S. Army Corps of Engineers.
Hydroscience, Inc. 1971. Simplified Mathematical Modeling of Water
Quality. U.S. Environmental Protection Agency, Washington, D.C.
Jones, H.G.M., H. Bronheim, and P.F. Palmedo. 1975. "Electricity
Generation and Oil Refining," Mesa New York Bight Atlas. New York Sea
Grant Institute, Albany, New York, Monograph No.25.
Kelly, M.G., G.M. Hornberger, B.J. Cosby. 1975. A Method for Monitoring
Eutrophication in Rivers. University of Virginia.
Karmondy, E.J. 1969. Concepts of Ecology. Prentice-Hall, Inc.; Englewood
Cliffs, New Jersey.
King, H.W. 1954. Handbook of Hydraulics. McGraw-Hill Book Company, Inc.,
New York.
Krenkel, P.A., and F.L. Parker. 1969. Biological Aspects of Thermal
Pollution. Vanderbilt University Press.
Lehman, J.T., D.B. Butkin, and G.E. Likens. 1975. "The assumptions and
Rationales of a Computer Model of Phytoplankton OPopulation Dynamics,"
Lj'mnology and Oceanography, Vol.20(3), pp 343-362.
Linsley, R.K., M.A. Kohler, and J.H. Paulhus. 1958. Hydrology for
Engineers. McGraw-Hill Book Company; New York.
567
-------
Lund, J.W.6. 1965. "The Ecology of Freshwater Phytoplankton," Biol.Rev.
Vol.40, pp 231-293.
Mastropietro, M.A. 1968. "Effects of Dam Reaeration on Waste Assimilation
Capacities of the Mohawk River," Proceedings of the 23rd Industrial
Waste Conference. Purdue University.
McElroy, A.D., S.Y. Chiu, J.W. Nebjen, A. Aleti, and F.W. Bennett. 1976.
Loading Functions for Assessment of Water Pollution from Nonpoint
Sources. U.S. Environmental Protection Agency, Washington, D.C.,
EPA-600/2-76-151.
McQuivey, P.S. and T.N. Keefer. 1976. Dispersion-Mississippi River Below
Baton Rouge, LA. Journal of Hydraulics Division, ASCE. Vol.102, HY10.
Miles, J.R.W. 1976. Insecticide Residues on Stream Sediments in Ontario,
Canada. Pesticide Monitoring Journal, Vol.10, No.3.
Mills, W.B. 1976. A computational Model That Simulates Biofilm
Nitrification in Streams. Engineer's Thesis, Stanford University.
Mills, W.B. 1979. A Nutrient Analysis of the Snake River and Its
Tributaries Within the State of Idaho. Tetra Tech Report TC-3923.
Prepared for Idaho Department of Health and Welfare.
Mills, W.B., V.H. Colber, and J.D. Dean. 1979. Hand Held Calculator
Programs for Analysis of River Quality Interactions. Supplemental
Volume to: Water Quality Assessment, A Screening Method for
Nondesignated 208 Areas.
Mills, W.B. 1981. Workshop on Screening Methods for Conventional and Toxic
Pollutants in Rivers; unpublished notes.
Montana State Dept. of Health and Environmental Sciences. 1973. Water
Quality Standards. MAC 16-2.14(10)-S14480.
Neely, W.B., G.E.Blau, and T.Alfrey,Jr. 1976. Mathematical Models Predict
Concentration-Time Profiles Resulting from Chemical Spill in a River.
Environmental Science and Technology, Vol.10, No.l, pp 72-76.
Nemerow, N.L. 1974. Scientific Stream Pollution Analysis. Scripta Book
Company; Washington, D.C.
Novotny, V. 1969. Boundary Layer Effects on the Course of the
Self-Purification of Small Streams, in S.H. Jenkins, Ed., Adv.Water
Pollution Res., Pergamon, pp 39-50.
Novotny, V., and P.A. Krenkel. 1975. "A Waste Assimilative Capacity Model
for a Shallow, Turbulent Stream," Water Research. Vol.9, pp 233-241.
Odum, H.T., W. McConnel, and W. Abbott. 1958. "The Chlorophylla of
Communities," Institute of Marine Science. Vol.V, pp 65-69.
Omernik, J.M. 1977. Nonpoint Source—Streams Nutrient Level Relationships:
A Nationwide Study. Corvallis Environmental Research Laboratory,
EPA-600/3-77-105.
568
-------
Parker, F.L., and P.A. Krenkel. 1969. Engineering Aspects of Thermal
Pollution. Vanderbilt University Press.
Pluhowski, E.J. 1970. Urbanization and Its Effect on Stream Temperature.
U.S. Geological Survey Professional Paper 627-D, pp 1-109.
Schumm, S.A. 1960. The Shape of Alluvial Channels in Relation to Sediment
Type. U.S. Geological Survey, Professional Paper 352-B, Washington,
D.C.
Sladecek, V. 1965. "The Future of the Saprobity System," .Hydrobilogia.
Vol.25, pp 518-537.
Sladecek, V. 1969. "The Measures of Saprobity," Verh. Int. Ver. Limnol.
Vol.17, pp 546-559.
Stall, J.B., N.L. Rupani, and P.K. Kandaswamy. 1958. "Sediment Transport
in Money Creek," American Society of Civil Engineers, Journal of the
Hydraulics Division. Vol.84, No. lm7~pTT53lTrT6~l531-27.
Stefan, H. and J.S. Gulliver. 1978. Effluent Mixing in a Shallow River.
Journal Environmental Engineering Division, ASCE. Vol.104, No.2. pp
199-213.
Stumm, W., and J.J. Morgan. 1970. Aquatic Chemistry. Wiley-Interscience;
New York.
Stumm, W., and Stumm-E. Zollinger. 1972. "The Role of Phosphorus in
Eutrophication," Water Pollution Microbiology. Chapter 3,
Wiley-Interscience; New York.
Task Committee on Preparation of Sedimentation Manual, 1971. "Sediment
Discharge Formulas," Journal of the Hydraulic Division. Vol.97,
No.HY4, Proc. Paper W.7755^
Tetra Tech, Inc. 1978. Methodology for Evaluation of Multiple Power Plant
Cooling System Effects, Volumes I and III. Tetra Tech Report TC-3810.
Thibodeaux. 1979. Chemodynamics: Environmental Movement of Chemicals in
Air, Water, and Soil. John Wiley & Sons, New York, pg 501.
Thibodeaux, L.J. 1980. Spill of Soluble, High Density Immiscible Chemicals
on Water. U.S. Department of Transportation Report No.
CG-UOA-80-011. pg 131.
Thibodeaux, L.G. 1981. Personal Communication on August 18.
Thomann, R.V. 1972. Systems Analysis and Water Quality Management.
Environmental Research and Applications, Inc., New York.
Thomas, N.A., and R.L. O'Connell. 1966. "A method for Measuring Primary
Production by Stream Benthos," Limnology and Oceanography. Vol.11,
No.3, pp 386-392.
Tsivoglou, E.G. and J.R. Wallace. 1972. Characterization of Stream
Reaeration Capacity. EPA-R3-72-012.
569
-------
Turk, J.T. 1980. Applications of Hudson River Basin PCB-Transport Studies.
In: Contaminants and Sediments, Vol. 1. Ann Arbor Science.
U.S. Bureau of Reclamation, Interim Report. 1958. Total Sediment Transport
Program, Lower Colorado River Basin. USBR. Denver, Colorado.
U.S. Department of Commerce. 1968. Climatic Atlas of the United States.
U.S. Dept. of Commerce, Environmental Sciences Services
Administration, Environmental Data Service; Washington, D.C.
U.S. Environmental Protection Agency. 1975. National Water Quality
Inventory. Report to Congress, EPA-440/9-75-014.
U.S. Environmental Protection Agency. 1976. Quality Criteria for Water.
U.S. Environmental Protection Agency. 1976. The Influence of Land Use on
Stream Nutrient Levels. Ecological Research Series, EPA-600/13-76-014.
U.S. Environmental Protection Agency. 1981. The Economic Impact of
Promulgating Toxic Standards for Indiana—A Case Study.
Wild, H.E., C.N. Sawyer, and T.C. McMahon. 1971. "Factors Affecting
Nitrification Kinetics," Journal of the Water Pollution Control
Federation. Vol. 43, pg 1845.
Wright, R.M., and A.J. McDonnell. 1979. In-Stream Deoxygenation Rate
Prediction. Journal Environmental Engineering Division, ASCE.
Vol. 105, No. EE2.
Yang, C.T. 1976. Minimum Unit Stream Power and Fluvial Hydraulics.
Journal of the Hydraulics Division, ASCE, Vol. 102, No. HY7.
Zison, S.W., K.F. Haven, and W.B. Mills. 1977. Water Quality Assessment:
A Screening Method for Nondesignated 208 Areas. Environmental
Protection Agency 600/9-77-023.
Zison, S.W., W.B. Mills, D. Deimer, and C.W. Chen. 1978. Rates, Constants,
and Kinetics Formulations in Surface Water Quality Modeling.
U.S. Environmental Protection Agency 600/3-78-105.
570
------- |