EPA/600/R-04/026
January 2004
A COMPREHENSIVE NONPOINT SOURCE FIELD STUDY FOR SEDIMENT,
NUTRIENTS, AND PATHOGENS IN THE SOUTH FORK BROAD RIVER
WATERSHED IN NORTHEAST GEORGIA
by
Charles N. Smith1, Frank E. Stancil1, David L. Spidle1, Paul D. Smith1, Brenda E. Kitchens1,
Heinz Kollig2, Linda Smith2, Sam Senter2, Mike Cyterski1, Lourdes Prieto1, Dermont Bouchard1,
Kurt Wolfe1, Rajbir Parmar1, Yusuf Mohamoud1, Morris Flexner3, Tom Cavinder3 and Bonita
Johnson3
1 National Exposure Research Laboratory
Ecosystems Research Division
Athens, Georgia 30605-2700
2Senior Service America, Inc.
Senior Environmental Employment (SEE) Program
Silver Spring, MD 20910
3U. S. Environmental Protection Agency
Region 4
Science and Ecosystem Support Division
Athens, Georgia 30605-2700
National Exposure Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
-------
NOTICE
This report was prepared and reviewed by the EPA's National Exposure Research
Laboratory's Ecosystems Research Division in Athens, Georgia, and the Region 4's Science and
Ecosystem Support Division in Athens, Georgia, and approved for publication.
This work reports the results of research only. Mention of trade names, products, or
services does not convey, and should not be interpreted as conveying, official EPA approval,
endorsement or recommendation by the U. S. Environmental Protection Agency.
-------
FOREWORD
As environmental controls become more costly to implement and the penalties of judgement
errors become more severe, environmental quality management requires more efficient management
tools based on greater knowledge of the environmental phenomena to be managed. The National
Exposure Research Laboratory's Ecosystems Research Division (ERD) in Athens, GA, conducts
process, modeling, and field research to assess the exposure risks of humans and ecosystems to
chemical and non-chemical stressors. ERD research includes studies of the behavior of contaminants,
nutrients, and biota in environmental systems, and the development of mathematical models to assess
the response of aquatic systems, watersheds and landscapes to stresses from natural and anthropogenic
sources. ERD field and laboratory studies support process research, model development and field
testing, and characterize variability and prediction uncertainty.
A cooperative field data collection project was developed in the South Fork Broad River
Watershed (SFBR), a 245.18 square mile area with 337.32 stream miles located in the Savannah River
Basin, that consists of intensive storm event stream sampling. In 1998, the State of Georgia listed the
SFBR watershed as biologically impaired (i.e. 303. (d) list), but the source of contamination was
unknown. This project would support: 1) developing sampling protocols to measure the Total
Maximum Daily Load (TMDL) of bedload and suspended sediment, nutrients (e.g. indicators total
nitrogen, nitrate, ammonia, ortho and total phosphorus), total organic carbon and pathogens (e.g.
indicators fecal coliform, E. coli. and enterococci); and 2) developing a comprehensive database to
develop, field test and apply mathematical models and protocols for calculating the TMDLs in this
watershed and its tributaries in a field setting not available elsewhere in the U.S. Six stream sites
were highly instrumented with specialized monitoring equipment (e.g. ISCO water samplers, YSI
multi-probes and cableway sampling systems) for collecting data before, during and after storm
events.
This project was a great example of joint Federal cooperation where special expertise and
technology were being pooled to achieve a common goal. This effort addressed several of the issues
identified in the "Twenty Needs Report: How Research Can Improve the TMDL Program", EPA,
2002 and will establish a scientific database for clean sediment and pollutant TMDLs. This project
will improve Regional involvement in research planning and enhance ORD's detailed knowledge of
the TMDL program. Participants in the study included: The U. S. Environmental Protection Agency,
Office of Research and Development, National Exposure Research Laboratory, Ecosystems Research
Division, Athens, GA; the EPA Region 4 Science and Ecosystem Support Division, Athens, GA; the
EPA Region 4 Water Management Division, Atlanta, GA; and other Federal, State and County
Agencies, and landowners.
The report is intended to provide a description of the field project design, quality control,
sampling and analysis methodology and standard operating procedures. The database design,
architecture, user access, quality control, security and other details will be made available in a
separate report.
Rosemarie C. Russo, Ph.D.
Director
Ecosystems Research Division
Athens, Georgia
iii
-------
ABSTRACT
This technical report provides a description of the field project design, quality control,
the sampling protocols and analysis methodology used, and standard operating procedures for
the South Fork Broad River Watershed (SFBR) Total Maximum Daily Load (TMDL) project.
This watershed is located in the Savannah River Basin and the project constitutes Task 12556,
Field Research Program. A TMDL is the sum of the individual pollutant waste load allocation
for point sources and load allocation for nonpoint sources and natural background, with a margin
of safety (CWA Section 303 (d)(l)(C), EPA 1999).
The field study reported was part of a project designed to develop sampling protocols and
predictive models, and to establish a comprehensive database to field test the developed models
in a field setting not available elsewhere in the U.S. The protocols and models will then be
applied to calculate a series of TMDLs for contaminants of concern (e.g. sediment, nutrients and
pathogens). These protocols can be used by the EPA Regions, Office of Water, and States to
meet the national requirements for TMDL development and implementation under the Clean
Water Act. The field study will establish scientific basis for clean sediment and pollutant
TMDLs.
Our selected field site was the SFBR watershed, a 245.18 square mile area located in the
Savannah River Basin near Athens, GA. Six sites within the SFBR were highly instrumented
with specialized monitoring equipment (e.g. ISCO, YSI multi-probes, cableway sampling
system) to collect data before, during and after storm events on stream depth, turbidity, specific
conductance, pH, dissolved oxygen (DO), Oxidation Reduction Potential (ORP), and
temperature. Rain-event stream sampling was conducted for bedload sediment, total suspended
solids (TSS), nutrients (e.g. nitrate, ammonia, total nitrogen, ortho and total phosphorus), total
organic carbon and pathogens (e.g. indicators fecal coliform, E. coll and enterococci). Stream
hydrographic data were also collected including stage-discharge relationships, water stage
records and velocity profiles. The water stage data was obtained at a continuously recording
USGS station (SFBR70) plus several non-recording stations (i.e. SFBR10, 20, 30, 40 and 60).
The stage data and the attendant stage-discharge relationships provide the basis for assessing
stream discharge. When coupled with analytical measurements, pollutant loadings can then be
calculated for the various sampled runoff events.
Two hundred seventy nine stream cross-sectional sites located along stream channels in
the SFBR were surveyed and sampling/analysis completed for particle size distribution at 90 of
these sites.
Slope-elevation estimates have been developed to generate longitudinal profiles at
selected cross-sectional sites along the SFBR and at sites 1500 feet above and below existing
bridges, using bridge elevation data as benchmarks obtained from the Georgia Department of
Transportation.
IV
-------
A comprehensive relational database system has been designed and populated with data
collected from the South Fork Broad River Watershed sampling sites. The database resides in
MySQL database server. Field data are recorded in various formats including hand written field
and lab sheets, text files, and excel worksheets. A software system has been developed for
transferring data from text files and excel worksheets to the relational database. The database
created in the SFBR will be unique; there is no other study site with a comparable collection of
data in the U.S.
-------
CONTENTS
NOTICE ii
FOREWORD iii
ABSTRACT iv
CONTENTS vi
LIST OF TABLES viii
LIST OF FIGURES ix
ACKNOWLEDGMENTS xi
Chapter 1 1
INTRODUCTION 1
Statement of Problem 1
Background 1
Chapter 2 8
SUMMARY AND CONCLUSIONS 8
Chapter 3 11
DESCRIPTION OF SFBR WATERSHED PROJECT 11
Project Objectives 11
Specific Project Objectives 11
Watershed Characteristics 11
Experimental Design 13
Landowner Approval 13
Field Site Name and Location 18
Stream Monitoring Site Design 22
Installation of Cableway Sampling Systems 23
Site Maintenance 24
Communication at Field Sites 24
Laboratory Modifications 24
Vehicle Requirements 26
Battery Charging 26
Stream Cross Sections 26
Weather Station 28
Storm Event Sampling 28
Pathogen Sampling 30
Monthly Base Flow Sampling 32
Field Sampling procedures 32
Stream Flow 32
Stream Velocity Profiles 33
Laboratory Analysis Procedures 33
Training of Field Sampling and Laboratory Analysis Teams 48
Quality Assurance 48
Database Development 50
vi
-------
Chapter 4 51
Supporting Projects in SFBR 51
Interagency Agreement with the U.S. Geological Survey 51
Task Order through Region 4 contractor, Integrated Laboratory Systems, Inc. . 51
Interagency Agreements with the U.S. Army Corps of Engineers 51
Cooperative Agreement with the University of Georgia 52
In-house research Project by Roger Burke 52
Chapter 5 55
RESULTS OF STUDY 55
REFERENCES 73
Appendices 75
1. Standard Operating Procedures 75
EAB-001.0 Bridge Sampling Using Bedload andDH-59 Samplers 75
EAB-002.0 Use of Laboratory Balances 77
EAB-003.0 Chemical Analysis of Nutrients Using the AutoAnalyzer 3 79
EAB-004.0. Analysis of Total Suspended Solids (TSS) 83
EAB-005.0 Operation of the Cableway System at SFBR30 87
EAB-006.0 Operation of the Cableway System at SFBR60 92
EAB-007.0 Operation of the Cableway System at SFBR70 97
EAB-008.0 Field Sample Numbering System for Storm Event Sampling .... 101
EAB-009.0 Stream Cross Section Sample Numbering System 105
EAB-010.0 Particle Size Analysis Using the Coulter LS200 PSA 107
EAB-011.0 Analysis of Total Organic Carbon (TOC) 112
EAB-041.0 Fecal Coliform Membrane (MF) Technique 114
EAB-042.0 Simultaneous Detection of Total Coliform and Escherichia coli .118
EAB-043-0 Detection and Quantification of Enterococci 122
2. Storm Event Sampling Data Analysis 125
vn
-------
LIST OF TABLES
Table: 1. Loadings for SFBR sites for March 12, 2002 Rain Event 9, 67
Table: 2. Stream Site Identification and Drainage Areas 17
Table: 3. Loading rates for SFBR10 Rain Event on March 12, 2002 68
Table: 4. Loading rates for SFBR20 Rain Event on March 12, 2002 69
Table: 5. Loading rates for SFBR30 Rain Event on March 12, 2002 70
Table: 6. Loading rates for SFBR40 Rain Event on March 12, 2002 71
Table: 7. Loading rates for SFBR60 Rain Event on March 12, 2002 72
Vlll
-------
LIST OF FIGURES
Figure 1. Fecal Coliform Counts in the Savannah River Basin and the South Fork Broad River
Watershed 5
Figure 2.Location of South Fork Broad River Watershed in Clarke, Madison and Oglethorpe
Counties, GA 6
Figure 3. South Fork Broad River Watershed Stream Sample Sites, Rain Gage and Weather
Station Locations 7
Figure 4: South Fork Broad River Watershed with USGS 12 Digit HUC's 14
Figure 5: Land Cover in the South Fork Broad River Watershed 15
Figure 6: Digitized Soils Map of Oglethorpe County Portion of SFBR 16
Figure 7: SFBR10 19
Figure 8: SFBR20 19
Figure 9: SFBR30 20
Figure 10: SFBR40 20
Figure 11: SFBR60 21
Figure 12: SFBR70 21
Figure 13: ISCO Setup at Each Sampling Site 25
Figure 14: YSI Multi-Probe 25
Figure 15. Stream Cross Sectional Sampling Sites, 305 are Scheduled for Investigation 27
Figure 16: Weather Station 29
Figure 17: Rain Gage 29
Figure 18: 6-inch Helley-Smith Bedload Sampler 31
Figure 19: Depth Integrated Sampler 31
Figure 20a. USGS Precipitation Data at SFBR70 site, 02-13-2003 35
Figure 20b. USGS Gage Height Data at SFBR70 site, 02-13-2003 36
Figure 20c. USGS Discharge Data at SFBR70 Site, 02-13-2003 37
Figure 20d. Stream Velocity Profiles for SFBR10 Site, 02-8 & 12-2002 38
Figure 20e. Stream Velocity Profiles for SFBR20 Site, 02-8-2003 40
Figure 20f. Stream Velocity Profiles for SFBR30 Site, 02-7 & 12-2002 42
Figure 20g. Stream Velocity Profiles for SFBR60 Site, 02-7, 8, & 12-2002 44
Figure 20h. Stream Velocity Profiles for SFBR70 Site, 02-7 & 12-2002 46
Figure 21: South Fork Broad River Laboratory Analysis Flow Chart 49
Figure 22. Location of Seventeen Small Watershed Sampling Sites in SFBR Study Area. ... 56
Figure 23. Comparison of Monthly Rainfall in SFBR with 20 Year Average 57
Figure 24. SFBR Average vs 2000 and 2001 57
Figure 25. Comparison of TSS vs Stream Level at SFBR 10 Site for Storm Event on March 12,
2002 58
Figure 26. Comparison of Turbidity by YSI vs Stream Level at SFBR 10 Site for Storm Event on
March 12, 2002 60
Figure 27. Comparison of TSS vs Turbidity at SFBR 10 Site for Storm Event on March 12,
2002 61
IX
-------
Figure 28. Comparison of Total Suspended Solid Measurements of Depth Integrated and ISCO
samplers 62
Figure 29. Estimated linear relationship between Depth Integrated and ISCO Total Suspended
Solid measurements 62
Figure 30. Comparison of one hour versus two hour ISCO Total Suspended Solid
measurements 64
Figure 31. Linear relationship between one hour and two hour ISCO Total Suspended Solid
measurements 64
Figure 32. Relationship between the standard deviation and the mean of Depth Integrated Total
Suspended Solids measurements 65
Figure 33. Relationship between the coefficient of variability and the mean of Depth Integrated
Total Suspended Solids measurements 65
Figure 34. Relationship between Depth Integrated Total Suspended Solid measurements taken at
various locations within a stream cross-section 66
-------
ACKNOWLEDGMENTS
The authors express sincere appreciation for support of the project by EPA at the
National Exposure Research Laboratory, Ecosystems Research Division (ERD), Athens, GA; at
the EPA Region 4 Science and Ecosystem Support Division (SESD), Athens, GA; and at the
EPA Region 4 Water Management Division (WMD), Atlanta, GA. Specifically, we would like
to recognize Dr. Rosemarie C. Russo, Director, Dr. Harvey Holm, Dr. Dave Brown and Dr. Mac
Long of Athens ERD; Mr. Antonio Quinones, Mr. Bill Bokey and Mr. Philip Murphy of Region
4 SESD; Dr. Bruce Pruitt, formerly with SESD now with Nutter and Associates, Athens, GA;
and Mr. Jim Greenfield and Ms. Gail Mitchell of Region 4 WMD.
Special thanks are extended to the following people for their conscientious help: Mr.
Robert Carousel for his guidance and technical leadership as the original project leader; Dr. Earl
Hayter for his sediment modeling data input and selection of stream cross section sites; Ms. Pam
Gunter for her efforts in purchasing all of the equipment and supplies in a timely manner; Mr.
Alan Tasker, Mr. Jimmy Pierce and Mr. Ricky Hardigree for their untiring efforts in conducting
many laboratory modifications and upgrades and construction work on two sheds to house field
equipment and personal rain gear; Mr. Jimmy Pierce again, who designed and developed an
electrical system for direct and generator backup power and phone systems at the remote
cableway stream sampling sites needed for the operation of the cableways, and for providing
excellent lighting for night time sampling.
The authors gratefully acknowledge and appreciate the cooperation of several landowners
for allowing stream monitoring equipment, weather station and rain gages to be installed on their
property, specifically: Mr. Lee Moon, Park Manager, Watson Mill Bridge State Park, GA
Department of Natural Resources; Mr. Larry Edge, Georgia Department of Transportation; Mr.
Gerald Kemp, Superintendent of Public Works, City of Comer; Mr. Wesley Chandler; Mr.
Albert Stovall; Mr. Jack Hammond; Mr. David Spidle; Mr. Jim Wilcox; Dr. Lee Wolfe; Mr.
Randy Patman; and Mr. Bob Ambrose. Without their cooperation, this study would not have
been possible.
More special thanks are extended to the volunteer ERD sampling team members who
participated in storm event sampling: Mr. Robert Carousel; Dr. Earl Hayter; Dr. John M.
Johnston; Mr. Robert Ambrose; Dr. Roger Burke; Ms. Linda Exum; Mr. Gerry Laniak; Mr.
Chris Mazur; Dr. Marirosa Molina; Dr. Brenda Rashleigh; and Dr. Melike Gurel.
We greatly appreciate the work of the staff of the Integrated Laboratory Systems, Inc., in
particular Mr. Bill Simpson, who conducted stream cross-sectional surveys for about 300 sites,
and developed longitudinal slope elevations at the stream cross sections using existing bridge
benchmarks.
Additional special thanks are extended to the following special people: Mr. Tim
XI
-------
Fagerburg, with the U.S. Army Corps of Engineers for his assistance in establishing the stream
monitoring sites with ISCO support tables and stilling wells to house the YSI multi-probes, as
well as for his technical/mechanical skills and leadership during the installation of three
cableway sampling systems; Mr. Brian E. McCallum, Mr. Jack Neighbor and Mr. Cal Day with
the U.S. Geological Survey, for their assistance in the operation and maintenance of a real-time
gaging station at the SFBR70 site, and for developing stream discharge rating curves for each of
the six sites; Ms Rose Kress and Linda Peyman-Dove, with the U.S. Army Corps of Engineers,
for their assistance in developing a SFBR data repository; and Mr. Martin Stancil, who
participated in a summer hire program at Athens ERD during 2000, assisting in the installation
of specialized monitoring equipment at several sampling sites and the construction of wooden
support structures for raingages, extension of cableway platforms and shelving for storage
buildings at the Field Research Annex (FRA).
We thank three technical reviewers for their helpful suggestions to improve the report:
1. Dr. Bruce Pruitt, formerly with EPA Region 4 SESD, now with Nutter and Associates,
Athens, GA.
2. Dr. Robert Swank, formerly with EPA Athens ERD, now Senior Environmental Employee
(SEE)
3. Mr. Bill Bokey, EPA Region 4 SESD, Chief, Ecological Support Branch, Athens, GA
xn
-------
Chapter 1
INTRODUCTION
Statement of Problem
There is an urgent need for EPA to develop protocols for establishing Total Maximum
Daily Load (TMDL) in streams, lakes and estuaries. Prior to the SFBR project, there was limited
scientific data available to support TMDL development. The results of the SFBR project will
accomplish the following:
o Provide a comprehensive dataset that allows for the development, field testing, and
calibration of mathematical models addressing water quality and quantity in a watershed.
The dataset created in SFBR will be unique; there is no other study site with a comparable
collection of data in the U.S.
o Provide robust data and models that establish a scientific basis for clean sediment and
pollutant TMDLs
o Provide a means of testing field and laboratory instrumentation, methodolgy, and
development of standard operating procedures for sampling protocols, sample processing and
analytical analyses
o Develop procedures for site selection, field instrumentation, maintenance and servicing,
frequency of sampling, data requirements, safety and QA
Background
A TMDL is an estimate of the maximum pollutant loading from both point and nonpoint
sources that receiving waters can accept without exceeding water quality standards. Point source
loadings are essentially continuous in time, while most nonpoint source (NPS) loadings occur
intermittently. Under §303(d) of the Clean Water Act, each State must: 1) produce and provide
EPA with a list of waters where water quality standards are not being attained, 2) prioritize the
development of TMDLs for these water bodies that will result in attainment of standards, and 3)
develop and implement the TMDLs. In the event a State fails to develop the list or to develop
TMDLs, EPA is obligated to do so. Although the TMDL requirement has been in existence for
twenty years, most implementation to date has focused on point source requirements rather than
nonpoint loadings. Environmental groups have become impatient with the TMDL process. In
the lawsuit, Sierra Club et al. vs. U.S. EPA, Browner, and Hankinson, the court found that EPA's
failure to disapprove Georgia's inadequate TMDL submissions was in violation of the
Administrative Procedure Act and the Clean Water Act. Sierra Club et al 1996. On September 2,
1996, the district court ordered EPA to ensure that TMDLs were established for all 303(d) listed
waters within five years and to ensure that they are implemented through the National Pollutant
Discharge Elimination System permitting program of the states. In July, 1997, the parties signed
a consent decree that would supersede the court's order of September 2, 1996. The decree sets
out a schedule for establishing TMDLs in each of Georgia's watershed basins between 1998 and
2005.
-------
The decree provides that EPA would ensure that the TMDLs are established if Georgia
does not. If Georgia fails to propose the TMDLs for the Savannah/Ogeechee Basins by 6/30/04,
then EPA shall propose them by 8/30/04 including the Broad River. TMDLs have already been
estimated by EPA Region 4 for the South Fork Broad River, and will be reevaluated based on
results of this research study.
Protocols are urgently needed by the States and EPA to develop over 10,000 watershed
sediment TMDLs that must be established over the next 8-15 years to attain the goals of the
Clean Water Act and to meet various court ordered deadlines. EPA Region 4 alone has over
1100 sediment-related TMDLs to develop within this time frame. Recent trends in current and
evolving environmental regulatory strategies dictate that EPA will have to rely more heavily on
predictive modeling technologies in carrying out the increasingly complex array of exposure and
risk assessments necessary to develop scientifically defensible TMDL's. There is a pressing
need for a comprehensive data set to be developed that will enable such models to be developed
and tested with actual field data. The main goal of sediment TMDL analysis is to protect
designated or existing uses of natural resource systems in watersheds by:
a. Characterizing properly functioning watershed processes that influence the erosion, transport
and storage of sediment;
b. Evaluating the degree to which the current and expected future functioning of these
processes is impaired. The impairment usually results from the transport of excessive
sediment loads to water bodies (e.g., streams, lakes) within the watershed. The excess
sediment loads are usually generated by changes in watershed processes that result from both
natural (e.g., wildfires) and anthropogenic (e.g., logging, agriculture) causes; and
c. Identifying land and water management restoration actions that should be implemented to
restore the proper functionality of the impacted watershed processes.
An environmental group found the Savannah River Basin to be the seventh most toxic
body of water in the nation during a 5-year period (i.e. 1992 through 1996). According to
industrial discharge reports to the EPA, 17.4 million pounds of chemicals were discharged into
the river during this period (Athens Daily News, September 11, 1998).
From Raschke et al. 1998, twenty streams in twelve, eleven-digit-size watersheds in the
Savannah River basin were identified as waters potentially impacted by agricultural nonpoint
source pollution. The listing of these waters as potentially impacted by nonpoint source
pollution was based on a statewide assessment of watersheds conducted in 1991 by the Soil
Conservation Service and Forest Services of the U.S. Department of Agriculture, and the
Georgia Soil and Water Conservation Commission. This assessment was based on a number of
factors including animal population density, topography, fertilizer and pesticide application
rates, and rainfall. To determine the actual condition of the watersheds, the EPA conducted a
screening of these same watersheds during 1997. This screening assessed the status of the
aquatic biological community, habitat conditions, and included water chemistry measurements.
-------
Based on the results of the EPA assessment, eleven streams in eight of the eleven-digit-sized
watersheds in the Savannah River Basin were identified for further study including: 1) South
Fork Broad River; 2) South River; 3) North, 4) Middle, and 5) Lower North forks of the Broad
River; 6) Broad River; 7) Lake Hartwell tributaries (i.e. Crawford Creek, Little Crawford Creek,
Little Shoal Creek, Flat Shoals Creek); and 8) Toccoa Creek.
During 1994 through 1997, EPA Region 4's Science and Ecosystem Support Division
conducted a field screening study of wadeable streams in the Savannah River Basin to evaluate
ecological conditions as part of the Regional Environmental Monitoring and Assessment
Program (REMAP), Raschke et al. 1999. Four ecological response indicators were monitored
including: 1) fish community (Index of Biological Integrity-IB I); 2) macro invertebrates (Rapid
Bioassessment Protocol-RBP method); 3) habitat (RBP method); and 4) algal growth (Algal
Growth Potential Test, AGPT). These indicators represent societal values of biological integrity
and trophic condition for small wadeable streams (i.e. 1st to 3rd order) and the tributary
embayments of large reservoirs. Sixty stream sites per year were targeted for sampling using the
EMAP approach, but only a total of 119 sites were actually sampled due to both property access
issues and the fact that some stream sites failed to meet design site criteria. Results of the
REMAP study indicated that with respect to habitat, community integrity, and trophic condition,
63% to 95% of the basin's wadeable streams were in fair to poor condition. This meant that
about 52% of the stream miles involved in the study were in poor condition based on impacts on
the fish community. Two study areas had an unusually high concentration of poor ecological
condition, one in Georgia (i.e. SFBR area) and one in South Carolina. The impacted Georgia
area had a high population of poultry and cattle production, but the South Carolina impact area
had no obvious landscape features to explain the findings. The SFBR watershed, located in
Georgia's Madison, Oglethorpe and Clarke Counties, was identified as a sediment impacted
stream and found to be in poor ecological condition.
Based on the EPA Region 4 findings and the close proximity of the basin, the Athens
ERD selected the Savannah River Basin as it's Near Laboratory Ecological Research Area
(NLERA) for field research. Athens ERD conducted a stream water quality study of 40 stream
sites in the SRB from June, 1998, to November, 1999, Smith et al. 2002 draft. Results of the
Savannah River Basin study indicated that the SFBR watershed had the highest number of fecal
coliform counts (i.e. as a pathogen indicator) in the basin, exceeding EPA's recreational water
limit of 200 counts per 100 ml, Figure 1.
In 1998, the State of Georgia listed (i. e. the 303 (d) list of the Clean Water Act) the
SFBR watershed as being biologically impaired but the source of the contamination was not
identified.
Research is now underway in this watershed because it is ideal for simultaneously
measuring contaminants and their impacts on ecological health. Athens ERD normally conducts
field studies for at least three years so that results account for seasonal variability in weather
-------
patterns and other important factors. Site location (i.e. Madison, Oglethorpe and Clarke counties)
characteristics include topography, meteorology, hydrology, point sources of pollution,
agricultural production, cropping patterns, agricultural chemical application rates and stream
cross-sectional profiles at segmented river reaches, plus other multifaceted data and information
in order to relate the causes and effects of the changes in contaminant transport measured.
Data obtained through Athens ERD's field research will be widely used for developing
and field-testing models, and providing improved field assessment methodologies and
monitoring protocols. Field research will also provide a useful data resource to both the
Agency's research and regulatory components, local governmental agencies and the general
public for planning and other uses by providing an improved and more realistic understanding of
the impacts of contaminants released into the environment.
In response to the Agency's need, researchers at the EPA's National Exposure Research
Laboratory operation in Athens, GA, (i.e. the Ecosystems Research Division (ERD)), together
with the Region 4 Science and Ecosystem Support and Water Management Divisions have
designed and implemented a sampling plan for a comprehensive watershed study in the SFBR
watershed within the Savannah River Basin, Figures 2, 3.
This interim report provides a description of the field project sampling and analysis
methodology. The database design, architecture, user access, quality control, security and other
details will be made available in a separate report.
-------
Savannah River Basin
Fecal Coliform Counts
w
N
• Fecal Coliforms > 199 Colonies/100 ml
o Fecal Coliforms < 200 colonies/100ml
^| SF Broad River Watershed
^] North Carolina Watersheds
| South Carolina Watersheds
f Georgia Watersheds
50 Miles
Figure 1. Fecal Coliform Counts in the Savannah River Basin and the South Fork Broad River
Watershed
-------
SAVANNAH RIVER BASIN
South Fork Broad River Watershed
JACKSON
WAT KINS VI LL
10 Miles
Cities
EPA Facilities
/ Streams
OGLETHORPE '/\f SF Broad River
/\/Broad River
"LEXINGTON y'A\/RoadS
J|~n SFBR Watershed
Counties
Savannah River Basin
Figure 2. Location of South Fork Broad River Watershed in Clarke, Madison and Oglethorpe
Counties, GA.
-------
South Fork Broad River Watershed
EPA FRP TMDL Sample Sites
BOWMAN
CENTER
COMER
Weather
CARL TON
WINTERVILLE
• ATHENS
EPA Field Research Annex
•
*
EPA ORD/Region 4
WATKINSVILLE
CRAWFORD
10 Miles
LEXINGTON
Rain Gauges
• Cities
• Sample Sites
• EPA Facilities
Streams
y\/SF Broad River
Broad River
Roads
SFBR Watershed
Counties
Figure 3. South Fork Broad River Watershed Stream Sample Sites, Rain Gage and Weather
Station Locations
-------
Chapter 2
SUMMARY AND CONCLUSIONS
We have successfully designed and are implementing the SFBR watershed study to
provide the best available database for developing TMDL sampling protocols, for the field
testing of exposure models, and for the development of TMDLs now and in the future. The
project had six stream monitoring sites highly instrumented and operational, with extensive
background data already collected. In addition, a weather station was installed in the watershed
and additional tipping-bucket rain gages located throughout the watershed. Many of the designs
used in this study will become part of standard operating procedures (SOP) for data collection in
support of TMDL development. Installation of cableway sampling systems was critical to this
effort in order to provide a safe environment for sampling teams to collect the required bedload
and depth-integrated samples during intensive storm events. To our knowledge, these are the
only automated, unmanned stream sampling systems of their size in the United States.
This study has been a most challenging field project to implement due to its complexity
and unexpected delays. The effort involved: getting both the field equipment and laboratory
operational (e.g. new equipment, lab renovation); getting field staff trained (e.g. sampling and
safety); selecting and purchasing the appropriate state-of-the-art equipment for unattended
collection of stream samples; the installation of bank-operated cableway systems; obtaining
permission from land owners to locate equipment on their property; and obtaining through the
General Services Administration (GSA) and purchasing the required vehicles (i.e. 4x4 trucks,
crane trucks, 4x4 mules, trailers) for transporting staff and equipment to field sites.
Rain events tracked thus far have not produced significant runoff and sediment loads due
to the project area being in its fourth year of a severe drought with stream levels at their lowest
in many years. The study area was greater than 30 inches below the 20-year rainfall average of
51.5 inches/year during the period 2000 to 2001. Historically, 100 years ago in March 1902,
such great rainfall fell in the project area as reported in a local newspaper (March 14, 2002 issue
of The Comer News, Comer, GA 30629), that the Broad River stage rose to levels of 25 to 30
feet above base flow.
Nine (9) storm events have been sampled since January 2001; however, only limited data
were collected because of the brevity of the storms. To obtain the needed manpower to
manually collect samples during storm events at 2-hour sampling intervals, an additional work
request was added to an existing task order through a Region 4 contractor, Integrated Laboratory
Systems, Inc. These task orders would provide the staff needed to collect storm event samples
for bedload and depth-integrated samples at four high priority sites (i.e. SFBR10, 30, 60 and 70).
ISCO sampling was done at 1-hour intervals for all six sites.
More rainfall (i.e. number of events and the amount) occurred during 2003 than any other
year since the beginning of the field research project. A large rain event was predicted to occur
during early April 2003 that produced a week of field sampling with 4-days of intensive
sampling. This rain event was not as intensive as expected for producing heavy runoff but it did
-------
provide a low intensity rain over several days. Unfortunately, during this sampling event, two
sites equipped with the Rickly Hydrological Company type cableway systems malfunctioned and
became inoperable and unsafe and therefore eliminated further rain event sampling until
replacement winches could be installed. After many discussions with the Rickly Company, they
finally agreed to replace both winch systems, but not until year 2004. The cableway system
problems have caused unexpected delays in obtaining the required data during a period of
greater rainfall.
TSS values obtained from an ISCO sampler located at a fixed depth and cross-sectional
location in the stream channel were statistically different (i.e. at the 95% confidence level) from
depth-integrated samples taken across the stream channel at the same site. However, these
values were highly correlated. It is valuable to have a statistical relationship between the
automated ISCO sampler produces TSS measures and the more time and labor intensive depth-
integrated sampling procedure. However, additional data collection efforts during a variety of
different size storm events would be needed to confirm this relationship before a definitive
conclusion could be drawn. Data analysis also showed that taking ISCO samples every two
hours versus every hour would result in approximately a 26% drop in information content. The
methodology behind these conclusions is explained in Chapter 5 and Appendix 2.
An example of stream loading rates for the SFBR based on samples collected from the
ISCO samplers were calculated for TSS, nutrients and carbon for each of the 5 sites during a
storm event that occurred on March 12, 2002. This storm represents a low-runoff event with
rainfall amounts of 0.93 inches. Table 1 provides a summary of the calculated storm event
loadings (kg) for each of the five sites for each measured parameter. Estimated TSS loadings at
sites SFBR 30 and 60 were the highest at 23,642 and 34,248 kg, respectively. The highest
loadings for ammonia, nitrate, phosphorus and total organic carbon were also calculated at these
two sites. As expected, lower TSS loadings were observed at site SFBR20 which is a small
tributary of the SFBR. One would expect even higher loads to occur during greater rainfall
events.
Table: 1. Loadings for SFBR sites for March 12, 2002 Rain Event
Site
SFBR10
SFBR20
SFBR30
SFBR40
SFBR60
TSS
kg
2179.7
195.9
23641.6
2244.8
34248.2
NHg-N
kg
2.0
0.7
14.0
5.5
8.0
NO' 3-N
kg
19.6
1.9
175.4
5.5
118.4
o-PO4 as P
kg
1.6
0.3
8.4
3.0
6.5
Total-P
kg
2.4
0.4
9.1
4.3
18.1
TOC
kg
168.0
19.8
1592.7
484.2
1121.8
-------
The SFBR watershed is an excellent area to continue watershed research given the
diversity of land usage, landowner cooperation, background stream water quality and
hydrographic data, and longitudinal stream slope elevations and characterization of about 300
stream cross sections. The SFBR covers an area of 245.18 square miles with 337.32 stream
miles. It would take many man-years to establish/characterize a comparable test watershed at
another location.
10
-------
Chapter 3
DESCRIPTION OF SFBR WATERSHED PROJECT
Project Objectives
EPA's Office of Research and Development, National Exposure Research Laboratory,
Ecosystems Research Division in Athens, GA (Athens-ERD), together with the Agency's Region
4's Science and Ecosystem Support and Water Management Divisions, have designed and are
implementing a sampling plan for a field study in the SFBR watershed to calculate/establish
TMDLs. The study involves the collection of storm event samples at six stream sites for the
analysis of bedload and suspended sediment, nutrients (i.e. nitrate, ammonia, total nitrogen,
ortho and total phosphorus), organic carbon and pathogens (i.e. indicators fecal coliform, E. coll
and enterococci). The objective of the SFBR study is to develop sampling protocols and
predictive models, and to establish a comprehensive database to field test the developed models.
These protocols and models will then be applied for calculating the TMDLs.
Specific Project Objectives
I. Develop a comprehensive database for SFBR including: 1) instream storm event data for
measured bedload and total suspended solids, nutrients, organic carbon, and pathogens;
2) meteorology; 3) stream hydrography (i.e. stage-discharge relationships, water stage
records and velocity profiles); 4) stream cross-section characterization; 5) longitudinal
stream slope elevations; 6) soil characterization; 7) land cover; and 8) contaminant
loading rates.
II. In conjunction with EPA Region 4, develop and implement an intensive rainfall/runoff
event driven field sampling program, including development of the sampling protocols
and the database needed to calibrate and conduct field performance testing of predictive
exposure models.
Watershed Characteristics
The SFBR watershed, a 245.18 square mile area (156,915 acres), (Figure 3), is a tributary
of the Broad River Watershed located in the Savannah River Basin. The Savannah River Basin is
a 10,577 square mile area located along the border of three states with 5,821 square miles in
Georgia, 4,581 square miles in South Carolina and 175 square miles in North Carolina. The
Savannah River originates in the mountains of Georgia, South Carolina and North Carolina and
flows south-southeasterly about 300 miles to the Atlantic Ocean near the port city of Savannah,
GA. Approximately 45 miles of the lower Savannah River are influenced by tidal action. The
Broad River Watershed is located in the Piedmont area of North Georgia, is the largest
watershed within the Savannah River Basin, and is among the last free-flowing rivers
(undammed) in Georgia. It flows from the foothills to the fall line in a southeastly direction
from its headwaters in Stephens and Banks Counties through Franklin, Madison and Elbert
Counties to its confluence with the Savannah River at the Clark's Hill Reservoir. Poultry, cattle
and forestry production, with some row crops, are the major agricultural activities in the
watershed. Runoff is the major mode of transport for pollutants entering surface water streams.
11
-------
Chicken manure is widely used as a fertilizer source on largely sloped pasture land. The Broad
River's runoff problems are attributed to effluent from septic systems, landfill leachate, litter,
riverbank erosion, destruction of the vegetative buffer, lack of tributary protection and, most
importantly, non-point source agricultural runoff. Animals (e.g. cattle, deer) obtaining their
drinking water from the river contribute to water pollution, sedimentation, and degradation and
destabilization of riverbanks.
The SFBR watershed (Figure 3) is located in the southern Piedmont area near Athens,
GA, primarily in Madison and Oglethorpe counties with a small portion in Clarke county. The
watershed contains 337.32 stream miles (Figure 4). The watershed had severe erosion problems
during the 1900 to 1960's due to the conventional tillage practices (i.e. moldboard plowing)
without adequate conservation measures (e.g. terraces, grass waterways). This occured
principally in the upland areas of the watershed that were primarily used in the production of
cotton (Broad River Soil Conservation District and County Government of Madison County,
1961). In addition, when the use of tractors began in the area (i.e. replacing horses and mules)
this added to the erosion problem since farmers would plow the land parallel to slope (personal
communication with local farmer). During this time the boll weevil was a serious pest in the
production of cotton and several pesticides were used including arsenic, DDT and toxaphene that
caused fish kills to occur in nearby streams. Today, most of the land cover in the watershed is in
deciduous, evergreen and mixed forests interspersed with pasture, hay and row crops (Figure 5).
Many of the agricultural activities today integrate chickens (i.e. broilers, layers), cattle (i.e. beef
and dairy), hogs, row crops (i.e. cotton, corn, soybeans, wheat, rye), hay/silage and forest
production (i.e. pine and hardwood). Small towns (population numbers obtained from the
Georgia County Guide, 2002) located in the watershed include Ila (328), Danielsville (457),
Colbert (488), Comer (1052), Hull (160) and Carlton (233). The populations of Madison,
Oglethorpe and Clarke Counties are 25,730, 12,635 and 101,489 respectively based on year 2000
census. Madison County has the 5th largest number of poultry houses (616) and the 5th largest
number of cow head (15,800) in the State (Georgia County Guide, 2002). Oglethorpe County
has 235 poultry houses and 7,500 cow head and Clarke County has 11 poultry houses and 2000
cow head. Runoff delivery of large amounts of sediment, nutrients, pesticides and pathogens are
the primary causes of receiving water impairment.
The SFBR watershed consists of well-drained, upland soils with a loamy surface layer
and a red clay subsoil (USDA, 1968, 1979, 1991). These soils have a slow to moderate
infiltration rate and a good water holding capacity. The soil loss due to erosion during the 1900
to 1960 era was estimated to be 65 percent of the topsoil (Broad River Soil Conservation District
and County Government of Madison County, 1961). A digitized soils map for that portion of the
watershed in Oglethorpe county is shown in Figure 6. Digitized soils maps for Madison and
Clarke counties are not currently available, but are expected to be published within 2 years
(personal communication, USD A).
12
-------
Experimental Design
Statistical design was not used to determine site locations. The following factors were
considered in selecting the stream sampling sites:
• site location to measure contaminant loads by runoff at specific upstream and
downstream points within the watershed, including the headwaters and confluence or
watershed outlet
• stream mixing
• existing USGS stream gaging stations
• vehicle accessibility to site
availability of an existing highway bridge (in use or abandoned)
landowner approval
Once a decision was made relative to where to locate a stream monitoring site, contacts
were made with the local landowner for approval to allow EPA to install long term monitoring
equipment on their property. In some cases, this required going to the Madison County
Courthouse to determine the landowner, and then contacting them for approval. In several cases,
the landowner lived outside the watershed. Approval for three (i.e. SFBR30, 60, 70) of the more
difficult and larger stream sites required approval by the Georgia Department of Transportation
(GA-DOT), Georgia Department of Natural Resources and a landowner located in middle
Georgia. Each of these sites required the purchase and construction of bank-operated cableway
systems requiring additional time and resources.
Landowner Approval
A number of unexpected delays and problems were encountered in selecting and
establishing the stream monitoring sites, including difficulties in obtaining landowners approval
to install equipment on their property.
Site SFBR30 - bridge at highway 172. Permission to sample from existing bridges in the SFBR
watershed was denied by the Georgia Department of Transportation because blocking one lane
of the bridges and roads during day and night time rain storm sampling for several days would
create a safety hazard. Their concern was due to the liability issue in case an accident occurred
causing injuries while following GA-DOT road blocking safety protocols. The GA-DOT
required protocol is for their staff to get off the road during these critical times. GA-DOT did,
however, provide EPA a permit to install the cableway system on their right-of-way.
Site SFBR60 - Clouds Creek in Watson Mill Bridge State Park - we originally evaluated the
possibility of restoring the flooring of an old, abandoned steel bridge for use in sampling.
Estimated cost by the Georgia Department of Natural Resources to restore the metal
frame, bridge flooring, and guard rails was about $250K. The TMDL project staff reevaluated
this site and determined that the bridge would also cause restrictive flow and that it would be
better to move upstream and construct a cableway system. To obtain approval at this site for a
13
-------
South Fork Broad River Watershed
12 - Digit HUC's
ELBERT
JACKSON
EPA Field Research A
•
•
EPA ORD'Region 4
OGLETHORPE
10 Miles
12 -Digit HUG
030601040401
030601040402
030601040403
030601040404
030601040405
030601040406
Sq. km
129.07
94.33
96.12
123.19
46.81
145.51
Sq. miles
49.84
36.42
37.11
47.56
18.07
56.18
Stream miles
69.39
51.49
49.04
61.88
22.59
82.93
EPA Facilities
HUC's 12 Digit
/ Streams
/X/'SF Broad River
/\/ Broad River
| | SFBR Watershed
Counties
Figure 4. South Fork Broad River Watershed with USGS 12 Digit HUC' s.
14
-------
South Fork Broad River Watershed
1996 Landsat TM Land Cover Image
N
•• *sl >.'..• -•1. -'**-«ffS«A«i(v
Class
Deciduous Forest
Evergreen Forest
MiKed Forest
Pasture/ Hay
Row Crops
Transitional
Woody Wetlands
Open Water
Low Intensity Residential
Quarries/ St rip Mines/Gravel Pits
Commercial/Industrial/Transpor
Urban / Recreational Grasses
Bare Rock / Sand/Clay
Emergent Herbaceous Wetlands
High Intensity Residential
Total
Sq . km
IBS. 10
156. 32
103. 05
101. 68
61.04
3.80
3. 31
2. 51
2. 68
1. 80
2. 66
0. 83
0.38
0. 27
0. 15
635. 03
Sq . miles
73. 01
60.36
39.79
39.26
23.57
3. 40
1.28
0. 97
1. 03
0. 69
1. 03
0. 32
0. 15
0 . 10
0. 06
245. 18
% Cover
29.78
24. 62
16.23
16. 01
9. 61
1. 39
0. 52
0. 40
0. 42
0. 28
0. 42
0. 13
0. 06
[i . 04
0. 02
100. 00
012345 Miles
Land Cover SFBR 1996
| | Deciduous Forest
^B Evergreen Forest
| Mixed Forest
| Pasture/Hay
| Row Crops
| Transitional
| Woody Wetlands
^^ Open Water
^^ Low Intensity Residential
| Quarries/Strip Mines/Gravel Pits
| Commercial/lndustrial/Transportation
| Urban/Recreational Grasses
| | Bare Rock/Sand/Clay
| Emergent Herbaceous Wetlands
^B High Intensity Residential
Figure 5. Land Cover in the South Fork Broad River Watershed
15
-------
Oglethorpe County Soils Map
SFBR Watershed Portion
Oglethorpe CountySoils
I | AltaVista Loam
| | Appling coarse sandy loam
j ] Ashlar, Louisburgand Pacolet soils
gCartecay loam
Cecil sandy clay loam
Cecil sandy loam
Chewacla silt loam
I I Ruuaquents. ponded
| | Helena sandyloam
| | Hd••.',•• ass-i* clay loam
SHawassee loam
Madison sandy clay loam
Madison sandy loam
I | Mecklenburgfine sandy loam
| | Pacolet sandy clay loam
| ] Pacolet sandy loam
I | Pacolet-gull led land complex
S Quarries
RSuerview sift loam
| | Toccoafine sandy loam
I 1 Water
| | Wedowee sandy loam
| | Vvlckham fine sandy loam
SFBR Watershed
Figure 6. Digitized Soils Map of Oglethorpe County Portion of SFBR
16
-------
bank-operated cableway system, we had to develop an access agreement that took about 9
months to complete and get approved. Mr. John Cline of EPA- NERL-RTP provided excellent
help in obtaining the access agreement by working with EPA's Office of General Council and the
Georgia Department of Natural Resources. The agreement, effective February 7, 2000, states
that EPA would have access to a 1-acre site for 10 years. Installation of a cableway system at
this site also required approval from the private landowner of the opposite bank. This approval
necessitated extensive inquires and phone calls to locate the owner.
Site SFBR70 - Carlton site on the Carlton-Lexington road. Initially, a decision was made to try
two approaches; 1) find the local landowner as well as 2) contacting the Georgia Department of
Transportation for a permit to install a cableway on their right-of-way as was previously done at
the SFBR30 site. The Georgia Department of Transportation notified us that this was a county
road, not state-owned. Upon identification of the landowner, he was contacted by phone to
discuss the possibility of locating our cableway system on his property. He informed us that he
owned the land on both sides of the river, that he would be willing to cooperate with EPA, and
that there was already an agreement (dated February 10, 1998) with EPA Region 4 that could be
revised and used. With the help of Walter Stutts, NERL-Cincinnati, a 10-year access agreement
effective October 10, 2001 was obtained.
Six stream sites were ultimately selected in the SFBR watershed for collecting data
before, during and after storm events. Originally four sampling sites were selected along the
main stem SFBR (SFBR10, 30, 50, and 70) and three more along the major tributaries of the
SFBR (SFBR20, 40 and 60, see Table 2). Site SFBR50 was finally canceled due to resource
limitations in that this site required the installation of another cableway sampling system. Along
the main stem of the SFBR, there are three existing dams constructed many years ago: 1) Rogers
Mill dam below the SFBR10 site; 2) Bullock Mill; and 3) Watson Mill. Numerous farm ponds
are located in the watershed, and several beaver dams have been located in the streams.
Table 2. Stream Site Identification and Drainage Areas
Site Number Site Name Drainage area (mi2)
1 SFBR10 (Ila) 16.9
2 SFBR20 (Double Branch) 4.8
3 SFBR30 (Highway 172) 85.9
4 SFBR40 (Brush Creek) 34.1
5 SFBR50 (Beaverdam Creek) site canceled
6 SFBR60 (Clouds Creek) 47.5
7 SFBR70 (Carlton) 224.0
Five additional sampling sites were selected by Region 4's Science and Ecosystems
Support Division (SESD); stations Bl, B1A, B2, B3 and B4 (Figure 3.) located on Biger Creek,
a tributary of Brush Creek in the SFBR watershed. A reference watershed site was located in
Hart County, GA, at Lightwood Log Creek. This site had already been sampled by Region 4's
SESD. Field site information and stream sampling regarding these latter six sites are not
provided in this report.
17
-------
Field Site Name and Location
The following is a brief description of the stream sampling site names and their locations
in the SFBR watershed.
SFBR10 - This is an upstream headwater site located south of Ila, GA, off highway 106 on Old
Ila Road. There is an old steel bridge at this site, located just downstream of the currently used
concrete bridge, from which sampling is conducted using a truck outfitted with a crane system
(Figure 7).
SFBR20 - located south of Danielsville, GA, along highway 29 about 150 feet downstream of
the highway culvert. Two branches merge upstream and above highway 29 and form Double
Branch Creek. This is a wadeable stream site for sampling (Figure 8).
SFBR30 - located north of Colbert, GA, near a bridge on Highway 172 immediately downstream
from the confluence of Brush Creek with the SFBR. Sampling is conducted with a cableway
system (Figure 9).
SFBR40 - Brush Creek sampling site is located at a wooden bridge on the McCarty-Dodd Road
between highway 172 and Bullock Mill road. The Georgia Department of Transportation plans
to replace the wood bridge with a concrete structure in the near future. The Brush Creek site is
unique in that runoff from an unpaved road conveys a large amount of sediment into the stream
during rain events (Figure 10).
SFBR50 - located south of Comer, GA, near a bridge on Highway 22 immediately downstream
from the confluence of Beaverdam Creek and the SFBR. This site was canceled due to resource
limitations because it required the installation of another cableway sampling system.
SFBR60 - this site is located on Big Clouds Creek approximately 800 feet upstream from its
confluence with the SFBR in the Watson Mill Bridge State Park primitive camping area.
Sampling is conducted with a cableway system (Figure 11).
SFBR70 - located next to the Carlton-Lexington road bridge (near Sandy Cross) over the SFBR
near Carlton, GA. This site is near the outlet of the SFBR into the Broad River, which is located
about 2 miles downstream. This site is the nearest access point to the Broad River for
establishing a monitoring site. Sampling is conducted by cableway system (Figure 12).
18
-------
Figure 7: SFBR10
Figure 8: SFBR20
19
-------
Figure 9: SFBR30
Figure 10 SFBR40
20
-------
Figure 11: SFBR60
Figure 12: SFBR70
21
-------
Stream Monitoring Site Design
Each of the six stream sites was instrumented with two automatic ISCO samplers (Figure
13) and a YSI multi-probe (Figure 14) located at a fixed point (i.e. position and depth) in the
stream. The two ISCO samplers are used to: 1) provide a duplicate set of samples for analysis in
the laboratory for sediment and one set of samples for nutrient analysis; and 2) to provide
backup samples in case one of the ISCO samplers fails to collect or overfills. A software
interface problem surfaced when the ISCO sampler was coupled to the YSI probe. This required
lengthy manufacturer efforts to resolve. A wooden support platform was constructed for the two
ISCO's using treated 2-inch x 4-inch wood frame with plywood deck (1 foot high by 2-feet wide
by 6 feet long) secured to ground by concrete. For security purposes, galvanized metal boxes
were designed and constructed with treated plywood bottoms to house each ISCO and a 12-v
battery power source.
A 6-inch diameter PVC stilling well, [with 1-inch x 6-inch slots cut at several locations
along the side wall to a height of estimated stream rise (i.e. 5 feet) to allow water in/out] is used
to surround and fix the position of the YSI probe to protect it from debris, and to support the two
ISCO strainers positioned (one on each side) on the outside of the stilling well, about 2 inches
from the stream bottom. A support stand for the YSI probes was designed and constructed to
secure and position the YSI probe in the stilling well at 2 inches from the stream bottom. The
bottom of the stilling well is secured by large clamps to metal fence posts driven into the stream
bed. The top of the stilling well is secured to a bridge rail at SFBR10, the bridge pillow at
SFBR30, and trees at SFBR20, SFBR40, and SFBR60. A 4-inch PVC pipe was run from the
stilling well to the ISCO sampler to house and provide access to the ISCO tubing and YSI cable,
as needed. The PVC pipe was installed with a gradual slope to prevent potential water ponding
in the ISCO tubing that could cause contamination and purging problems. The ISCO tubing is at
the maximum length of 99 feet (as provided by the manufacturer) at SFBR30 and 85 feet at
SFBR60.
Due to drought conditions causing low stream water levels, the YSI probe was removed
from the stilling wells at sites SFBR20 and 60 and secured in a horizontal position using a 3-inch
PVC pressure coupling. For the location of the pressure transducer on the YSI to measure
stream level, water levels must be at least 8 inches deep located in a vertical position or at least 4
inches deep in a horizontal position.
At the watershed outlet site SFBR70, installation of the support structure to secure the
two ISCO strainers and YSI multi-probe near the center of the 180-foot wide river at this point
was very difficult because the river bed consisted of loose sand (2 feet deep) overlying bedrock.
This installation required a contractor to bore into the river bed rock with an air drill and install
ten, 8 feet long drill bits located at 10-foot intervals. At each drill bit, a 2-inch wide unistrut was
placed over the drill bits and driven into the stream bottom as deeply as possible. A 4-inch PVC
pipe was then placed over each drill bit/unistrut and driven into the stream bottom; the unistrut
extended above the PVC pipe approximately 6 inches. The PVC pipe was then filled with
22
-------
concrete for extra strength. A 4-inch PVC pipe, containing ISCO tubing and YSI vented
connecting cable, was installed from the cableway platform along the top of the unistrut posts to
the support frame for the YSI probe that was located 70 feet into the river from the bank. The
support frame was designed to allow easy access to the YSI probe for servicing. The support
frame consists of a 2-inch wide metal unistrut frame connected to 6-inch PVC couplings that
slide over 4-inch PVC support posts located 4-feet apart.
During rain events, the support structure as described above at SFBR70 site, trapped
floating trees and debris. This caused a logjam that partially destroyed the supporting structure
at this site including drill bits and breaking the pvc pipe. The damage by the trees required
moving the YSI and ISCO sampling equipment within 30 feet from the river bank. Later,
another large tree lodged on top of the sampling equipment and deposited a large area of sand.
However, the sampling equipment was not damaged. One idea was to solve this problem by
installing a large pipe along the bottom of the river from the cableway platform and the sampling
equipment positioned at the end of the pipe. This idea was eliminated when large deposits of
sand was observed at SFBR70 that would have caused the sampling equipment to be buried
underneath. All supporting equipment that was originally installed was later removed from the
river and new supporting equipment was relocated near the stream bank to eliminate potential
problems with logjams and damage to sampling equipment. It is very difficult to keep
monitoring equipment in the river to collect representative stream data without causing serious
trapping problems.
Installation of Cableway Sampling Systems
Three cableway systems were installed under an Interagency Agreement with the U.S
Army Corps of Engineers, Vicksburg, MS, working in concert with staff from Athens ERD's
Field Research Annex. Sites SFBR30 and 60 were the first cableways installed using the model
DDT 900 system from Rickly Hydrological Company, 1700 Joyce Avenue, Columbus, Ohio
43219. The original design of the equipment when the order was placed with Rickly was to be
powered by 12-v power. The purchase order was changed later to be powered by 110-v
electrical system. After installation, many mechanical problems became apparent using the first
two cableway s, specifically the systems were found not to be capable of lifting the 170-pound
bedload sampler. This required disassembling the system for repair and modification by the
manufacturer, including installation of a more powerful AC motor and gear drive mechanism.
Extensive time and effort was then required to calibrate the cableways for collecting stream
samples at designated locations across the stream channel, and to develop SOP's for sample
collection using the bedload and depth-integrated samplers.
The last site, SFBR70, became operational May 16, 2002. An OTT (OTT Hydrometry,
Ludwigstr 16, PO Box 2140, Kempton, HI D-87411) model SK-V-G/W cableway system was
installed at this site.
At field sites SFBR30 and 70, direct AC electrical power is used to power the cableways.
23
-------
At the SFBR70 site, an 1 Ikw generator, operating on propane fuel, has been installed for backup
power with an automatic start-up and transfer switch to provide continuous power in the event of
electrical power loss during sampling events. The SFBR60 site has a similar 1 Ikw
propane-fueled generator setup to provide all the electrical power for equipment and lights.
There is no backup generator power at the SFBR30 site since the overhead electrical lines are the
major power source for that area and the chances of extended power outage are minimal. Each
of the three cableway sites is designed with proper lighting at the cableway platform and across
the river for night time sampling and worker safety.
Site Maintenance
Maintenance of the 6 sites required frequent visits to remove and exchange the YSI probe
for cleaning, calibration and individual probe replacement. Data was being collected
continuously from each site; however, drought conditions resulted in low water levels that
caused severe damage to the YSI probes. In addition, the stream bed sediment constantly shifts
causing "silting-in" of the monitoring equipment. The state-of-the-art field equipment being used
does not appear to be sufficiently rugged for continuous instream monitoring. After each storm
event sampling, debris has to be removed from the site, and in some cases logs have to be cut
with a chain saw for removal.
Communication at Field Sites
Several methods were used to communicate between field and laboratory staff and for
emergency use. These methods included cell phones walkie-talkies, pagers and a base station
located at the FRA. Four field sites, SFBR10, 30, 60, 70, and the weather station had telephones
(i.e. voice and modem); however, the weather station had only a modem line. Newer model
ISCO samplers have been installed with modems and remote access control via phone line
connection between the four high priority stream sites (i.e. SFBR10, 30, 60, 70) in the watershed
and the FRA. This saved the FRA staff valuable time since the samplers could be accessed
remotely instead of requiring personnel to go to the field to activate the instruments before each
rain event. It also ensured that the beginning of each rain event would be sampled since no time
would be lost in driving to the different sampling sites while the event was progressing. This
allowed more control during the most critical time of the rain event by reducing response time
for sampling start-up. It also allowed remote monitoring of the sampling operation and
minimized the number of samples lost because of equipment malfunction.
Laboratory Modifications
In order to conduct the large number of required analyses for this project, extensive
modifications were required at the FRA laboratory. Installation of individual laboratory cooling
and vacuum systems was required to handle the heat load of the many ovens and muffle furnaces
needed to conduct TSS analyses.
24
-------
Figure 13: ISCO Setup at Each Sampling Site
Figure 14: YSI Multi-Probe
25
-------
To properly store the large number of analytical samples required the design, purchase
and installation of a walk-in freezer and cooler.
Storage of all sampling equipment and personnel protective safety equipment for
sampling team members required installation and modification of two storage buildings (12 feet
X 50 feet).
Vehicle Requirements
A fleet of 4-wheel drive vehicles, some with specialized equipment such as cranes
mounted on flat bed trucks for sampling from bridge structures, were required to collect storm
event samples at the 6 sites. Some sites required the use of a Kawasaki Mule (4x4) to transport
equipment for routine maintenance and access during storm event sampling. A boat was
required to service the stream monitoring equipment at the SFBR70 site.
Battery Charging
A large quantity of 12-v battery power sources were required to operate equipment at
remote field sites. A battery-charging facility was constructed at the FRA to support this need in
a safe environment for our staff.
Stream Cross Sections
In addition to the six stream monitoring sites, 279 stream cross sectional sampling sites
were selected, surveyed and samples collected from 90 sites of these sites for characterization
(Figure 15). Engineering surveys were conducted using a TOPCON total station laser surveying
instrument. Locational data was obtained using a Global Positioning System (GPS). Stream
elevations are being determined between sites. In addition, sediment core samples are being
collected at 6-inch depth increments for characterization of particle size and organic carbon
content by Loss on Ignition (LOI). Core samples were collected at each cross section at five
locations across the stream channel; at the center of right bank, the thalweg, equal distant
between thalweg and right bank sample, at the center of the left bank and equal distant between
the left bank sample and the thalweg. Digital pictures were taken at each location documenting
the sampling site; field notes were recorded in field log books. Laboratory analyses of the core
samples include particle size characterization for < 2-mm size using a LS 200 Beckman Coulter
Particle Size Analyzer, with the results shown as a histogram and volume % for selected particle
diameters (urn) (0.375, 2, 4, 16, 31, 63, 125, 250, 500, 1000, 2000). Particle sizes > 2 mm (2-4,
4-8, >8) are shown as fractions collected manually using wet sieving techniques, with the results
reported in dry weight and after LOI. LOI and dry weight are also reported for the total
sediment fraction < 2 mm.
26
-------
South Fork Broad River Watershed
Stream Cross Sectional Sampling Sites
N
I LA
w *5te£- E
s
SFBR10
• Cross Section Sites
• Cities
O Sam pie Sites
Streams
/\/SF Broad River
Roads
V,
DANIELSVILLE
o O«.
SFBR20 /
3 4 Mjie's
SFBR40
V
COMER
V
^.COLBERT
0 1 2 Miles
CARtTOH
---•'' O
SFBR60
Figure 15. Stream Cross Sectional Sampling Sites, 305 are Scheduled for Investigation
27
-------
Weather Station
A Class A weather station (Figure 16) was located near the town of Comer, GA, to
measure meteorological parameters at hourly intervals that include; precipitation rate, solar
radiation, relative humidity, temperature, wind speed/direction, barometric pressure, and
evaporation rate. The town of Comer provided the EPA with access to their property for the
installation of this weather station. A fence was constructed around the site for security. The
nearest existing weather station with long term records is located in Athens, GA, near the airport,
and is not in the watershed.
In addition to the weather station, seven tipping-bucket rain gages (RG1 through RG7)
were installed at other locations throughout the watershed to provide basin-wide spatial coverage
of rainfall quantities (Figure 17). Approval was obtained from the various land owners for the
installation of this equipment on their property.
The project area had four years of a severe drought condition beginning June, 1998, (i.e.
15 inches below the normal 51.5 inches/year rainfall for this area) as recorded by Mark Jenkins,
a cooperative observer of Madison County for the National Weather Service.
Storm Event Sampling
The sampling scheme consists of two automatic ISCO water samplers at each of six
stream sites; one collects samples for sediment and the other for nutrient/organic carbon analyses
at 1-hour sampling intervals. For the initial sampling of storm events during 2001 at sites
28
-------
"3k "*
it?!tt CB
'*%:$#&?£ ^-^^^Q
^A.» "*"-V>M» 7 ua;<^ «'t^.v*SS
S9/28/2000 14:40
Figure 16: Weather Station
w-.ite'x1 v-nV.AI. -I: F...
Figure 17: Rain Gage
29
-------
SFBR10 and 20, the sampling frequency for the ISCO, was 30 minutes. A YSI multi-probe at
each of the same six sites collected continuous instream data on depth (stage), turbidity, specific
conductance, temperature, dissolved oxygen, pH, and oxidation-reduction potential at 15-minute
intervals. A 6-inch Helley-Smith bedload sampler (Figure 18) and a DH- 59 depth-integrated
sampler (Figure 19) were used to collect samples at several points across the stream channel for
sediment analysis at the four previously listed priority sites at 2-hour intervals. A grab sample
was also collected for pathogen analysis at 2-hour intervals at mid-stream and mid-depth at the
four priority sites.
The number of expected rain events (2 inches or greater) to sample during the year is
difficult and uncertain to determine based on predictions by the weather service. Based on
analysis of historical rainfall data, the number of events producing at least 2 inches for Athens,
GA, ranges from 3 to 15 with a mean of 12.
For each storm event tracked for potential sampling, Athens-ERD staff decided when to
make the call to initiate and stop sampling. An EPA contractor provides the staff required to man
the four priority stream monitoring sites in the SFBR. A minimum time notice of 24 hours was
provided to the contractor by Athens-ERD to alert, assemble and organize the sampling teams
for each of the sites. After the alert notice, EPA made the call to the contractor regarding when
to actually mobilize the teams for sampling. After the call for mobilization, the contractor was
expected to be at the four high priority stream monitoring sites within 2 hours.
The contractor collected the bedload and DH-59 depth-integrated sediment samples at
2-hour intervals at several established calibration points (5 to 10) across the stream channel. The
DH-59 samples are instantaneously collected samples whereas the bedload samples are
composites from across the stream channel over each 2-hour sampling interval. The DH-59
samples will be used to assess cross-sectional variability in fluvial sediment. The contractor also
collected the grab samples for pathogens at 2-hour intervals at mid-stream and mid-depth.
Sampling continued round-the-clock at each site during both the rising and falling stages of the
runoff hydrograph until the stage falls one half the stage rise and/or turbidity level is 100 NTU or
less. At the beginning of each 12-hour shift, the contractor records the staff gage reading for
quality control and to establish a relationship between staff gage readings and the YSI level
readings in order to calculate contaminant loadings.
Pathogen Sampling
Sampling plans were developed to collect bacteriological samples for pathogens analysis
(i.e. as represented by the indicators fecal coliform, E. coli. and enterococci) during rain events
over several days at SFBR10, 30, 60 and 70 sites beginning in December 2002. Samples were
collected for bacteriological analysis from each site such that the maximum 6-hour holding times
for the analyses are not exceeded. The DH-59 sampler was modified with a customized
removable bottle holder for the 250-mL sterilized sample bottle in order to collect
uncontaminated bacteriological stream samples. An EPA field mobile laboratory was set up
30
-------
Figure 18: 6-inch Helley-Smith Bedload Sampler
Figure 19: Depth Integrated Sampler
11
-------
in Comer, GA, to conduct these analyses. The analyst set-up samples for incubation
immediately upon receipt. Analysis was conducted according to Region 4 SOP's and the
manufacturer protocols. All samples collected are prepared and processed according to specified
Quality Assurance/Quality Control (QA/QC) procedures.
Monthly Base Flow Sampling
Each month, base flow stream samples were collected for background analysis of total
suspended solids, nutrients and organic carbon from each of the six stream monitoring sites.
Pathogen samples were also collected for analysis as part of the base flow sampling, starting in
December, 2002. Staff gage readings were recorded at time of sample collection.
Field Sampling procedures
Field and laboratory SOPs (see appendix 1) were developed by EPA Athens-ERD and are
being followed, that incorporate those of the Analytical Support Branch, Ecological Assessment
Branch and Enforcement and Investigation Branch of EPA Region 4 (2002).
Stream Flow
EPA had two Interagency Agreements with the USGS to collect stream flow data: 1) a
real-time station was established at the SFBR70 site to collect long-term continuous river stage
(i.e. level), stream flow (i.e. discharge), and precipitation data at the USGS station 02191743 at
Carlton GA; and 2) long-term, non-continuous river stage (level), and stream flow (discharge)
data were collected for the remaining five stream sites (SFBR10, 20, 30, 40, and 60).
At the real-time station, data was collected at 15-minute intervals and transmitted via
satellite every four hours to the USGS office in Atlanta, GA. The web site can be accessed for
real-time data at: http://water.usgs. gov/ga/nwis/uv?02191743. The data was usually checked on
a daily basis to ensure that the station was working properly. A USGS field person visits the
gage frequently to ensure that the sensors are properly cleaned and calibrated, and usually
performs a stream flow measurement to relate stream flow to river stage. Additional data for the
state collected during 1999, 2000 and 2001 are published in the Georgia District Annual Data
CD-Report.
For the collection of non-continuous stage data at the other five sites by the USGS, a staff
person visited each site to perform stream flow measurements at frequent intervals to develop
and maintain site-specific rating curves to relate stream flow to river stage. After a series of
measurements were made that covers a full range of river stages, a rating curve was created to
relate river stage to stream flow. Once this rating curve has been verified and entered into the
database, it can be used to calculate stream flow values associated with the staff gage readings or
the level data obtained by the YSI multi-probe. Continued stream flow measurements are
required after the determination of the initial rating curve to ensure that it is still valid. This
relationship can change due to changes in the channel, vegetation growth, and other factors. If a
series of measurements are found to deviate from the rating curve by more than five percent, a
new rating curve is usually called for. Stream stage-flow rating data for the five stations are
32
-------
presented in Appendix 10.
Once every three years, USGS staff from outside the Georgia District perform an external
review to ensure that the practices used are nationally accepted procedures.
Stream Velocity Profiles
Vertical stream velocity profiles were conducted at sites SFBR 10, 20, 30 and 70 during
February 7, 8 and 12, 2002 at varying stream stages for use in the calibration/verification of
mathematical modeling of stream bed scour and sediment transport. Precipitation of 2.2 inches
(Figure 20A) was recorded at the USGS real-time station at the SFBR70 site during the period
of 0600 on 2/6 to 1200 on 02/07/2002. Stage and discharge data also collected at SFBR70 site is
shown in Figures 20B and 20C. This rain event provided an opportunity to measure stream
velocities at a relatively high stage and the following stream recession at lower stages.
Accordingly, velocity measurements were made on Feb. 7 and 8, and then after the stream stage
receded on Feb. 12, 2002.
Measurements were made by a combination of acoustical doppler velocimeters and
bridge suspension apparatus coupled with Price AA current meters. Velocity profiles were
measured at stream quarter points at selected representative stream sections. All profiles
proceeded from near the stream bed to near the stream surface. Typically, the range of
observation points was from 0.2 feet above the stream substrate to 0.2 feet below the stream
surface. All measurements were accompanied by a determination of stream stage by either "tape
down" from an established reference point (RP) or by existing staff gage.
The velocity profiles and their attendant data tables are shown in Figures 20D, 20E, 20F,
20G and 20H. In most cases, as presented, the profile data supported a regression fit of the
velocity versus depth data.
Laboratory Analysis Procedures
Specific analysis methods for sediment and nutrients in the low milligrams per liter
range, and for pathogens in counts per milliliter, are required for TMDL development and model
testing. All standard operating procedures (SOP's) are in Appendix 1. Analysis of TSS was
conducted using a modified method in Standard Methods, 1998, SOP EAB-004.0 . The analysis
of nutrients was performed using a Bran+Luebbe AutoAnalyzer 3, SOP EAB-003.0 whereas the
analysis of organic carbon utilized Shimadzu's TOC 5050A instrument with an AST 5000A
autosampler attached, SOP EAB-011.0. Particle size analysis for particles less than 2 mm was
conducted with a Beckman Coulter LS 200 instrument, SOP EAB-010.0. The analysis of
particles greater than 2 mm was conducted manually by wet-sieving.
All laboratory analyses for sediment, nutrients, and organic carbon were conducted at
EPA's Ecosystems Research Division Field Research Annex by a team having extensive
laboratory experience. See Figure 21 for the Laboratory Analysis Flow Chart. Pathogen
analyses were conducted in a mobile laboratory on site and at EPA's Region 4 laboratories.
33
-------
Holding times before analysis for the different species are 6 hours maximum for pathogens; 24
hours preferably, but not longer than 7 days, for TSS; 48 hours for nutrients; and indefinite for
bedload samples after freezing.
34
-------
PRECIPITATION, TOTAL, INCHES
Most recent value: .00 02-13-2002 11:30
USGS 82191743 SOUTH FORK BRORD RIVER HT CflRLTON, Gfl
2.5
g 2.8
o
z
H
£ 1.5
o
1.8
0-
M
LJ
e.s
8.8
oc:
=3
CO
Feb 06 Feb 87 Feb 98 Feb 89 Feb 10 Fcb 11 Feb 12 Feb 13
DfiTES: 82/96/28B2 to 82/13/2882 11:38
Download a presentation-quality graph Parameter Code 00045; DD 03
Questions about data gs^w^ia_NWISWeb_Data_Inquiries.@usgs.gov
Feedback on this websitegs-w-ga_NW1 SWebJMaintainertSiusgs. gov
Real-time Data for Georgia
http://water.usgs.gov/ga/nwis/uv?
Retrieved on 2002-02-13 14:36:53 EST
Department of the Interior, U.S. Geological Survey
USGS Water Resources of Georgia
Privacy Statement || Disclaimer || Accessibility
1.78 1.34
Return to top
ofpase
Figure 20a. USGS Precipitation Data at SFBR70 site, 02-13-2003
35
-------
GAGE HEIGHT, FEET
Most recent value: 1.97 02-13-2002 11:30
USGS 92191743 SOUTH FORK BROflD RIVER RT CflRLTON, Gfl
3,88
l.Bfl
Feb 86 Feb 87 Feb 88 Feb 89 Fob 10 Feb 11 Feb 12 Feb 13
DflTESJ 02/86/2882 to 02/13/2802 11:38
Download a presentation-quality graph Parameter Code 00065; DD 01
Figure 20b. USGS Gage Height Data at SFBR70 site, 02-13-2003
36
-------
DISCHARGE, CUBIC FEET PER SECOND
Most recent value: 162 02-13-2002 11:30
USGS 02191743 SOUTH FORK BROfiD RIVER RT CHRLTON, Gfi
2086
8 180e
UJ
e.
u
M
a
a
ac
u
OT
H
B
180
Feb 86 Feb 87 Feb 88 Feb 89 Feb IB Feb 11 Feb 12 Feb 13
DBTES: 82/86/2882 to 82/13/2882 lit38
Download a presentation-quality graph Parameter Code 00060; DD 02
Daily mean flow statistics for 2/13 based on 0 year of record in ft /sec
'"™ !, : ':
Current |__. . L.
_. Minimum ;Mean
Flow 1 1 ;
162 | - ! - !
Maximum;
i
80 percent
exceedence j
i
50 percent
exceedence
\
20 percent
exceedence
—
Percent exceedance means that 80, 50, or 20 percent of all daily mean flows for 2/13
have been greater than the value shown.
Figure 20c. USGS Discharge Data at SFBR70 Site, 02-13-2003
37
-------
Left Quarter Point
Velocity Profile - SFBR10 @ Old Ma Rd
FebS and 12,2002
2.5
Feb8 -Left Qtr Pt - Surface at 1.8 Ft above Bottom _
Feb 12 - Left Qtr Pt - Surface at 1.5 Ft above Bottom
0.5
1.5
Water Velocity-Ft/Sec
Feb 8 Fe^-12
Center Line
Velocity Profile - SFBR10 @ Old Ma Rd
Feb 8 and 12,2002
2.5
1.5
0.5
0.5
1.5
Water Velocity - Ft/Sec
Feb 8 - CL - Surface at 0.9
Regression -PoMSjfit
Feb 12 - CL - Surface at 0.
Regression - Power Fit
Ft above Bottom
5 Ft above Bottom
/ /x ^>
_/ fy^
(7\- ^f\
" " "c^R - -^-^
^ff^^
SFBR10
Feb 8 and 12, 2002
Feb 8
Total Width at Section = 32 Ft
Staff Gage = 21.38 Ft
Feb 12
Total Width at Section = 29 Ft
Staff Gage = 20.96 Ft
Right Quarter point
Velocity Profile -SFBR10 @ Old Ma Rd
FebS and 12, 2002
£
E 2-5
re
& ,r
0>
8 n f.
« 0.5
5
n
-------
Quarter Point Velocity Profiles
gaff
Gage: 21.38 Ft
Left Qtr Pt
Depth*
Ft
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.7
Velocity
FPS
0
0.02
0.43
0.66
0.55
0.59
0.54
0.34
0.4
0.33
[tTapeDown: 19.47 Ft
Center Line
Depth
Ft
0
0.1
0.3
0.5
0.7
0.8
Velocity
FPS
0
0.91
1.04
1.37
1.38
1.52
(Stream Width: 32.0 Ft
Rt Qtr Pt
Depth
Ft
0
0.1
0.3
0.5
Velocity
FPS
0
0.47
0.86
0.96
gaff
Gage: 20.96 Ft
Left Qtr Pt
Depth
Ft
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Velocity
FPS
0
0.51
0.64
0.65
0.6
0.51
0.45
0.43
[Tape Down:
Center Line
Depth
Ft
0
0.2
0.4
0.5
Velocity
FPS
0
0.94
0.97
1.04
IStream Width: 29.0 Ft
Rt Qtr Pt
Depth
Ft
0
0.1
0.2
Velocity
FPS
0
0.57
0.57
Station: |
Staff Gage: |
Left Qtr Pt
Depth
Ft
Velocity
FPS
Lft Qtr Pt* = As looking up stream.
Depth* = Distance above stream bed.
Figure 20d. Continued
Date: |
Tape Down: |
Center Line
Depth
Ft
Velocity
FPS
Time: |
Stream Width: |
Rt Qtr Pt
Depth
Ft
Velocity
FPS
39
-------
Left Quarter Point
Velocity Profile - SFBR20 - Dbl Br CR @ Hyw 29
Feb 8,2002
£
E
0.5
1
*LeftQtrPt- Surface at 0.4 Ft above Bottom
0.5
>K
1 1.5 2
Water Velocity-Ft/Sec
Feb 8
2.5
SFBR20
Feb 8, 2002
Total Width at Section = 12 Ft
Staff Gauge = 21.36 Ft
fi 1.5
E
re
I 1
u 0.5
I
5
Center Line
Velocity Profile - SFBR20 - Dbl BR Rd @ Hyw 29
Feb 8, 2002
* CL - Surface at 0.85 Ft above Bottom
Regression - Power Fit
0.5
1 1.5 2
Water Velocity - Ft/Sec
Feb 8
2.5
Figure 20e. Stream Velocity Profiles for SFBR20 Site, 02-8-2003
E
E
u 0.5
S
w
5
Right Quarter Point
Velocity Profile - SFBR20 - Dbl Br Rd @ Hyw 29
Feb 8,2002
'Right QtrPt- Surface at 1.15 Ft above Bottom
0.5
1 1.5 2
Water Velocity - Ft/Sec
Feb,8
2.5
40
-------
Quarter Point Velocity Profiles
ptaffGage: 21.36 Ft
[Tape
(Stream Width: 12.0 Ft
Left Qtr Pt
Depth*
Ft
0
0.1
0.2
0.3
Velocity
FPS
0
1.39
1.46
1.45
Center Line
Depth
Ft
0
0.15
0.25
0.35
0.45
0.55
0.65
0.75
Velocity
FPS
0
1.05
1.22
1.25
1.45
1.49
1.47
1.6
Rt Qtr Pt
Depth
Ft
0
0.15
0.25
0.35
0.45
0.55
0.65
0.75
0.85
0.95
1.05
Velocity
FPS
0
0.82
0.88
1.04
0.82
0.94
0.68
0.93
0.73
0.7
0.68
Station: |
Staff Gage: |
Left Qtr Pt
Depth
Ft
Velocity
FPS
Date: |
Tape Down: |
Center Line
Depth
Ft
Velocity
FPS
Time: |
Stream Width: |
Rt Qtr Pt
Depth
Ft
Velocity
FPS
Station: |
Staff Gage: |
Left Qtr Pt
Depth
Ft
Velocity
FPS
gate
Lft Qtr Pt* = As looking up stream.
Depth* = Distance above stream bed.
Figure 20e. Continued
Center Line
Depth
Ft
Velocity
FPS
Time: |
Stream Width: |
Rt Qtr Pt
Depth
Ft
Velocity
FPS
41
-------
Left Quarter Point
Velocity Profile -SFBR30 @ Hyw 172
Feb 7 and 12, 2002
10
£
E
Feb 7 - LeftQtrPt- Surface at 4.3 Ft above Bottom
Regression - Power Fit
Feb 12 -LeftQtrPt- Surface at 3.45 Ft above Bottom
Water Velocity-Ft/Sec
Feb 7 Feb. 12
Center Line
Velocity Profile -SFBR30 @ Hyw 172
Feb 7 and 12,2002
10
Feb 7 -CL-
Regression
Feb 12 -CL
Regression
Surface at 3.5 Ft above Bottom
- Power Fit
- Surface at 2.15 Ft above Bottom
- Power Fit
<> <">
w-
£^Q ^r^\>
10
Water Velocity - Ft/Sec
FeJb 7 Feb^12
Figure 20f. Stream Velocity Profiles for SFBR30 Site, 02-7 & 12-2002
42
SFBR30
Feb 7 and 12, 2002
Feb 7
Total Width at Section = 54 Ft
Tape Down from RP = 28.09 Ft
Feb 12
Total Width at Section = 41 Ft
Staff Gage = 39.94 Ft
Right Quarter Point
Velocity Profile - SFBR30 @ Hyw 172
Feb7 and 12, 2002
Feb 7 - Right Qtr Pt - Surface at 1.5 Ft above Bottom
Feb 12 - Right Qtr Pt - Surface at 0.8 Ft above Bottom
Regression - Power Fit
Water Velocity - Ft/Sec
Feb7 Feb12
"XT' ^^
-------
Quarter Point Velocity Profiles
[gtaffC
Left Qtr Pt
Depth*
Ft
0
0.5
0.8
1.3
1.8
2.3
2.8
3.3
3.8
4.2
Velocity
FPS
0
2.9
1.3
2.7
2.9
2.9
3.2
3.2
2.9
2.4
[Tape Down:
Center Line
Depth
Ft
0
0.5
1
1.5
2
2.5
3
3.4
Velocity
FPS
0
1.6
2.1
2.4
2.7
2.7
2.7
2.1
(Stream Width: 54.0 Ft
Rt Qtr Pt
Depth
Ft
0
0.5
1
1.4
Velocity
FPS
0
1.8
2.4
1.6
gaff
Gage: 34.94 Ft
Left Qtr Pt
Depth
Ft
0
0.3
0.5
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
2.3
2.5
2.7
2.9
3.1
3.3
Velocity
FPS
0
0.01
0.8
0.72
0.68
0.68
0.92
0.86
0.9
1
0.86
0.92
1.05
1.02
1
1.32
0.93
[Tape
Center Line
Depth
Ft
0
0.3
0.5
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
Velocity
FPS
0
1.23
1.36
1.35
1.37
1.39
1.5
1.46
1.57
1.5
1.59
(Stream Width: 41.0 Ft
Rt Qtr Pt
Depth
Ft
0
0.2
0.4
0.6
Velocity
FPS
0
0.41
0.88
0.88
Station: |
Staff Gage: |
Left Qtr Pt
Depth
Ft
Velocity
FPS
Lft Qtr Pt* = As looking up stream.
Depth* = Distance above stream bed.
Figure 20f. Continued
Date: |
Tape Down: |
Center Line
Depth
Ft
Velocity
FPS
Time: |
Stream Width: |
Rt Qtr Pt
Depth
Ft
Velocity
FPS
43
-------
E
E
E
£ i
:e Above S
* i>
£ •
^
Feb 7 -Left
Regression
Feb 8 -Left
Regression
Feb 12 - Let
ji ^~~~~-~~
Left Quarter
Velocity Profile -SFBR40
Feb 7,8 and 12,
QtrPt- Surface at 2.4 1
- Power Fit
Qtr Pt- Surface at 2.25
- Power Fit
•t QtrPt- Surface at 1.4
S °"
:£^^
Point |
- Brush Creek 1
2002 1
-t above Bottom
Ft above Bottom
Ft above Bottom
O ~^
r __-— ^
i i
0.5
1 1.5 2
Water Velocity -Ft/Sec
2.5
Center Line
Velocity Profile - SFBR40 - Brush Creek
Feb 8 and 12,2002
4
E
E
re
£„
:e Above S
* rs
£ '
§
Q
"1
Feb 8-CL-
Regression
Feb 12 - CL
Regression
r^T
0.5
Surface at 2.3 Ft above Bottom
- Power Fit
- Surface at 1.5 Ft above Bottom
- Power Fit /-,
/x
^
^
n ^ Y^
J& ^^<^>
'O /\_^-^^ ^
-^ /L^^~^
1_ -^7^ i i
1 1.5 2 2.5 3
Water Velocity - Ft/Sec
Fejb 8 Feb^12
Figure 20g. Stream Velocity Profiles for SFBR60 Site, 02-7, 8, & 12-2002
44
SFBR40
Feb 7,8 and 12, 2002
Feb 7
Total Width at Section = 31.5 Ft
Tape Down from RP = 9.83 Ft
Staff Gauge = 12.70 Ft
Feb 8
Total Width at Section = 26 Ft
Tape Down from RP = 10.51 Ft
Staff Gauge = 12.03 Ft
Feb 12
Total Width at Section = 27 Ft
Staff Gauge = 11.44 Ft
Right Quarter Point
Velocity Profile - SFBR40 - Brush Creek
Feb8 and 12, 2002
Distance Above Stream Bottom - Ft
_=.„„*
Feb 8 - Rt QtrPt- Surface at 2.15 Ft above Bottom
Regression - Power Fit
Feb 12 - Rt
Regression
~
' 0.5
QtrPt- Surface at 1.4 Ft above Bottom
- Power Fit
P, Y\-
8 ^^o
/ fV ^-^X/
1 1.5 2 2.5 3
Water Velocity - Ft/Sec
Feb 8 Feb^12
-------
Quarter Point Velocity Profiles
ptaffGage:
Left Qtr Pt
Depth*
Ft
0
0.5
0.9
1.4
1.9
2.3
Velocity
FPS
0
1.6
2.7
2.7
2.8
2.1
Date: 02/07/2002I
Tape Down: 9.83 Ft |
Center Line
Depth
Ft
Velocity
FPS
Time: 1430-1500 |
Stream Width: 31.5 Ft |
Rt Qtr Pt
Depth
Ft
Velocity
FPS
gaff
Gage: 12.03 Ft
Left Qtr Pt
Depth
Ft
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
2.15
Velocity
FPS
0
0.95
1.06
1.52
1.67
1.58
1.57
1.44
1.67
1.82
|[TapeDown: 10.51 Ft
Center Line
Depth
Ft
0
0.2
0.5
0.75
1
1.25
1.5
1.75
2
2.2
Velocity
FPS
0
1
1.36
1.85
2.05
2.03
2.16
2.08
2.13
2.16
IStream Width: 26.0 Ft
Rt Qtr Pt
Depth
Ft
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
2.05
Velocity
FPS
0
1.19
1.51
1.81
1.85
1.73
2.01
1.96
2.11
2.16
|gtatiorr_
[gate
ptaffGage:
Left Qtr Pt
Depth
Ft
0
0.2
0.4
0.6
0.8
1
1.2
1.3
Velocity
FPS
0
0.02
0.71
0.84
0.84
0.88
0.88
0.89
[Tape
Lft Qtr Pt* = As looking up stream.
Depth* = Distance above stream bed.
Figure 20g. Continued
Center Line
Depth
Ft
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Velocity
FPS
0
0.58
0.84
1.04
1.03
1.02
1.13
1.18
(Stream Width: 27.0 Ft
Rt Qtr Pt
Depth
Ft
0
0.2
0.4
0.6
0.8
1
1.2
Velocity
FPS
0
0.63
0.88
0.91
0.91
0.96
0.97
45
-------
Left Quarter Point 1
Velocity Profile - SFBR70 @ Hyw 22-Near Carlton 1
Feb 7, 2002 1
Distance Above Stream Bottom - Ft
=> 10 .U 0> CO 0
Feb 7 - Left Qtr Pt- Surface at 6.25 Ft above
Regression - Power Fit
o ^
~Jr
Bottom
^/^
. ^^,
SFBR70
Feb 7 and 12, 2002
Feb 7, 2002
Total Width at Section = 183 Ft
Tape Down from RP = 31.14 Ft
Feb 12, 2002
Total Width at Section = 183 Ft
Tape Down from RP = 32.07 Ft
Water Velocity-Ft/Sec
Feb 7
10
Center Line
Velocity Profile -SFBR70 @ Hyw 22-Near Carlton
Feb 7 and 12,2002
Feb 7 - CL - Surface at 5
Regression - Power Fit
Feb 12 - CL - Surface at
-Re-gressiorr - Power Fit
w
. sf , _^>-
3 Ft above Bottom
4.0 Ft above Bottom
O /
--<^
<
,<-: <~2
^<
-
-------
Quarter Point Velocity Profiles
[gtaffC
Left Qtr Pt*
Depth*
Ft
0
0.5
0.75
1.25
1.75
2.25
2.75
3.25
3.75
4.25
4.75
5.25
5.75
6.15
Velocity
FPS
0
0.5
0.5
0.8
0.8
1
1
1.3
1.6
1.6
1.6
1.6
1.8
1.6
[tTapeDown: 31.14 Ft
Center Line
Depth
Ft
0
0.5
0.8
1.3
1.8
2.3
2.8
3.3
3.8
4.3
4.8
5.2
Velocity
FPS
0
1
1.3
1.6
1.6
1.8
1.8
1.8
2.1
1.95
1.8
1.6
(Stream Width: 183 Ft
Rt Qtr Pt
Depth
Ft
0
0.5
0.8
1.3
1.8
2.3
2.8
3.3
3.8
4.3
4.7
Velocity
FPS
0
1
1.3
1.3
1.6
1.8
1.8
1.8
1.8
2.1
1.6
Station: SFBR70 |
Staff Gage: |
Left Qtr Pt
Depth
Ft
Velocity
FPS
I
[[Tape Down: 32.07 Ft ||
IStream Width: 183 Ft
Center Line
Depth
Ft
0
0.5
0.75
1
1.25
1.5
1.75
2
2.25
2.5
2.75
3
3.25
3.5
3.75
Velocity
FPS
0
0.24
0.19
0.24
0.3
0.3
0.3
0.3
0.35
0.3
0.3
0.3
0.3
0.35
0.41
Rt Qtr Pt
Depth
Ft
0
0.5
0.85
1.1
1.35
1.6
1.85
2.1
2.35
2.6
2.85
3.1
3.35
Velocity
FPS
0
0.14
0.21
0.27
0.22
0.22
0.21
0.24
0.24
0.35
0.37
0.36
0.38
Station: |
Staff Gage: |
Left Qtr Pt
Depth
Ft
Velocity
FPS
Lft Qtr Pt *= As looking up stream.
Depth* = Distance above stream bed.
Figure 20h. Continued
Date: |
Tape Down: |
Center Line
Depth
Ft
Velocity
FPS
Time: |
Stream Width: |
Rt Qtr Pt
Depth
Ft
Velocity
FPS
47
-------
Training of Field Sampling and Laboratory Analysis Teams
Approximately 25 EPA volunteer field team members from Athens ERD were trained for
both day and night time sampling. In addition to the field team, a laboratory analysis and logging
team was organized possessing extensive laboratory experience in conducting these analyses and
in the logging of samples into the laboratory.
During October 2002, an EPA contractor was funded under an existing task order to
collect storm event bedload and depth-integrated sediment samples from four high priority sites as
discussed previously. The contract field staff were trained for sampling by EPA staff using the
EPA developed procedures and protocols.
Quality Assurance
Development of QA/QC procedures began in the project's planning stage and will continue
through sample collection, analyses, reporting and in the database development. The primary
objective of the project was to collect a representative set of high quality data for the SFBR
watershed TMDL study. One duplicate field sample was collected for pathogen analysis during
each 12-hour shift. One duplicate sample for sediment analysis was assigned by laboratory staff
from the ISCO samples approximately every 10th sample. Standards are run in the laboratory
before and after analyses and during selected (every 10th sample) analyses. Standard curves are
run before analysis. There was no TSS standard; however, balances were checked frequently.
Field duplicate samples of the sediment core samples were collected at every 1 Oth stream cross
sectional sampling site and noted with a "Q" at the end of the sample ID. Field notes were kept in
log books.
All field samples were kept on ice during sampling and transport to the FRA for analysis.
At the end of a 12-hour sampling shift, all samples were returned to the FRA, removed from the
ice chests and placed in cold storage. The DH-59 samples were placed in the walk-in-cooler,
grouped by site and accompanied with the following information: date and time of arrival; site
where samples were collected; number of DH-59 and bedload samples; and the names of the
sample collectors. The bedload samples were placed in the walk-in freezer, grouped by site, and
accompanied with the same documentation as with the DH-59 samples. All sample information is
recorded in a record book located in a secure box at the rear door of the FRA. ISCO samples were
also placed in the walk-in-cooler by FRA personnel, grouped by site, and labeled with the date and
time at which the last sample was collected.
All samples collected in the field for pathogen analysis were held on ice at <10°C. These
samples are picked-up frequently and analyzed as soon as possible to ensure that the maximum
allowable holding time of 6 hours is not exceeded.
Field records were kept in log books for each of the four high priority stream monitoring
sites to include: site shift leader; date and time of hourly stream stage data for depth; and turbidity
from the LCD of the ISCO sampler on site. In addition, any problems noted during sampling was
required to be recorded. All log books remained on site except at SFBR10 which remained in the
crane truck.
48
-------
ISGOA
TSS
TOC
Stream
Samples
ISGO B
DH 89/48
Centrifuge
Nitrate
Anmonia
Total
Phosphorus
Ortho
Phosphorus
TSS
Figure 21: South Fork Broad River Laboratory Analysis Flow Chart
49
PSD
Dry
Weight
LOI
Bedload
Wet Sieve
2nm C 4mm
4mm CSrrm
Dry
Weight
Dry
Weight
LOI
Dry
Weight
LOI
LOI
-------
Database Development
A comprehensive relational database system has been designed and populated with data
collected from the South Fork Broad River Watershed sampling sites. The database resides in
MySQL database server. Field data are recorded in various formats including hand written field
and lab sheets, text files, and excel worksheets. A software system has been developed for
transferring data from text files and excel worksheets to the relational database. The database
design, architecture, user access, quality control, security, and other details will be made
available in a separate report.
50
-------
Chapter 4
Supporting Projects in SFBR
The following were projects that supported the South Fork Broad River Watershed
Research Project.
1. Interagency Agreement with the U.S. Geological Survey
a. Through an IAG with the USGS, a real-time gaging station was installed near
Carlton, GA at the watershed outlet (SFBR70) to provide real-time
stage-discharge and rainfall data. The station web site can be accessed at:
http://water.usgs. gov/ga/nwis/uv?02191743.
b. Through another IAG with the USGS, stage-discharge rating curves are being
developed by installing staff gages at the six stream monitoring sites and then
measuring the discharges corresponding to at least ten different stages during rain
events. Data collected under the original IAG consisted of low flow stream rating
curves. A new IAG is in effect that includes a range of low and high stream flow
conditions to provide critical data for estimating contaminant loading rates.
2. Task Order through Region 4 contractor, Integrated Laboratory Systems, Inc.
This task order was funded to conduct sediment investigation in the SFBR watershed as
part of the development of a sediment TMDL protocol. The task order includes: conducting
stream cross-sectional surveys for about 300 sites; developing elevations of the stream cross
sections relative to bridge benchmarks; and collecting storm event samples for bedload and
depth-integrated samplers at four high priority sites (SFBR10, 30, 60, and 70).
3. Interagency Agreements with the U.S. Army Corps of Engineers.
a. Through an IAG with the U.S. Army Corps of Engineers, three cableway stream
sampling systems were successfully designed and constructed. The installation of
the cableway sampling systems was critical to our field research project in order
to provide a safe environment for sampling teams to collect the required bedload
and depth-integrated samples during intensive storm events. To our knowledge,
these are the only automated, unmanned stream sampling systems of their size in
the United States.
b. Through another IAG with the U.S. Army Corps of Engineers, a computerized
data repository is being developed that can incorporate information from different
sources and hardware platforms, that will provide data exploration capabilities,
and that make the data accessible to users via the Internet. The repository will
accommodate geospatial data, tabular data, and electronic documents.
Organizational goals and needs, including watershed level extrapolation and
modeling, multi-model integration, regional scale analyses, and interagency
collaboration and coordination are being considered in this development.
51
-------
4. Cooperative Agreement with the University of Georgia.
A Cooperative Agreement was developed with the University of Georgia to measure
sediment at selected sites in the SFBR watershed using a pressure transducer and to compare
these data with actual stream sediment data collected during storm events. Results of this
research effort thus far indicate that development and use of pressure transducers are in a
research mode and not ready at this time to produce reliable and defensible data in the field
during storm events for our TMDL project.
5. In-house research Project by Roger Burke.
The research project described below is currently ongoing in the SFBR and will provide
additional data.
Nutrient and Trace Gas Concentrations in Headwater Streams of the South Fork Broad
River Watershed
Introduction
According to the U.S. Census Bureau (http://quickfacts.census.gov/qfd/), population
increases in Madison and Oglethorpe counties, which contain most of the SFBR watershed,
between 1990 and 2000 were 22% and 29%, respectively. By comparison, between 1990 and
2000 population increases in the state of Georgia and in the U.S. were 26% and 13%,
respectively. Population is expected to increase in the Savannah River basin in the future at a
similar rate. The additional growth is expected to place increased stresses on streams, notably
flashier hydrographs, due to increases in water use and levels of impervious surfaces in
developed watersheds, and reduced water quality due to greater contamination of streams by
sediments, nutrients, organic material, agricultural chemicals and other toxics [Mulholland and
Lenat, 1992; Georgia Dept. of Environmental Protection, 2001]. We have chosen to work on
headwater streams for several reasons. Largely because of their shallow depths and high
surface-to-volume ratios, headwater streams are highly efficient at retaining nutrients and
organic matter (Peterson et al., 2001). First- and second-order streams are a critical part of the
overall river network, comprising an estimated 95% of the total stream channels and 73% of the
total stream channel length in the US (Meyer and Wallace, 2001). Because of their importance
in the river network and high rate of biogeochemical cycling, headwater streams provide
valuable ecosystem services by reducing transport of various pollutants to downstream
ecosystems such as rivers and lakes (Meyer and Wallace, 2001). Headwater streams also
provide refuge for species from downstream ecosystems and are an important food source for
some birds (Meyer and Wallace, 2001). Headwater streams are inadequately protected, and
because of their small size, their ecosystem function is easily impaired by human disturbance of
their catchment, riparian zone, and channel (Meyer and Wallace, 2001). For these reasons,
headwater streams probably deserve more intensive study and greater protection.
Goals
The overall goals of this research are to: (1) relate watershed land use to concentrations
of trace gases, nutrients, oxygen, chl a (planned), dissolved organic carbon and nitrogen,
alkalinity, and conductivity and temperature in first order streams of the Georgia Piedmont; (2)
elucidate the instream and watershed processes responsible for the concentration and distribution
52
-------
of these parameters; and (3) evaluate the utility of trace gas concentrations as indicators of
stream ecosystem function. Land use changes induced by the growing human population will
continue to have marked influences on important watershed processes such as C and N cycling,
trace gas emission, and hydrology that are expected to directly impact stream ecosystem function
because of altered water flows and decreased water quality.
Approach and Results to Date
Seventeen headwater watersheds within the SFBR watershed, ranging from 0.5 to 3.4
km2 were selected (Figure 22). Percentages of forested land, agricultural and pasture land,
residential areas, wetlands and open water surfaces within the watershed were calculated from
the National Land Cover Data (NLCD) database. Water samples have been collected monthly
from November 2001 to the present at the outlet of each watershed.
Nitrogen Concentrations
Linear models relating land uses to the nitrogen concentrations in the streams were
developed by step-wise regression. Preliminary analysis of the data indicates that the percentage
of forested land in a watershed is a good indicator of nitrogen concentration in streams,
especially for total nitrogen (TN) and dissolved inorganic nitrogen (DIN). Similar relationships
between land-cover and stream N content have been observed at much larger scales by others
[e.g., Omernik, 1977; Jones et al., 2001]. Up to 70% of the TN in streams draining residential
areas was DIN. In contrast, streams in forested watersheds had relatively less nitrogen (45%) in
inorganic form and about twice as much nitrogen (45%) in organic form compared to streams
that drain residential areas. Streams in agricultural and pasture land-dominated watersheds had
about twice as much nitrogen (15%) in particulate form as those from either forested or
residential watersheds. The amount (%) of forested land within the watershed was the best
predictor to the concentrations of the different forms of nitrogen in the streams.
Trace Gas Concentrations
Nitrous oxide concentrations varied widely from 10 nM (atmospheric equilibrium
concentration) to nearly 80 nM among the streams. Overall, the streams draining watersheds
dominated by developed land use had the highest dissolved nitrous oxide concentrations,
although the difference is statistically significant only for comparisons between the forest and
mixed land use watersheds. Also, the streams draining watersheds dominated by pasture had
significantly greater nitrous oxide concentrations overall than streams draining forested
watersheds. These results suggest that small streams could be a significant source of nitrous
oxide to the atmosphere in some watersheds. Carbon dioxide concentrations in the streams
range from about 30 to 900 mM (about 3 to 75 supersaturated with respect to the atmosphere).
Similarly to that for nitrous oxide, the streams draining residential areas had the highest overall
carbon dioxide concentrations, although the only statistically significant comparison was
between them and streams draining forested areas. Methane concentrations ranged from about
0.06 to 40 mM (about 30 to 20,000 supersaturated with respect to the atmosphere). Overall,
streams draining watersheds dominated by pastures had significantly higher methane
concentrations than streams draining any other land use type. Similar to the case for nitrous
oxide and carbon dioxide, the streams from forested watersheds had the lowest methane
concentrations.
53
-------
Summary of Nutrient and Trace Gas Concentration Project
The observations discussed here support the contention that land uses within the
watershed have a significant impact on nitrogen concentrations and forms and trace gas
concentrations in headwater stream waters. Also, use of data derived from the NLCD database to
assess the impact of land uses on nitrogen and trace gas concentrations in streams at a local scale
is a promising approach. The data collected in this project, including a more detailed analysis,
will be reported at a later date in separate publications and a report.
54
-------
Chapter 5
RESULTS OF STUDY
Data from the weather station have been collected hourly since January 1, 2000 to the
present. Twenty (20) year monthly rainfall averages for the Danielsville, Georgia, area of the
watershed were provided by Mark A. Jenkins, a National Weather Service Cooperative
Observer. Comparison of these monthly averages with years 2000, 2001 and 2002 through July
is presented in Figure 22. These later year total monthly rainfall amounts are corresponding less
than the 20-year monthly averages except in September 2000 and March and July 2001. Figure
24 shows that the total rainfall for 2000 and 2001 was each about 15 inches less than the 20-year
annual average. The reduction in total rainfall was over 40 inches from January 2000 to August
2002.
More rainfall (number of events and the amount) occurred during 2003 than any other
year since the beginning of the field research project. A large rain event was predicted to occur
during early April 2003 that produced a week of field sampling with 4-days of intensive
sampling. This rain event was not as intensive as expected for producing heavy runoff but it did
provide a low intensity rain over several days. Unfortunately, during this sampling event, two
sites equipped with Rickly Hydrological Company type cableway systems malfunctioned and
became inoperable and unsafe and therefore eliminated further rain event sampling until
replacement winches could be installed. After many discussions with the Rickly Company, they
finally agreed to replace both winch systems, but not until next year, 2004. The cableway
system problems has caused unexpected delays in obtaining the required data during a period of
greater rainfall.
The YSI multi-probe instream data [depth or level, turbidity, specific conductance, pH,
DO, ORP, and temperature] was collected continuously at 15-minute intervals at each of the six
sites. We have attempted to collect continuous instream data; however, this was very difficult to
accomplish due to probe malfunctioning (mainly due to extremely low water flow condition), or
the probes being silted-in by sand during a storm event. Judgement calls were made to
determine when to omit erroneous data.
Baseflow (or background) sampling and analysis for TSS, NH3-N, NO-3-N, O-PO4 as P,
Total-P and TOC contents were conducted monthly to provide background data between rain
events for the six sites. Also, prior to an event, background samples were collected for analysis.
Rain event samples were collected by the ISCO samplers (i.e fixed location and depth in stream
cross-section) and analyzed for TSS, NH3-N, NO-3-N, Total-N, o-PO4 as P, Total-P and TOC.
In addition, rain event samples were collected at three to five locations across the stream
cross-section by the bedload and depth-integrated samplers. The depth-integrated samples were
analyzed for TSS and the bedload samples were analyzed for particle size distribution and loss
on ignition.
55
-------
South Fork Broad River Watershed
EPA Upper Watershed Sample Sites
EPA ORD/Region4
10 Miles
N
W -tfJbpi E
s
• Cities
o Upper Watershed Sample Sites
• TMDL Sample Sites
• EPA Facilities
Roads
'Streams
F Broad River
/\/ Broad River
I I SFBR Watershed
Figure 22. Location of Seventeen Small Watershed Sampling Sites in SFBR Study Area.
56
-------
4 --
n
XjS«
^
Figure 23. Comparison of Monthly Rainfall in SFBR with 20 Year Average
60
50
40
30
20
10
20 yr Average 2000
Figure 24. SFBR Average vs 2000 and 2001
2001
57
-------
250.0
E 200.0
(/>
•o
5 150.0
*
1 100.0
Q.
I/)
« 50.0
S
0.0
0
10 20 30
Elapsed Time (hr) from Time 00:00
40
o
10 15 20 25 30
Elapsed Time (hr) from Time 00:00
35
40
Figure 25. Comparison of TSS vs Stream Level at SFBR 10 Site for Storm Event on March 12,
2002
58
-------
The comparison of TSS values from samples collected during the March 12, 2002 rain
event by the ISCO samplers versus stream level (stage) measured by the YSI ( 25) indicate that
the highest concentrations of TSS match the observed stream peak flow levels. The comparison
of measured turbidity versus stream level (stage) rise by the YSI multi-probes does not appear to
be as well defined as the TSS comparison (Figure 26). Figure 27 shows the comparison of
measured TSS versus turbidity.
The data obtained from the ISCO samplers located at a fixed depth and cross-sectional
location representing a single-point measurement in the stream channel compare favorably with
the average data from the depth-integrated samplers taken at several locations across the stream
channel at the same site. Figure 28 shows a comparison of TSS values using these two
measurement methods for a single storm event. Figure 29 shows a scatter plot of each DI TSS
value versus the corresponding ISCO value. Because DI samples were not taken at exactly the
same time as the ISCO samples, linear interpolation between temporally adjacent ISCO samples
was used to estimate an ISCO TSS value that matched the time of a DI TSS value. These data
represent six rain-sampling events, and 75 total points.
The regression coefficient (R2) is highly significant (F(123) = 883, p<0.0001). Since the
intercept is greater than zero and the slope greater than one, there is an indication of bias
between the two values, namely that ISCO measures tended to be higher than DI values. But in
analysis of data from a later storm event, this bias was reversed (Appendix 2). Note the large
circular outlier at an ISCO TSS value of about 100. This could have been due to malfunction of
the ISCO sampling gear. Due to its location, this outlier actually depresses the estimate of the
slope somewhat. With this point removed, the slope estimate increases to 1.111 from 1.098.
Additional data derived from various size storm events are needed to confirm these results
before any reliable conclusions can be drawn.
Our results indicated that the information loss when automated ISCO samples are taken
every two hours, instead of every hour, is measureable, not severe. Figure 30 presents a typical
storm-event pollutograph, with the two-hour time series simply equivalent to the one-hour time
series, with every other value removed.
Figure 31 depicts a scatter plot that matches every other actual TSS value from the
one-hour time series to an interpolated value based on the two hour interval ISCO time series.
For example, if a one-hour ISCO time series had TSS values for hours 3, 4, and 5, the
corresponding two-hour time series would only have TSS values for hours 3 and 5. In Figure 30,
the actual TSS value of the one-hour series at hour 4 would be compared to a TSS value derived
from linear interpolation of the hour 3 and hour 5 values. This interpolated value is the mean of
the hour 3 and hour 5 values. A total of 145 data points from eight storm events were included
in the analysis.
59
-------
1000
0
10 20 30
Elapsed Time (hr) from Time 00:00
40
2 A
.*\
20
.O
9 9
£-.£.
24
. I
g
<•>
5 2
0)
— 1 1 Q
i .y
(/)
*• 1 R
1 .0
1 7
I . /
1 R
1 .0
1 ^
^
/\
/ \
/ X^_
/ ^-^^
/ "^ — ^^_
^^~
1 .0 n i i i i i i
0 5 10 15 20 25 30 35 40
Elapsed Time (hr) from Time 00:00
"igure 26. Comparison of Turbidity by YSI vs Stream Level at SFBR 10 Site for Storm Event on
March 12, 2002
60
-------
0
10 20 30
Elapsed Time (hr) from Time 00:00
40
1000
0
10 20 30
Elapsed Time (hr) from Time 00:00
40
Figure 27. Comparison of TSS vs Turbidity at SFBR 10 Site for Storm Event on March 12, 2002
61
-------
I
1800.0 -
1600.0 -
1400.0 -
1200.0
1000.0 -
w 800.0
600.0 -
400.0
200.0 H
0.0
1700
3/12/2001, Site 2
1800
1900
2000
Time (hours)
2100
2200
2300
Figure 28. Comparison of Total Suspended Solid Measurements of Depth
Integrated and ISCO samplers.
1800 -i
1600 -
1400 -
1200 -
1000 -
g> 800
600 -
400 -
200 -
y= 1.0983X +8.289
R2 = 0.9236
200
400
600 800
ISCO TSS (mg/l)
1000
1200
1400
Figure 29. Estimated linear relationship between Depth Integrated and ISCO
Total Suspended Solid measurements.
62
-------
The R2 value of 0.94 from the Figure 31 comparison plot indicates that only about 6% of
the total deviation was left unexplained by the regression relationship (which was highly
significant (F(l,143)=2154, p<0.0001)). This is one way to measure the cost incurred by
moving to a two-hour interval sampling regime. The 95% confidence interval on the slope
(0.847,0.922) does not contain 1.0, however. This means that even though very little
unexplained variation remained, the TSS values estimated by interpolation (y-values) tended to
be lower than the TSS values actually measured (x-values). We do not have an explanation for
this phenomenon at the moment, and more data need to be collected to investigate this
relationship.
As stated previously, when sampling for TSS using the DI methodology, TSS was
measured at three to five locations across the stream, and then a mean was computed. We were
interested in quantifying the cross-sectional variability of these multiple TSS measures and
determining any patterns existing in the variability. Figure 32 shows that the higher the mean
TSS, the more variability in the TSS measures. This is a typical occurrence and presents no
cause for alarm. Standardized variability, however, as measured by the coefficient of variation
(the standard deviation divided by the mean) showed no relationship to the mean TSS, Figure 33.
Finally, TSS values taken in midstream were not statistically different than TSS values
taken along the edges of the stream (Student's t test, p > 0.05). When 3 sites at a cross section
were sampled, site 2 (i.e. midstream) was compared to the mean of sites 1 and 3 (i.e. edge).
When 4 sites were sampled, the mean of sites 2 and 3 (i.e. midstream) was compared to the mean
of sites 1 and 4 (i.e. edge). When 5 sites were sampled, the mean of sites 2, 3, and 4 (i.e.
midstream) was compared to the mean of sites 1 and 5 (i.e. edge). A scatter plot showing the
midstream versus edge TSS measures is shown in Figure 34.
Bedload sampling represents the composite of three to five samples taken at the stream
cross section with time. Analyses include dry weight of sample, particle size distribution and
LOI for each size fraction. These samples consisted of a large quantity of leaves and woody
debris that were washed to remove the sediment. The debris was removed and not included in
the sample analysis. These data indicate that the sediment being transported during storm events
in the SFBR was primarily less than 2 mm.
About 90 of the 279 stream cross-sectional sites have been completed. Sediment core
samples were collected at 6-inch depth increments and analyzed including dry weight of sample,
particle size distribution and LOI for each size fraction. GPS locational data for the sampled
stream cross sections are available in the database.
Slope elevation estimates are being developed for longitudinal profiles at the sampled
cross-sectional sites along SFBR and at 1500 feet above and below existing bridges using bridge
elevation data from the Georgia Department of Transportation as benchmarks. This field survey
work is not complete. The stream slope appears to be less than one percent based on an estimate
between stream channel bottom thalweg points.
63
-------
160 -,
140-
120-
— 100-
^
O)
E
OT
OT
80-
60-
40-
20-
0
SiteS, 1/25/2002
2300 2800 3300
3800 4300
Time (hours)
4800 5300 5800
Figure 30. Comparison of one hour versus two hour ISCO Total Suspended
Solid measurements.
700 -,
600 -
500 -
OT
V)
400 -
•a
01
4-1
ro
I 300
V)
200 -
100 -
1 hour versus 2 hour ISCO sampling
y = 0.8846x+9.1656
R2 = 0.9378
0 100 200 300 400 500 600 700 800
Observed TSS
Figure 31. Linear relationship between one hour and two hour ISCO Total
Suspended Solid measurements.
64
-------
500 -I
0 200 400 600 800 1000 1200 1400 1600 1800
Mean TSS
Figure 32. Relationship between the standard deviation and the mean of Depth
Integrated Total Suspended Solids measurements.
1.20 -i
1.00 -
| 0.80 -
re
£ 0.60 -
'o
o
o
0.40 -
0.20 -
0.00
y = 2E-05x + 0.1463
R2= 0.0021
* *
v
0 200 400 600 800 1000 1200 1400 1600 1800
Mean TSS
Figure 33. Relationship between the coefficient of variability and the mean of
Depth Integrated Total Suspended Solids measurements.
65
-------
LU
1,800
1,600
1,400
1,200
1,000
800
600
400
200
0
Center/Edge TSS Comparison
y = 0.978x+0.0588
R2 = 0.9678
0 200 400 600 800 1,000 1,200 1,400 1,600 1,800
Center TSS
Figure 34. Relationship between Depth Integrated Total Suspended Solid
measurements taken at various locations within a stream cross-section.
In support of the South Fork Broad River Watershed Project, EPA Region 4 staff
conducted measurements of water current velocities at selected locations in the watershed. The
purpose of these measurements was to provide an assessment of the stream velocities at varying
stream stages for use in the calibration/verification of mathematical modeling of stream bed
scour and sediment transport. Vertical velocity profiles were measured at designated stream sites
(i.e. SFBR 10, 20, 30, 40, and 70) on February 7, 8 and 12, 2002. Rainfall of 2.2 inches was
recorded at SFBR 70 site during the period of 0600 on 02/02/2002 to 1200 on 02/07/2002. This
rainfall event provided an opportunity to collect stream velocities at a relatively high stage and
then following stream recession, at a lower stage. Accordingly, velocity measurements were
made on Feb. 7 and 8 and then after the streams receded on Feb. 12, 2002. Measurements were
made by a combination of acoustical doppler velocimeters and bridge suspension apparatus
coupled with Price AA current meters. Velocity profiles were made at stream quarter points
along representative stream sections. All profiles proceeded from near the stream bed to near the
stream surface. Typically the range of observation points was from 0.2 feet above the stream
substrate to 0.2 feet below the stream surface. All measurements were accompanied by a
determination of stream stage by either "tape down" from an established reference point "RP" or
by existing staff gage. The attached graphs and their attendant data tables shown in Figures 20A
through 20H present the results of the measurements for stream sites SFBR10, 20, 30, 40 and 70.
In most cases, as presented, the profile data supported a regression fit of the velocity versus
depth data.
66
-------
Hydrographic data, including stage-discharge relationships, water stage records and
velocity profiles, are available for each site. The water stage data includes both continuously
recording USGS station (SFBR70) input as well as that from non-recording stations (SFBR10,
20, 30, 40 and 60). These data and the attendant stage-discharge relationships provide the basis
for assessing stream discharge. When coupled with pollutant analytical results, pollutant
loadings can then be calculated for the various sampled runoff events.
Stream contaminant loading rates were determined by matching the date/time when
samples were collected by the ISCO sampler (30-minute or 1-hour increment) with the YSI
multi-probe level readings at 15-minute intervals adjusted to the staff gage readings provided by
the USGS. Then the contaminant loading rates were estimated by multiplying the flow in cubic
feet per second (CFS), obtained from the rating curve for a particular stream site times the
measured contaminant concentrations (mg/L) to obtain the loading rate in kg/hour as follows:
(ft3/sec) (mg/L) (28.32L/ft3) (3600 sec/hr) (gm/lOOOmg) (kg/1000 gm) or
(ft3/sec) (mg/L) 0.102 = kg/hr
Example contaminant loading rates, based on samples collected from ISCO samplers
were calculated for TSS, nutrients and carbon for each of the 5 instream sites during a storm
event that occurred on March 12, 2002, (see Tables 1,3,4, 5, 6, and 7). This event represents a
low -runoff-yielding event due to a rainfall amount of 0.93 inches. Because of the lack of a
stage-discharge rating curve for higher stages (due to the drought), contaminant loadings could
not be estimated for these type events.
Table 1 provides a summary of the calculated total event loadings (kg) for each of the
five sites and for each measured contaminant for the March 12, 2002 storm sampling event.
Estimated TSS loadings were highest at sites SFBR 30 and 60 at 23,642 and 34,248 kg,
respectively. In addition, highest loadings were calculated for ammonia, nitrate, phosphorus and
total organic carbon at these two sites. As expected, lower TSS loadings were estimated at site
SFBR20, which is a small tributary to the SFBR. One would expect even higher loads to occur
during higher rainfall events of 2 inches or more.
Table: 1. Loadings for SFBR sites for March 12, 2002 Rain Event
Site
SFBR10
SFBR20
SFBR30
SFBR40
SFBR60
TSS
kg
2179.7
195.9
23641.6
2244.8
34248.2
NH3-N
kg
2.0
0.7
14.0
5.5
8.0
NO' 3-N
kg
19.6
1.9
175.4
5.5
118.4
o-PO4 as P
kg
1.6
0.3
8.4
3.0
6.5
Total-P
kg
2.4
0.4
9.1
4.3
18.1
TOC
kg
168.0
19.8
1592.7
484.2
1121.8
67
-------
Table 3. Loading Rates for SFBR10 Rain Event on March 12,2002
YSIData
Date/Time
3/12/200221:30
3/12/200222:30
3/12/200223:30
3/13/20020:30
3/13/20021:30
3/13/20022:30
3/13/20023:30
3/13/20024:30
3/13/20025:30
3/13/20026:30
3/13/20027:30
3/13/20028:30
3/13/20029:30
3/13/200210:30
3/13/200211:30
3/13/200212:30
3/13/200213:30
3/13/200214:30
3/13/200215:30
3/13/200216:30
3/13/200217:30
3/13/200218:30
3/13/200219:30
3/13/200220:30
3/13/200221:30
3/13/200222:30
3/13/200223:30
3/14/20020:30
3/14/20021:30
3/14/20022:30
3/14/20023:30
3/14/20024:30
3/14/20025:30
3/14/20026:30
3/14/20027:30
3/14/20028:30
Level
(ft)
1.784
1.968
2.095
2.231
2.317
2.324
2.268
2.199
2.147
2.099
2.066
2.036
2.013
1.996
1.990
1.976
1.961
1.947
1.929
1.915
1.900
1.887
1.872
1.858
1.844
1.833
1.822
1.808
1.801
1.789
1.779
1.767
1.758
1.750
1.742
1.736
Stage
(ft)
20.924
21.108
21.235
21.371
21.457
21.464
21.408
21.339
21.287
21.239
21.206
21.176
21.153
21.136
21.130
21.116
21.101
21.087
21.069
21.055
21.040
21.027
21.012
20.998
20.984
20.973
20.962
20.948
20.941
20.929
20.919
20.907
20.898
20.890
20.882
20.876
Flew
(cfe)
9.02
10.70
11.80
13.11
13.87
13.94
13.38
12.78
12.26
11.88
11.55
11.25
11.03
10.85
10.80
10.75
10.61
10.46
10.28
10.15
10.00
9.92
9.79
9.67
9.54
9.45
9.36
9.24
9.18
9.07
8.99
8.88
8.81
8.74
8.67
8.62
ISCO Data from SFBR10
Sample*
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
36
Sample ID
1lsA031 20221 36
1lsA031 2022236
1lsA031 2022336
1lsA031 3020036
1lsA031 30201 36
1lsA031 3020236
1lsA031 3020336
1lsA031 3020436
1lsA031 3020536
1lsA031 3020636
1lsA031 3020736
1lsA031 3020836
1lsA031 3020936
1lsA031 3021 036
1lsA031 3021 136
1lsA031 3021 236
1lsA031 3021 336
1lsA031 3021 436
1lsA031 3021 536
1lsA031 3021 636
1lsA031 3021 736
1lsA031 3021 836
1lsA031 3021 936
1lsA031 3022036
1lsA031 30221 36
1lsA031 3022236
1lsA031 3022336
1lsA031 4020036
1lsA031 40201 36
1lsA031 4020236
1lsA031 4020336
1lsA031 4020436
1lsA031 4020536
1lsA031 4020636
1lsA031 4020736
1lsA031 4020836
Date
3/12/2002
3/12/2002
3/12/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
Time
(24hr)
2136
2236
2336
0036
0136
0236
0336
0436
0536
0636
0736
0836
0936
1036
1136
1236
1336
1436
1536
1636
1736
1836
1936
2036
2136
2236
2336
0036
0136
0236
0336
0436
0536
0636
0736
0836
TSS
mg/l
74.8
172.2
78.2
226.2
174.4
134.2
149.8
87.4
70.8
68.4
56.4
39.6
39.2
0.0
28.6
15.6
17.2
25.2
22.6
27.0
25.6
23.8
17.4
21.0
21.6
25.6
20.0
16.2
19.4
19.6
15.6
18.2
17.4
17.2
18.0
17.0
TSS
kg/hr
68.8
187.9
94.1
302.5
246.7
190.8
204.4
113.9
88.5
82.9
66.4
45.4
44.1
0.0
31.5
17.1
18.6
26.9
23.7
28.0
26.1
24.1
17.4
20.7
21.0
24.7
19.1
15.3
18.2
18.1
14.3
16.5
15.6
15.3
15.9
14.9
NH3-N
mg/l
0.09
0.12
0.13
0.12
0.12
0.11
0.11
0.10
0.10
0.10
0.09
0.08
0.08
0.03
0.02
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.01
0.01
0.01
0.02
0.01
0.01
0.02
0.01
0.02
0.01
0.03
0.03
0.03
NHj-N
kg/hr
0.08
0.13
0.16
0.16
0.17
0.16
0.15
0.13
0.13
0.12
0.11
0.09
0.09
0.03
0.02
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.01
0.01
0.01
0.02
0.01
0.01
0.02
0.01
0.02
0.01
0.03
0.03
0.03
NO"3-N
mg/l
0.54
0.49
0.47
0.49
0.51
0.57
0.57
0.57
0.58
0.60
0.57
0.56
0.54
0.53
0.53
0.52
0.52
0.50
0.50
0.49
0.49
0.48
0.47
0.47
0.47
0.47
0.47
0.46
0.46
0.46
0.46
0.46
0.46
0.47
0.47
0.47
NO" 3-N
kg/hr
0.50
0.53
0.57
0.66
0.72
0.81
0.78
0.74
0.73
0.73
0.67
0.64
0.61
0.59
0.58
0.57
0.56
0.53
0.52
0.51
0.50
0.49
0.47
0.46
0.46
0.45
0.45
0.43
0.43
0.43
0.42
0.42
0.41
0.42
0.42
0.41
o-PO4asP
mg/l
0.06
0.08
0.10
0.09
0.10
0.07
0.07
0.06
0.05
0.05
0.05
0.04
0.05
0.03
0.03
0.03
0.02
0.03
0.02
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
68
-------
Table 4. Loading Rates for SFBR20 Rain Event on March 12,2002
YSIData
Date/Time
3/12/200222:00
3/12/200223:00
3/13/2002
3/13/20021:00
3/13/20022:00
3/13/20023:00
3/13/20024:00
3/13/20025:00
3/13/20026:00
3/13/20027:00
3/13/20028:00
3/13/20029:00
3/13/200210:00
3/13/200211:00
3/13/200212:00
3/13/200213:00
3/13/200214:00
3/13/200215:00
3/13/200216:00
3/13/200217:00
3/13/200218:00
3/13/200219:00
3/13/200220:00
3/13/200221:00
3/13/200222:00
3/13/200223:00
3/14/2002
3/14/20021:00
3/14/20022:00
3/14/20023:00
3/14/20024:00
3/14/20025:00
3/14/20026:00
3/14/20027:00
3/14/20028:00
3/14/20029:00
3/14/200210:00
Level
(ft)
0.98
1.114
1.231
1.301
1.344
1.266
1.186
1.161
1.115
1.067
1.03
1.001
0.98
0.96
0.942
0.929
0.915
0.904
0.894
0.883
0.873
0.865
0.857
0.85
0.844
0.838
0.831
0.827
0.822
0.818
0.821
0.817
0.808
0.808
0.803
0.797
0.792
Stage
(ft)
21.004
21.138
21.255
21.325
21.368
21.290
21.210
21.185
21.139
21.091
21.054
21.025
21.004
20.984
20.966
20.953
20.939
20.928
20.918
20.907
20.897
20.889
20.881
20.874
20.868
20.862
20.855
20.851
20.845
20.842
20.845
20.841
20.832
20.832
20.827
20.821
20.816
Flew
(cfs)
0.85
2.49
4.59
5.88
7.08
5.21
3.63
3.12
2.49
1.80
1.34
1.03
0.85
0.57
0.46
0.29
0.22
0.17
0.12
0.09
0.06
0.04
0.03
0.02
0.02
0.01
0.01
ISCO Data from SFBR20
Sample*
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
36
37
Sample ID
2lsB031 20221 57
2lsB031 2022257
2lsB031 2022357
2lsB031 3020057
2lsB031 30201 57
2lsB031 3020257
2lsB031 3020357
2lsB031 3020457
2lsB031 3020557
2lsB031 3020657
2lsB031 3020757
2lsB031 3020857
2lsB031 3020957
2lsB031 3021 057
2lsB031 3021 157
2lsB031 3021 257
2lsB031 3021 357
2lsB031 3021 457
2lsB031 3021 557
2lsB031 3021 657
2lsB031 3021 757
2lsB031 3021 857
2lsB031 3021 957
2lsB031 3022057
2lsB031 30221 57
2lsB031 3022257
2lsB031 3022357
2lsB031 4020057
2lsB031 40201 57
2lsB031 4020257
2lsB031 4020357
2lsB031 4020457
2lsB031 4020557
2lsB031 4020657
2lsB031 4020757
2lsB031 4020857
2lsB031 4020957
Date
3/12/2002
3/12/2002
3/12/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
Time
(24hr)
2157
2257
2357
0057
0157
0257
0357
0457
0557
0657
0757
0857
0957
1057
1157
1257
1357
1457
1557
1657
1757
1857
1957
2057
2157
2257
2357
0057
0157
0257
0357
0457
0557
0657
0757
0857
0957
TSS
mg/l
53.0
90.0
75.4
69.0
59.0
66.2
no data
no data
no data
no data
37.8
30.6
25.2
24.8
13.0
10.2
15.2
15.8
9.4
10.0
11.0
11.8
15.2
1.6
11.0
9.2
9.2
10.4
10.8
6.0
8.4
9.2
9.2
7.8
9.2
6.4
6.8
TSS
kg/hr
4.6
22.9
35.3
41.4
42.6
35.2
no data
no data
no data
no data
5.2
3.2
2.2
1.4
0.6
0.3
0.3
0.3
0.1
0.1
0.1
0.1
0.0
0.0
0.0
0.0
0.0
NH3-N
mg/l
0.08
0.09
0.09
0.08
0.08
0.15
0.26
0.34
0.34
0.30
0.25
0.21
0.19
0.16
0.05
0.04
0.03
0.02
0.02
0.00
0.01
0.01
0.00
0.01
0.02
0.02
0.02
0.02
0.04
0.03
0.04
0.05
0.03
0.04
0.03
0.03
0.02
NHj-N
kg/hr
0.01
0.02
0.04
0.05
0.06
0.08
0.10
0.11
0.09
0.06
0.03
0.02
0.02
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
NO'3-N
mg/l
0.32
0.33
0.32
0.30
0.33
0.40
0.56
0.69
0.71
0.65
0.61
0.58
0.56
0.55
0.57
0.59
0.58
0.59
0.59
0.61
0.61
0.59
0.60
0.60
0.59
0.59
0.57
0.58
0.57
0.56
0.55
0.55
0.54
0.54
0.54
0.53
0.53
NO" 3-N
kg/hr
0.03
0.08
0.15
0.18
0.24
0.21
0.21
0.22
0.18
0.12
0.08
0.06
0.05
0.03
0.03
0.02
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
o-PO4asP
mg/l
0.03
0.03
0.03
0.02
0.03
0.06
0.01
0.19
0.18
0.15
0.11
0.11
0.08
0.06
0.04
0.04
0.04
0.04
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.04
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
69
-------
Table 5. Loading Rates for SFBR30 Rain Event on March 12, 2002
YSI Data
Date/Time
3/12/200221:15
3/12/200222:15
3/12/200223:15
3/13/20020:15
3/13/20021:15
3/13/20022:15
3/13/20023:15
3/13/20024:15
3/13/20025:15
3/13/20026:15
3/13/20027:15
3/13/20028:15
3/13/20029:15
3/13/2002 10:15
3/13/200211:15
3/13/2002 12:15
3/13/2002 13:15
3/13/2002 14:15
3/13/2002 15:15
3/13/2002 16:15
3/13/200217:15
3/13/2002 18:15
3/13/2002 19:15
3/13/200220:15
3/13/200221:15
3/13/200222:15
3/13/200223:15
3/14/20020:15
3/14/20021:15
3/14/20022:15
3/14/20023:15
3/14/20024:15
3/14/20025:15
3/14/20026:15
3/14/20027:15
3/14/20028:15
3/14/20029:15
3/14/2002 10:15
3/14/200211:15
Level
(ft)
2.878
3.001
3.117
3.326
3.477
3.693
3.76
3.811
3.841
3.826
3.786
3.748
3.726
3.715
3.704
3.689
3.679
3.667
3.649
3.626
3.605
3.585
3.56
3.541
3.522
3.501
3.48
3.46
3.44
3.422
3.405
3.389
3.374
3.355
3.342
3.325
3.31
3.295
3.284
Stage
(ft)
34.814
34.937
35.053
35.262
35.413
35.629
35.696
35.747
35.777
35.762
35.722
35.684
35.662
35.651
35.640
35.625
35.615
35.603
35.585
35.562
35.541
35.521
35.496
35.477
35.458
35.437
35.416
35.396
35.376
35.358
35.341
35.325
35.310
35.291
35.278
35.261
35.246
35.231
35.220
Flow
(cfs)
53.9
64.0
73.8
95.6
112.9
141.0
150.6
157.7
162.0
159.1
153.5
147.9
145.1
143.8
142.4
139.7
139.7
137.0
147.9
131.7
129.2
126.6
124.0
121.5
119.0
116.6
114.1
111.7
109.3
107.0
104.7
103.5
101.2
98.9
97.8
95.6
94.5
92.3
91.2
ISCO Data from
Sample*
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
36
37
38
39
Sample ID
3lsA031 20221 22
3lsA031 2022222
3lsA031 2022322
3lsA031 3020022
3lsA031 30201 22
3lsA031 3020222
3lsA031 3020322
3lsA031 3020422
3lsA031 3020522
3lsA031 3020622
3lsA031 3020722
3lsA031 3020822
3lsA031 3020922
3lsA031 3021022
3lsA031 3021 122
3lsA031 3021222
3lsA031 3021322
3lsA031 3021422
3lsA031 3021522
3lsA031 3021622
3lsA031 3021 722
3lsA031 3021822
3lsA031 3021922
3lsA031 3022022
3lsA031 30221 22
3lsA031 3022222
3lsA031 3022322
3lsA031 4020022
3lsA031 40201 22
3lsA031 4020222
3lsA031 4020322
3lsA031 4020422
3lsA031 4020522
3lsA031 4020622
3lsA031 4020722
3lsA031 4020822
3lsA031 4020922
3lsA031 4021022
3lsA031 4021 122
Date
3/12/2002
3/12/2002
3/12/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
Time
(24hr)
2122
2222
2322
022
0122
0222
0322
0422
0522
0622
0722
0822
0922
1022
1122
1222
1322
1422
1522
1622
1722
1822
1922
2022
2122
2222
2322
0022
0122
0222
0322
0422
0522
0622
0722
0822
0922
1022
1122
TSS
mg/l
16.0
37.8
58.0
78.4
72.6
84.8
131.6
151.2
108.6
76.2
65.8
60.2
58.8
51.4
47.8
80.8
41.0
42.2
27.4
36.0
34.2
34.2
30.8
29.8
30.6
26.8
25.4
22.0
23.6
13.6
22.4
20.8
20.6
26.6
15.6
19.6
18.1
18.4
16.4
TSS
kg/hr
88.0
246.8
436.6
764.5
836.0
1219.6
2021.5
2432.1
1794.5
1236.6
1030.2
908.2
870.3
753.9
694.3
1151.4
584.2
589.7
413.4
483.6
450.7
441.6
389.6
369.3
371.4
318.7
295.6
250.7
263.1
148.4
239.2
219.6
212.6
268.3
155.6
191.1
174.8
173.2
152.6
NH3-N
mg/l
0.04
0.04
0.05
0.05
0.04
0.05
0.06
0.06
0.06
0.05
0.06
0.05
0.05
0.15
0.06
0.06
0.00
0.00
0.00
0.00
0.01
0.00
0.01
0.01
0.00
0.00
0.00
0.01
0.00
0.01
0.01
0.01
0.02
0.01
0.01
0.01
0.01
0.00
0.00
NH3-N
kg/hr
0.2
0.3
0.4
0.5
0.5
0.7
0.9
1.0
1.0
0.8
0.9
0.8
0.7
2.2
0.9
0.9
0.0
0.0
0.0
0.0
0.1
0.0
0.1
0.1
0.0
0.0
0.0
0.1
0.0
0.1
0.1
0.1
0.2
0.1
0.1
0.1
0.1
0.0
0.0
NO'3-N
mg/l
0.35
0.42
0.35
0.33
0.36
0.37
0.37
0.34
0.37
0.37
0.35
0.33
0.33
0.35
0.35
0.37
0.38
0.38
0.38
0.38
0.39
0.39
0.39
0.39
0.38
0.38
0.38
0.38
0.37
0.38
0.37
0.38
0.38
0.37
0.37
0.38
0.37
0.38
0.37
NO" 3-N
kg/hr
1.9
2.7
2.6
3.2
4.1
5.3
5.7
5.5
6.1
6.0
5.5
5.0
4.9
5.1
5.1
5.3
5.4
5.3
5.7
5.1
5.1
5.0
4.9
4.8
4.6
4.5
4.4
4.3
4.1
4.1
4.0
4.0
3.9
3.7
3.7
3.7
3.6
3.6
3.4
o-PO4asP
mg/l
0.02
0.02
0.02
no data
0.02
0.03
0.03
0.03
0.03
0.02
0.02
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
70
-------
Table 6. Loading Rates for SFBR40 Rain Event on March 12, 2002
YSI Data
Date/Time
3/12/200221:30
3/12/200222:30
3/12/200223:30
3/13/20020:30
3/13/20021:30
3/13/20022:30
3/13/20023:30
3/13/20024:30
3/13/20025:30
3/13/20026:30
3/13/20027:30
3/13/20028:30
3/13/20029:30
3/13/200210:30
3/13/200211:30
3/13/200212:30
3/13/200213:30
3/13/200214:30
3/13/200215:30
3/13/200216:30
3/13/200217:30
3/13/200218:30
3/13/200219:30
3/13/200220:30
3/13/200221:30
3/13/200222:30
3/13/200223:30
3/14/20020:30
3/14/20021:30
3/14/20022:30
3/14/20023:30
3/14/20024:30
3/14/20025:30
3/14/20026:30
3/14/20027:30
3/14/20028:30
3/14/20029:30
3/14/200210:30
Level
(ft)
2.029
2.211
2.394
2.377
2.431
2.435
2.448
2.426
2.416
2.424
2.435
2.448
2.46
2.469
2.475
2.475
2.47
2.462
2.454
2.443
2.431
2.418
2.405
2.39
2.376
2.364
2.347
2.335
2.322
2.31
2.298
2.285
2.272
2.259
2.248
2.237
2.227
2.215
Stage
(ft)
11.455
11.637
11.820
11.803
11.857
11.861
11.874
11.852
11.842
11.850
11.861
11.874
11.886
11.895
11.901
11.901
11.896
11.888
11.880
11.869
11.857
11.844
11.831
11.816
11.802
11.790
11.773
11.761
11.748
11.736
11.724
11.711
11.698
11.685
11.674
11.663
11.653
11.641
Flow
(cfs)
29.6
41.1
54.4
52.9
57.7
57.7
58.5
56.9
56.1
56.9
57.7
58.5
60.2
61.0
61.0
61.0
61.0
60.2
59.3
58.5
57.7
56.1
55.2
54.4
52.9
52.1
50.5
49.8
49.0
48.3
46.8
46.0
45.3
43.9
43.2
42.5
41.8
41.1
ISCO Data from
Sample*
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
36
37
38
Sample ID
4lsA031 20221 32
4lsA031 2022232
4lsA031 2022332
4lsA031 3020033
4lsA031 30201 33
4lsA031 3020233
4lsA031 3020333
4lsA031 3020433
4lsA031 3020533
4lsA031 3020633
4lsA031 3020733
4lsA031 3020833
4lsA031 3020933
4lsA031 3021 033
4lsA031 3021 133
4lsA031 3021 233
4lsA031 3021 333
4lsA031 3021 433
4lsA031 3021 533
4lsA031 3021 633
4lsA031 3021 733
4lsA031 3021 833
4lsA031 3021 933
4lsA031 3022033
4lsA031 30221 33
4lsA031 3022233
4lsA031 3022333
4lsA031 4020033
4lsA031 40201 33
4lsA031 4020233
4lsA031 4020333
4lsA031 4020433
4lsA031 4020533
4lsA031 4020633
4lsA031 4020733
4lsA031 4020833
4lsA031 4020933
4lsA031 4021 033
Date
3/12/2002
3/12/2002
3/12/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
Time
(24hr)
2132
2232
2332
0033
0133
0233
0333
0433
0533
0633
0733
0833
0933
1033
1133
1233
1333
1433
1533
1633
1733
1833
1933
2033
2133
2233
2333
0033
0133
0233
0333
0433
0533
0633
0733
0833
0933
1033
TSS
mg/l
693.0
154.8
135.6
153.2
330.8
129.4
68.8
52.8
43.4
41.2
36.8
36.6
32.0
31.2
17.2
22.2
13.8
10.0
18.6
18.6
8.2
14.2
15.6
7.0
11.6
13.8
14.6
14.4
14.2
12.0
9.8
14.6
9.4
13.6
8.8
9.6
5.8
7.6
TSS
kg/hr
2092.3
649.0
752.4
826.6
1946.9
761.6
410.5
306.4
248.3
239.1
216.6
218.4
196.5
194.1
107.0
138.1
85.9
61.4
112.5
111.0
48.3
81.3
87.8
38.8
62.6
73.3
75.2
73.1
71.0
59.1
46.8
68.5
43.4
60.9
38.8
41.6
24.7
31.9
NH3-N
mg/l
0.05
0.04
0.04
0.05
0.05
0.05
0.04
0.05
0.04
0.04
0.04
0.04
0.04
0.03
0.02
0.01
0.02
0.01
0.02
0.01
0.01
0.01
0.02
0.01
0.00
0.02
0.02
0.02
0.03
0.03
no data
0.02
0.00
0.02
0.03
0.03
0.03
0.03
NH3-N
kg/hr
0.2
0.2
0.2
0.3
0.3
0.3
0.2
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.0
0.1
0.1
0.1
0.1
0.1
no data
0.1
0.0
0.1
0.1
0.1
0.1
0.1
NQ-3-N
mg/l
0.05
0.04
0.04
0.05
0.05
0.05
0.04
0.05
0.04
0.04
0.04
0.04
0.04
0.03
0.02
0.01
0.02
0.01
0.02
0.01
0.01
0.01
0.02
0.01
0.00
0.02
0.02
0.02
0.03
0.03
no data
0.02
0.00
0.02
0.03
0.03
0.03
0.03
NO" 3-N
kg/hr
0.2
0.2
0.2
0.3
0.3
0.3
0.2
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.0
0.1
0.1
0.1
0.1
0.1
no data
0.1
0.0
0.1
0.1
0.1
0.1
0.1
o-PO4asP
mg/l
0.04
0.02
0.02
0.03
0.04
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
no data
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.02
0.01
0.01
0.01
0.01
0.01
71
-------
Table 7. Loading Rates for SFBR60 Rain Event on March 12, 2002
YSI Data
Date/Time
3/12/200222:15
3/12/200223:15
3/13/20020:15
3/13/20021:15
3/13/20022:15
3/13/20023:15
3/13/20024:15
3/13/20025:15
3/13/20026:15
3/13/20027:15
3/13/20028:15
3/13/20029:15
3/13/200210:15
3/13/200211:15
3/13/200212:15
3/13/200213:15
3/13/200214:15
3/13/200215:15
3/13/200216:15
3/13/200217:15
3/13/200218:15
3/13/200219:15
3/13/200220:15
3/13/200221:15
3/13/200222:15
3/13/200223:15
3/14/20020:15
3/14/20021:15
3/14/20022:15
3/14/20023:15
3/14/20024:15
3/14/20025:15
3/14/20026:15
3/14/20027:15
3/14/20028:15
3/14/20029:15
3/14/200210:15
3/14/200211:15
Level
(ft)
2.502
2.518
2.542
2.598
2.73
3.073
3.447
3.666
3.775
3.814
3.817
3.788
3.742
3.694
3.633
3.571
3.521
3.472
3.425
3.384
3.344
3.31
3.276
3.256
3.234
3.203
3.177
3.147
3.116
3.092
3.065
3.041
3.018
2.995
2.971
2.951
2.929
2.913
Stage
(ft)
33.620
33.636
33.660
33.716
33.848
34.191
34.565
34.784
34.893
34.932
34.935
34.906
34.860
34.812
34.751
34.689
34.639
34.590
34.543
34.502
34.462
34.428
34.394
34.374
34.352
34.321
34.295
34.265
34.234
34.210
34.183
34.159
34.136
34.113
34.089
34.069
34.047
34.031
Flow
(cfs)
35.3
36.0
36.7
38.9
48.2
58.8
76.9
88.6
94.7
97.0
97.5
95.8
93.0
90.2
86.9
83.7
81.1
78.4
75.9
73.8
71.8
70.3
68.3
67.3
66.4
64.9
64.0
62.1
60.7
59.7
58.3
57.4
56.5
55.2
54.3
53.4
52.5
51.6
ISCO Data from
Sample*
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
36
37
38
Sample ID
6lsA031 2022220
6lsA031 2022320
6lsA031 3020020
6lsA031 30201 20
6lsA031 3020220
6lsA031 3020320
6lsA031 3020420
6lsA031 3020520
6lsA031 3020620
6lsA031 3020720
6lsA031 3020820
6lsA031 3020920
6lsA031 3021 020
6lsA031 3021 120
6lsA031 3021 220
6lsA031 3021 320
6lsA031 3021 420
6lsA031 3021 520
6lsA031 3021 620
6lsA031 3021 720
6lsA031 3021 820
6lsA031 3021 920
6lsA031 3022020
6lsA031 30221 20
6lsA031 3022220
6lsA031 3022320
6lsA031 4020020
6lsA031 40201 20
6lsA031 4020220
6lsA031 4020320
6lsA031 4020420
6lsA031 4020520
6lsA031 4020620
6lsA031 4020720
6lsA031 4020820
6lsA031 4020920
6lsA031 4021 020
6lsA031 4021 120
Date
3/12/2002
3/12/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/13/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
3/14/2002
Time
(24hr)
2220
2320
0020
0120
0220
0320
0420
0520
0620
0720
0820
0920
1020
1120
1220
1320
1420
1520
1620
1720
1820
1920
2020
2120
2220
2320
0020
0120
0220
0320
0420
0520
0620
0720
0820
0920
1020
1120
TSS
mg/l
34.8
31.0
32.6
43.6
80.0
176.0
318.4
484.6
401.0
247.2
265.8
154.6
225.8
116.0
87.6
136.4
127.0
101.0
120.6
111.8
36.4
25.8
84.0
105.8
50.6
47.4
71.4
57.6
45.4
64.0
47.6
52.2
54.8
62.2
44.0
47.4
42.8
42.6
TSS
kg/hr
125.3
113.8
122.0
173.0
393.3
1055.6
2497.5
4379.4
3873.4
2445.8
2643.4
1510.7
2141.9
1067.2
776.5
1164.5
1050.6
807.7
933.7
841.6
266.6
185.0
585.2
726.3
342.7
313.8
466.1
364.8
281.1
389.7
283.1
305.6
315.8
350.2
243.7
258.2
229.2
224.2
NH3-N
mg/l
0.03
0.03
0.03
0.04
0.03
0.03
0.04
0.07
0.06
0.07
0.08
0.09
0.08
0.07
0.07
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
0.00
0.00
0.03
0.02
0.00
0.01
0.02
0.01
0.03
0.02
0.03
0.04
NH3-N
kg/hr
0.1
0.1
0.1
0.2
0.1
0.2
0.3
0.6
0.6
0.7
0.8
0.9
0.8
0.6
0.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.2
0.1
0.0
0.1
0.1
0.1
0.2
0.1
0.2
0.2
NQ-3-N
mg/l
0.34
0.38
0.33
0.35
0.36
0.32
0.37
0.44
0.45
0.47
0.49
0.52
0.53
0.51
0.51
0.51
0.51
0.50
0.49
0.49
0.48
0.48
0.47
0.47
0.47
0.46
0.45
0.44
0.43
0.44
0.43
0.43
0.42
0.42
0.41
0.41
0.40
0.40
NO" 3-N
kg/hr
1.2
1.4
1.2
1.4
1.8
1.9
2.9
4.0
4.3
4.7
4.9
5.1
5.0
4.7
4.5
4.4
4.2
4.0
3.8
3.7
3.5
3.4
3.3
3.2
3.2
3.0
2.9
2.8
2.7
2.7
2.6
2.5
2.4
2.4
2.3
2.2
2.1
2.1
o-PO4asP
mg/l
0.01
0.01
0.04
0.05
0.02
0.01
0.02
0.03
0.03
0.03
0.04
0.04
no data
0.04
0.06
0.04
0.04
0.03
0.03
0.04
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
72
-------
REFERENCES
Broad River Soil Conservation District and County Government of Madison County, 1961.
Watershed Work Plan South River Watershed, Clarke and Madison Counties, Georgia, 38 pp.
EPA. 2002. The Twenty Needs Report: How Research Can Improve the TMDL Program. EPA
841-B-02-002. Washington, DC.
EPA. 2002. Region 4 Analytical Support Branch SOP and QA Manual. SESD, EPA, Region 4,
Athens, GA.
EPA. 2002. Region 4 Ecological Assessment Branch SOP and QA Manual. SESD, EPA, Region
4, Athens, GA.
EPA. 2002. Region 4 Enforcement and Investigation Branch SOP and QA Manual. SESD, EPA,
Region 4, Athens, GA.
Georgia Department of Environmental Protection. 2001. Savannah River Basin Plan. Atlanta,
GA.
Jones, K. B. et al. 2001. Predicting nutrient and sediment loadings to streams from landscape
metrics: A multiple watershed study from the United States Mid-Atlantic Region. Landscape
Ecology 16: 301-312.
Meyer, J. L. and J. B. Wallace. 2001. Lost linkages and lotic ecology: rediscovering small
streams. In: Press, M. C. et al. (Eds.) Ecology: Achievement and Challenge, pp. 295-317,
Blackwell Science.
Mulholland, P. J. and D. R. Lenat. 1992. Streams of the Southeastern Piedmont, Atlantic
Drainage. In Hackney, C. T. et al. (Eds.) Biodiversity of the Southeastern United States, pp.
193-231, John Wiley and Sons, New York.
Omernik, J. M. 1977. Nonpoint source - stream nutrient level relationships: A nationwide study.
EPA-600/3-77-105, Corvallis, OR, 151 pp.
Peterson, B. J. et al. 2001. Control of nitrogen export from watersheds by headwater streams.
Science 292: 86-90.
Raschke, R. L., Bruce Pruitt, Tom Cavinder, Hoke Howard, David Melgaard, David Hill and
Nancy Bethune. 1998. Development of TMDL Protocols Using Selected Watersheds in the
Lower Piedmont Ecoregion of Georgia. EPA Region 4 Internal Report.
Raschke, R. L., H.S. Howard, R. J. Lewis, R. L. Quinn, and B. L. Berrang. 1999. Savannah River
Basin REMAP: A Demonstration of the Usefulness of Probability Sampling for the Purpose of
Estimating Ecological Condition in State Monitoring Programs. EPA/904-R-99-002.
73
-------
Sierra Club et al., 1996. Plantiffs, v. USEPA; Carol M. Browner, Administrator, USEPA, & John
H. Hankinson Jr., Regional Administrator, USEPA Region 4, defendants, civil action file # 1:
94-CV-2501-MHS.
Smith, Charles N., William R. Payne, Jr., John D. Pope, Heinz Kollig, Paul D. Smith, Donald
Brockway, Brenda E. Kitchens and David Spidle. 2002. A Stream Water Quality Study in the
Savannah River Basin. In Draft.
Standard Methods for the Examination of Water and Wastewater, 20th Edition 1998,
pp. 2-54 to 2-58.
The Georgia County Guide, 2002, 21st edition. Edited by Susan R. Boatright and Douglas C.
Bachtel, Center for Agribusiness and Economic Development in the College of Agricultural and
Environmental Sciences, ISSN# 1044-0976, University of Georgia.
USDA-SCS. 1968. Soil Survey of Clarke and Oconee Counties, Georgia.
USDA-SCS, 1979. Soil Survey of Elbert, Franklin, and Madison Counties, Georgia.
USDA-SCS, 1991. Soil Survey of Oglethorpe County, Georgia.
74
-------
Bridge Sampling SFBR10
EAB-001.0
Rev.#: 0
Date: September 08, 2003
Pagel of 2
Standard Operating Procedures (SOP's) for Bridge Sampling
PURPOSE
To collect storm event stream samples for developing sampling and modeling protocols for
determining sediment and nutrient TMDL's.
1. Unlock security gate.
2. Back the crane truck to the center of the bridge near ISCO sampler.
3. For night sampling, start the generator on truck to power the flood lights and prepare for
sampling with the DH-59 depth integrated sampler initially followed by the bedload
sampling.
4. Sampling with the DH-59 and Bedload sampler shall be conducted at a time interval of 2
hours.
A telephone is located on the bridge rail near the ISCO for sampling team use. The telephone
number: 789-2903.
Sampling with the DH-59:
Sampling with the DH-59 depth integrated sampler at this site is conducted by hand with
rope attached to sampler. Sample at the same locations as marked on the downstream side of the
bridge. This sampler weighs about 25 pounds. Field personnel shall collect samples on the
downstream side of the bridge, not to interfere with the intake of the ISCO sampler and YSI
multi-probe located on the upstream side of the bridge.
Follow the procedures outlined below to use the manually operated depth-integrated
sampler with rope:
1. Secure a rope to sampler.
2. Install the intake nozzle in the sampler. Start with the large (1/4 inch) nozzle. There are five
nozzles for use, two of the 1/4 inch, two of the 3/16 inch, and one 1/8 inch diameter bore,
threaded for assembly into the sampler head.
3. Place a 1-pint milk bottle into the sampler by pulling the end handle to open, make sure
gasket is in position for the sample bottle opening.
4. Move to the marked sampling site on the bridge from the right bank.
5. Lower the DH-59 sampler into the water until the rear fins are just in the water, allow
sampler to orient itself with current flow, lower unit into water and start counting (1001-
1030 depending on the depth of water), bubbles will appear on the water surface, lower the
sampler at a steady speed until the rear of the sampler hits the stream bottom and raise the
sampler upwards to the surface of the water to allow the milk bottle to fill, and retrieve the
75
-------
Bridge Sampling SFBR10
EAB-001.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 2
sampler. Repeat this process, if necessary, until the collected sample is to the required
volume between 300 and 400 ml. DO NOT COMPOSITE SAMPLES.
6. Record the sampling time, date, and cross stream location on sample bottles and in field log
book. See numbering SOP.
7. Empty the milk bottle by swirling the sample and transfer to labeled sample bottles.
8. Store sample on ice in ice chest.
9. Move to next stream cross section, and continue to collect samples at each location for that
time interval.
Sampling with crane truck using bedload sampler
1. Back crane truck to the center of the third sampling site (stream cross section locations)
marked on bridge from right-edge-of-water (looking downstream) for bedload sampling.
2. Prepare crane for sampling by obtaining the electrically operated remote control for
winch behind the seat of the Dodge flatbed truck, 4x4. Plug the remote cord into the side
of the crane support stand. Run the winch in reverse to loosen the cable to provide slack
for extending the arm to the maximum length and to raise the arm to the 3rd hole so that
the winch arm will extend over the bridge top rail. Lift the bedload sampler and lower to
ground and stand the sampler in an upright position at the rear of the truck.
3. Install sampling bag on sampler by placing the rubber ring into groove, tighten stainless
steel bar clamp with wing nuts. Secure other end of bag to chain.
4. Raise the sampler, loosen the lock handle to allow the crane arm to swing over the top
bridge rail.
5. Lower the sampler at the first sampling point (marked on bridge with orange paint) until
the rear fins are just in the water, allow sampler to orient itself with current flow and
lower unit to stream bottom carefully. Do not drop the sampler too fast to cause bottom
disturbance.
6. Sampling time is estimated at 10 minutes for each stream cross section sampling interval
(one composite sample per sampling time). Use stop watch to determine the 10 minute
sampling interval. The sampler shall be deployed at three to five pre-determined
(depending on stream width) equal intervals across the stream section.
7. Retrieve the sampler, lower the sampler to an upright position. Fill a garden sprayer with
tap water and wash the sediment attached to the inside of the sampler and bag side walls
to concentrate the sediment into the lower corner of the nitex bag.
8. Remove the nitex bag from the sampler.
9. Using a 5-gallon bucket of tap water (about 3- gallons) immerse the sample bag several
times to concentrate the sediment into the bottom of the bag for transfer to labeled
sample bottle using a large stem funnel.
10. Record the sampling time, date, and cross stream location on sample bottles and in field
log book. See numbering SOP.
76
-------
Balances (use of)
EAB-002.0
Rev.#: 0
Date: Sept. 08, 2003
Pagel of 2
INTRODUCTION
The balance is the most important single piece of apparatus available to the chemist and is used
to determine the mass (weight) of objects. Because the mass (weight) to be determined ranges
from kilograms to micrograms, the choice of the balance depends on the total mass (weight) to
be determined and the sensitivity required.
PURPOSE
Use of the balance is part of most chemical laboratory procedures. The weight obtained from the
weighing process is used in the computation of the respective final result.
GENERAL
The following rules should be observed in caring for and using a balance:
1. Level the balance.
2. Make sure the balance is working properly. Use calibrated and undamaged weights to check.
3. Check the balance zero.
4. Keep the balance clean.
5. Handle all weights and objects with forceps, never with fingers. Place the weight at the
center of the pan.
6. Avoid weighing unconditioned objects, hot or cold.
7. Do not overload the balance.
8. Do not place moist objects or chemicals directly on the balance pan.
9. Close the balance case and wait for constant weight.
10. Record weight in notebook for addition. Never add mentally.
ERRORS IN DETERMINING WEIGHTS
1. Changes in moisture or carbon dioxide content. Such materials must be weighed in a closed
system.
2. Volatility of sample. Such materials must be weighed in a closed system.
3. Electrification. A charged object is attracted to various parts of the balance. An antistatic
brush might help.
4. Temperature. Convection currents of a warm object causes the pan to be buoyed up which
makes the object weigh less. An object should be weighed at room temperature.
APPLICATION
All balances at the Field Research Annex are electronic balances. There is the single-pan
analytical balance which has a maximum weight capacity of 200 grams and weighs to four
places (O.OOOOg); there is the single-pan top-load balance which has a maximum weight capacity
77
-------
Balances (use of)
EAB-002.0
Rev.#: 0
Date: Sept. 08, 2003
Page 2 of 2
of 220 grams and weighs to three places (O.OOOg); there is the single-pan top-load balance which
has a maximum weight capacity of 310 grams and weighs to three places (O.OOOg); and there is
the single-pan top-load balance which has two ranges with maximum weight capacities of 16 kg
and 3.2 kg and weighs to zero and one places respectively (Og/O.Og).
The operation of each balance is simple:
1. Zero the balance (and open the door).
2. Place the object on the pan (and close the door).
3. Wait for constant weight and take digital readout.
Most of the balances can be coupled directly to computers or recording devices if necessary.
Each individual procedure, whether it is for Total Suspended Solids, for Bedload samples, or for
Nutrients, etc., defines which one of the balances is appropriate depending on the maximum
weight and the sensitivity required.
78
-------
Nutrient Analysis
EAB-003.0
Rev. #: 0
Date: Sept. 08, 2003
Page 1 of 4
PURPOSE
The data of the nutrient analyses will be applied to the modeling of Total Maximum Daily Loads
(TMDL).
SUMMARY OF PROCEDURES
ortho-phosphate:
The automated procedure for the determination of ortho- phosphate is based on the colorimetric
method in which a blue color is formed by the reaction of ortho- phosphate, molybdate ion and
antimony ion followed by reduction with ascorbic acid at an acidic pH. The reduced blue
phosphomolybdenum complex is read at 660 nm.
see Bran+Luebbe AutoAnalyzer method No. US-696B-82
Nitrate:
The automated procedure for the determination of nitrate utilizes the reaction whereby nitrate is
reduced to nitrite by an alkaline solution of hydrazine sulfate containing a copper catalyst. The
stream is then treated with sulfanilamide under acidic conditions to form a soluble dye which is
measured colorimetrically. The final product measured represents the nitrite ion originally
present plus that formed from the nitrate.
see Bran+Luebbe AutoAnalyzer method No. US-696F-82W
Total phosphorus:
The automated procedure for the determination of total phosphorus takes place in three stages.
First, the sample is mixed with persulphate and irradiated in a UV digestor. In this digestion step
organic bound phosphorus is released. Second, polyphosphates are converted to ortho-phosphate
by acid hydrolysis at 95°C. Third, the ortho-phosphate is determined by reaction with
molybdate, antimony and ascorbic acid, producing a phospho-molybdenum blue complex which
is measured colorimetrically at 660 or 880 nm.
see Bran+Luebbe AutoAnalyzer method No. G-092-93 Rev.l, multitest MT 17
Ammonia:
The automated procedure for the determination of ammonia utilizes the Berthelot Reaction, in
which the formation of a blue colored compound believed to be closely related to indophenol
occurs when the solution of an ammonium salt is added to sodium phenoxide, followed by the
addition of sodium hypochlorite. A solution of EDTA is added to the sample stream to eliminate
79
-------
Nutrient Analysis
EAB-003.0
Rev. #: 0
Date: Sept. 08, 2003
Page 2 of 4
the precipitation of the hydroxides of calcium and magnesium. Sodium nitroprusside is added to
intensify the blue color.
see Bran+Luebbe AutoAnalyzer method No. US-696D-82X
Total nitrogen:
Inorganic and organic nitrogen are oxidized to nitrate by sulfate radicals. Sulfate radicals are
produced by the photolytic decomposition of persulfate in an on-line UV digestor. The nitrate is
reduced to nitrite and then determined using the sulfanilamide/ NEDD reaction with detection at
520 nm.
See Bran+Luebbe AutoAnalyzer method No. G-157-96 Rev. 3, multitest MT17
INTERFERENCES
Phosphorus:
As much as 50 mg Ferric ion/L, 10 mg Copper/L, and 10 mg Silica/L can be tolerated. High
silica concentrations cause positive interference.
Nitrate
Chloride, sulfide, ferric ion and phosphate ion interfere.
Ammonia:
Calcium and magnesium ions in sea water concentrations will precipitate if EDTA is not used.
INSTRUMENT
The Bran+Luebbe AutoAnalyzer 3 is a modern wet-chemistry analyzer that is used in industrial
laboratories for automation of complex chemical reactions. Most types of aqueous samples can
be analyzed, such as water, solid extracts, beverages, or chemicals.
It uses the principle of air-segmented continuous-flow analysis (CFA) for fully automatic sample
analyses. Samples are mixed with reagents in a continuously flowing stream. The individual
sample segments are kept separate by means of air bubbles.
With more than 700 chemical methods, the AutoAnalyzer 3 combines the advantages of proven
methodology with state-of-the-art technology.
Due to its modular design, the AutoAnalyzer 3 can be easily adapted to the specific requirements
of a laboratory. The main system components include Sample, Pump, Chemistry Module and
Digital Colorimeter.
80
-------
Nutrient Analysis
EAB-003.0
Rev. #: 0
Date: Sept. 08, 2003
Page 3 of 4
System control is provided by the Bran+Luebbe AACE software. AACE stands for Automated
Analyzer Control and Evaluation Software and is designed to control all Bran+Luebbe
AutoAnalyzer 3 systems. It also provides comprehensive and easy-to-use tools to process and
evaluate the data collected from the analyzer. Designed as control software for Bran+Luebbe
Continuous Flow analyzers, AACE provides all the necessary tools for communication with the
AutoAnalyzer modules, easy run programming, and data processing and retrieval.
AACE is characterized by the following features:
• Windows 95 operating system
• Easy analyzer control
• Real-time information
Post-run reporting
Customer-specific quality control
• Network connection and LIMS
• Fast and reliable data storage
The AutoAnalyzer 3 modules are completely controlled by the AACE software. The
programming of a run is divided into five menus. The File menu, the Configure menu, the Set
Up menu, the Run menu, and the Retrieve menu. Each one is programmed to satisfy the
requirements of a run. See Bran+Luebbe Operation Manuals for the AutoAnalyzer 3 and for the
AACE software.
METHODOLOGY
The present setup consists of two parameters running simultaneously. First, ortho-phosphate and
nitrate, followed by total phosphorus and ammonia. Total nitrogen is run by itself. All five
parameters are mixed in one stock standard with equal concentrations. This set up only requires
the changing of reagents between runs.
The Set Up programming requires a run to start with the highest standard to adjust the
concentration range from zero to 100%, followed by five standards to establish the calibration
curve. Baseline and lamp intensity are set automatically. If the baseline is too noisy, a run will
not start. A high standard and a zero standard are inserted periodically for auto corrections.
Once a run is finished, the system shuts down automatically and prints a complete report of
peaks, concentrations and calibration curves including the correlation coefficients.
81
-------
Nutrient Analysis
EAB-003.0
Rev. #: 0
Date: Sept. 08, 2003
Page 4 of 4
METHOD CALIBRATION
Method calibration is discussed under Methodology.
Quality Control
Quality control measures should include:
a. Analysis of a duplicate sample every 10 to 15 samples if sample volume is available.
b. Analysis of a standard every 10 to 15 samples.
Samples for duplicate analysis will be marked with red tape. If QA samples are not
identified, check with log in personnel. The duplicate sample for QA shall be labeled
with a Q in back of the label. The following is an example label:
2KB0125011215Q
c. Calibration of the analytical balance once a year and weekly checks of the accuracy of the
balance by the analyst.
SAMPLE COLLECTION
Sample Collection is discussed in a separate SOP.
SAMPLE HANDLING AND PRESERVATION
When samples are collected, they should be put on ice, delivered to the laboratory as soon as
possible and stored under refrigeration until analysis. The time between collection and analyses
should be less than 48 hours. Preservation of the sample with acid should be avoided to
minimize the handling and preserve the integrity of the sample.
SAMPLE PREPARATION FOR ANALYSIS
The refrigerated sample is poured into a 15- mL labeled polypropylene centrifuge tube and
centrifuged at 4000 rpm for 20 minutes. It is then poured into a 5-mL AutoAnalyzer sample cup
and inserted into the sample tray according to the programmed Set Up.
TROUBLESHOOTING
Troubleshooting sections are included in both manuals, the AutoAnalyzer 3 manual and the
AACE manual.
PERSONNEL QUALIFICATIONS
The AutoAnalyzer 3 is a highly sophisticated instrument and requires knowledge of chemistry
and computer literacy. It can be operated by experienced technicians who have gone through a
hands-on training period until they are thoroughly familiar with the many different aspects of
mechanical, hydraulic, and computer programming handling of the system.
82
-------
Suspended Solids Analysis
EAB-004.0
Rev.#: 0
Date: September 08, 2003
Page 1 of 4
INTRODUCTION
Solids refers to matter suspended or dissolved in water or wastewater. Total suspended solids is
the portion of total solids retained by a filter. Dissolved solids is the portion of solids that passes
through a filter of 2.0 jim (or smaller) nominal pore size under specified conditions (1). Because
flows and concentrations in streams are usually unsteady, samples represent conditions only at
the time and location of sample collection. This method implies sand concentrations of less than
about 10,000 mg/L and clay concentrations of less than 200 mg/L. If the concentration in the
sample is greater than these limits, a factor must be applied to convert the concentration from
mg/L to mg/kg. The density of the sample will be larger than 1.000 g/mL at the higher
concentration and must be considered in the final calculation. A specific gravity of 2.65 g/mL is
generally assumed for sediment.
PURPOSE
The data of the total suspended solids (TSS) analysis will be applied to the modeling of Total
Maximum Daily Loads (TMDL).
SUMMARY OF PROCEDURE
A well-mixed sample is filtered through a weighed standard glass-fiber filter and the residue
retained on the filter is dried to a constant weight at 103 to 105°C. The increase in weight of the
filter represents the total suspended solids.
INTERFERENCES
Exclude large floating particles or submerged agglomerates of nonhomogeneous materials from
the sample if it is determined that their inclusion is not representative.
Because excessive residue on the filter may form a water-entrapping crust, limit the sample size
to that yielding no more than 200 mg residue.
EQUIPMENT
2-L filter suction flask with sidearm.
Coors porcelain filter funnel with 90-mm diameter perforated plate.
90-mm diameter glass-fiber filter disk #934AH (Whatman).
Forceps for handling glass-fiber filter.
90-mm diameter watch glass.
Filter trap.
One liter graduated cylinder.
Vacuum pump developing a vacuum of 27-28 inches of mercury.
Drying oven, for operation at 103 to 105°C.
Desiccator, provided with a desiccant containing a color indicator of moisture concentration.
Analytical balance, capable of weighing to 0.1 mg.
83
-------
Suspended Solids Analysis
EAB-004.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 4
METHODOLOGY
Preparation of glass-fiber filter disk:
Because of the huge number of samples to be analyzed, a large number of glass-fiber filter disks
should be washed and stored in a desiccator before the sampling event.
Insert glass-fiber filter disk with wrinkled side up in filtration apparatus using forceps. Apply
vacuum and wash disk with three successive 20-50 mL portions of deionized (DI) water after
wetting disk with a small volume of DI water to seat it. Continue suction to remove all traces of
water, turn vacuum off, and discard washings. Transfer disk to a watch glass and dry in oven at
103-105°C for one hour. Cool in desiccator to condition to temperature for one hour.
Conditioned disks can be stacked in desiccator for future use.
Sample analysis:
Transfer washed disk to a watch glass using forceps and keep in desiccator for 30-60 minutes to
condition if disks were prepared and stacked in desiccator. Weigh watch glass with disk.
Transfer disk to filtering apparatus using forceps, turn pump on, and wet disk with a small
volume of DI water to seat it. Shake the sample bottle for several seconds to suspend all the
solids and immediately pour all the sample into a one liter graduated cylinder. Record the
sample volume. Pour about 50 mL of DI water into the sample bottle, shake well and
immediately pour into the one liter grad. cylinder. Repeat this washing two more times or until
all the solids have been transferred to the one liter grad. cylinder. Filter the sample and washings.
Partially invert the one liter grad. cylinder over the filter funnel and rinse all the remaining solids
with a wash bottle into the filter funnel. Wash disk with three successive 20-50 mL portions of
DI water, allowing complete drainage between washings. Continue suction to remove all traces
of water. Disconnect pump quick-disconnect. Remove disk and transfer to its watch glass (watch
glasses must be numbered for easy identification). Dry in oven at 103-105°C for 1.5 hours.
Transfer watch glass with disk to desiccator to balance temperature for 1.5 hours and weigh.
As filtering proceeds, inspect the filtrate. If it is turbid, pour the filtrate back through the filter a
second and possibly a third time. If the filtrate is still turbid, the filter may be leaking, in which
case a new filter must be used and the process repeated. If the filtrate is transparent but
discolored, a natural dye is present and refiltration is not necessary.
Calculation: report results in mg to one decimal point
mgTSS/L = (A-E} x 1QQQ
Sample volume, mL
84
-------
Suspended Solids Analysis
EAB-004.0
Rev.#: 0
Date: September 08, 2003
Page 3 of 4
where: A = weight of watch glass and disk + dried residue
B = weight of watch glass and disk empty, mg
The yield should be between 2.5 and 200 mg dried residue.
REFERENCES
1. Standard Methods for the Examination of Water and Wastewater, 20th Edition 1998, pp. 2-54
to 2-58.
2. Ecosystems Research Division Laboratory Research Notebook #979 of Heinz P. Kollig,
pp. 7 to 8.
MODIFICATION TO THE REFERENCE METHODOLOGY
The reference methodology calls for repeating the cycle of drying, cooling, desiccating, and
weighing until a constant weight is obtained or until the weight change is less than 4% of the
previous weight or 0.5 mg, whichever is less. This repetition of the cycle was eliminated because
of the huge number of samples to be analyzed. The time of drying and desiccating, however, was
increased from 1 hour to 1.5 hours. In addition, a study was performed to explore the weight
change of the disk in repeating the cycle (2). 24 disks were washed, dried, desiccated, and
weighed twice. The mean weight change was 0.2 mg and the 95% confidence interval was 0.2 ±
0.06.
QUALITY CONTROL
Analysis of a duplicate sample is not possible because all of the sample has been used for the
analysis. Collection of more (another) sample is impractical because it would represent a
different sample.
Calibration of the analytical balance once a year and weekly checks of the accuracy of the
balance by the analyst.
SAMPLE COLLECTION
Sample collection is discussed in a separate SOP.
SAMPLE HANDLING AND PRESERVATION
When samples are collected, they should be put on ice, delivered to the laboratory as soon as
possible and stored under refrigeration until analysis. Begin analysis as soon as possible because
of the impracticality of preserving the sample. The time between collection and analysis should
be less than 48 hours. In no case should samples be held for more than 7 days.
85
-------
Suspended Solids Analysis
EAB-004.0
Rev.#: 0
Date: September 08, 2003
Page 4 of 4
SAMPLE PREPARATION FOR ANALYSIS
The refrigerated sample should be brought up close to room temperature before filtering.
SOURCES OF ERROR AND VARIABILITY
Sample not mixed adequately.
Torn or holes in fiber-glass filter disk.
Improper vacuum.
Wrong oven temperature.
Desiccant exhausted.
Balance malfunctioning.
Mistake in calculation.
PERSONNEL QUALIFICATIONS
The analysis of total suspended solids is a fairly straight forward process. The analyst should
have knowledge of chemistry and be experienced with general laboratory operations. A short
hands-on introduction and training period should be required.
86
-------
Cableway SFBR30
EAB-005.0
Rev.#: 0
Date: September 08, 2003
Page 1 of 5
Field Standard Operating Procedures
Cableway System
Highway 172 Site
PURPOSE
To collect storm event stream samples for developing sampling and modeling protocols
for determining sediment and nutrient TMDL's.
When the sampling team arrives at the site, unlock and open the fence gate underneath
the cableway platform. This site has direct utility power from Georgia Power. On the right side
of the door is a light switch that will turn on flood lights underneath, lights on side of the
building platform at access ramp and light above the main entrance doors. Inside the building on
the right side near door there are two light switches, the first switch will turn on the overhead
light (look for rodents and snakes), the second switch will turn on the big sports light underneath
building platform to light up river view for cableway sampling at night. Once sampling is
complete, reverse the steps for turning off lights.
Sampling with the DH-59 and Bedload sampler shall be conducted at a time interval of 2 hours.
A telephone is located on the inside back wall of the cableway house for sampling team use. The
telephone number 788-3046.
Operating the cableway system
1. Open the control panel along the back wall of the cableway, turn on, push start button, and
the cableway is ready for operation. The speed is controlled by the rheostat in the control
box. Start with the speed control at mid range and increase or decrease as desired.
2. Controls for operating the cableway system for positioning the samplers are as follows:
a. toggle switch located on the support post arm:
i. reverse position takes the sampler out and lowers the sampler (down/out).
ii. forward position raises the sampler and brings the sampler into the platform (up/in).
b. lever on the winch assembly shifts the double drum from one drum to two drum
operation:
i. up position raises and lowers the sampler, direction being controlled by the toggle
switch located on the support post arm. Forward position of the toggle raises the
Cableway sampler and reverse lowers the sampler.
ii. down position controls the transverse motion of the sampler, direction also being
87
-------
Cableway SFBR30
EAB-005.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 5
controlled by the toggle switch. Forward position of the toggle brings the sampler in
and reverse takes the sampler out.
iii. speeds of the directions are controlled by the rheostat on the control box.
iv. The winch is equipped with an automatic Weston brake system that will allow the
sampler to be held in any position even with the loss of generator power. A loud
clicking sound will be heard as the sampler is raised. Do not adjust the brake
control knobs.
v. It is very important at the beginning of each shift for the mechanism of the Weston
brake system to be cleaned. Place Kimwipes under the mechanism and then apply a
liberal amount of WD-40 to the brake while the winch system is in motion. At the
termination of the sampling event, run the winch and tighten the cables so that the
sampler attachment (on end of cable) is pulled to the traveler block such that the
small cables have minimum slack over the river and the traveler is within the
confines of the fence. Do not overtighten the cables.
The winch has two counters to provide the distance that the sampler has traveled. The
counter on the right side facing the river registers in feet and the counter on the left side registers
in tenth's of feet. These are the counters to be used in positioning the samplers at selected
distances from the platform for sampling.
Cableway Calibration Check
DH-59
Zero the cableway using the DH-59 sampler. Attach the DH-59 to the cableway using the
shackle pins and clips. Suspend the sampler using the lifting mode of the mechanism and allow it
to travel to the edge of the platform. Lower the sampler to a point slightly below the platform
then bring the vertical cable back even with the platform. Once this position has been achieved,
bring the sampler up to the shackle mechanism on the traveler cable. This is point "0" and should
be indicated on both counters located on opposite sides of the winch system. If not at "0",
manually set both counters to "0" at this time.
BEDLOAD
Calibrate the cableway for the bedload sampler by lowering the sampler to the edge of the
platform floor and position it in the center of the 6-inch sample intake. Once this position has
been achieved, bring the sampler up to the shackle mechanism on the traveler cable. This is point
"0" and should be indicated on both counters.
88
-------
Cableway SFBR30
EAB-005.0
Rev.#: 0
Date: September 08, 2003
Page 3 of 5
Sediment Sampling with Cableway
Depth Integrated Sampling using DH-59
The DH-59 depth-integrating sampler accumulates a water-sediment mixture throughout the
depth of the water column. If the 1-pint bottle becomes completely filled during a sampling
operation, it is not representative and must be discarded. The time it takes to lower and raise the
sampler in the water column shall be sufficient to produce a sample volume between 300 and
400 ml.
1. Install the intake nozzle in the sampler. Start with the large (1/4 inch) nozzle. Check each
nozzle and sampler intake hole to make sure it is free of debris and not clogged. There are
five nozzles for use, two of the 1/4 inch, two of the 3/16 inch, and one 1/8 inch diameter
bore, threaded for assembly into the sampler head.
2. Place a 1-pint bottle into the sampler by pulling the end handle to open, make sure gasket is
in position for the sample bottle opening.
3. Beginning on the far-side (right side facing downstream) of the stream from the platform,
sampling points are designated A through E and are located at counter readings of 79, 71, 63,
55 and 47 foot, respectively (8-foot increments between sites with 5-foot from each side of
the stream bank). In the event of exceedingly high water levels above bank full, two
additional sample sites are located at 94 and 14 feet when the water depth at these sites is 6-
inches or greater. The sampling site locations are provided on a chart located on the wall of
the cableway platform. Beginning at the79 foot interval, position the sampler consecutively
at each of these locations on the cableway.
4. After positioning the sampler at the designated locations, allow the sampler to descend to
the bottom of the river under controlled speed. First lower the sampler until the rear fins are
just in the water, allow the sampler to orient itself with current flow and then lower the unit
to the
stream bottom. As the unit decends into the water, start counting from 1001- 1030
depending on the depth of water. Bubbles will appear on the water surface indicating that
the sample is entering the bottle. Lower the sampler at a steady speed until the rear of the
sampler hits the stream bottom, (look for a slight upward deflection of the traveler) then raise
the sampler at the same steady speed upwards to obtain the sample. Retrieve the sampler and
check the volume collected. If necessary, repeat this process until the collected volume is
between 300 and 400 ml. DO NOT COMPOSITE THE SAMPLES. The DH-59 are
individual samples of each stream cross section.
5. Record the site number, sampling time, date, and cross stream location on sample bottles and
in field log book See numbering SOP.
6. Empty the sample bottle by swirling the contents and then transfer to the labeled sample
bottles.
89
-------
Cableway SFBR30
EAB-005.0
Rev.#: 0
Date: September 08, 2003
Page 4 of 5
7. Store samples on ice in ice chest.
8. Move to next stream cross section, and continue to collect samples at each location for that
time interval.
Helley Smith Bedload Sampler
Bedload sampling is collecting sediment that is being transported along the stream bottom
during a storm event. The sampler weighs 170 pounds, consists of a 6-inch opening and a nitex
bag for the collection of sediment.
Attach the bedload sampler to the cableway and raise to an upright position so that it is
standing on its fins.
1. Install nitex sampling bag on the sampler by placing the rubber ring into groove with the two
pleats at bottom of sampler for alignment, tighten stainless steel bar clamp with wing nuts.
Secure other end of bag to chain.
2. Beginning at the 79 foot interval, lower the sampler to the bottom of the river until the rear
fins are just in the water. Allow the sampler to orient itself with current flow and lower to
stream bottom, being careful not to disturb the stream bottom by dropping the sampler too
fast. When the sampler makes contact with the stream bottom there is a slight upward
deflection of the traveler. Allow the sampler to remain at that stream cross sectional location
for 10 minute (if the water depth is 6-inches or greater). Use stop watch to determine the 10
minute sampling interval. After the 10 minutes is complete, move to the next cross section by
raising the sampler above the water and determining the distance to travel to the next
location.
3. Retrieve the sampler and lower it to an upright position on the platform. Keep the cable
secured to the sampler. Use the garden sprayer filled with tap water to wash the sediment
from the inside surface of the sampler and the side walls of the bag so that the sediment is
concentrated in the lower corner of the nitex bag.
4. Remove the nitex bag from the sampler.
5. Using a 5-gallon bucket with about 3-gallons of tap water, immerse the sample bag several
times to clean the bag and to concentrate the sample in the bottom of the bag for ease of
transfer to sample bottles.
6. Transfer the bedload sediment sample into a one-liter widemouth Nalgene bottle by using a
large stem funnel.
7. Label the sample bottle and record in the field log book the sampling time, date, and cross
section locations that were sampled. See numbering SOP.
8. Place the composite sample on ice for transport to the laboratory.
9. Reattach the bag to the sampler for the next sampling time interval.
90
-------
Cableway SFBR30
EAB-005.0
Rev.#: 0
Date: September 08, 2003
Page 5 of 5
10. Remove the bedload sampler from the cableway and attach the DH-59 sampler for the next
sampling time interval.
91
-------
Clouds Creek SFBR60
EAB-006.0
Rev.#: 0
Date: September 08, 2003
Page 1 of 5
Standard Operating Procedures
Cableway System
Clouds Creek Site, SFBR60
Located in Watson Mill Bridge State Park
PURPOSE
To collect storm event stream samples for developing sampling and modeling protocols for
determining sediment, nutrient and pathogen TMDL's.
When the sampling team arrives at the primitive camp area road inside the park, unlock
the EPA lock at gate for entry. Keep the park gate closed at all times. Drive the vehicle to an
area with trash cans at top of hill and stop. Unload the 4x4 mule from trailer and transport
supplies and sampling team to the site. Secure vehicle and trailer doors. Unlock the gate and
doors to the cableway platform and generator door. Check oil level in the generator. Open the
propane gas valve to the generator. Start the generator by switching to run and close door. On
the right side of the back entrance gate is a light switch that will turn on flood lights on side of
the building platform, inside and at the generator. Inside the building on the back wall is an
electrical raceway. Turn on the big sports light (mounted on roof) that provides light across the
river at the electrical raceway for night time sampling.
Once sampling is complete, reverse the steps for securing the sampling site, turning off lights,
then open the door of generator and turn off and lock and lock all gates. Be sure to turn off
propane gas valve.
Located on the inside back wall of the cableway house is a telephone for sampling team use. The
telephone number 783-2201.
Operating the cableway system
1. Open the control panel along the back wall of the cableway, turn on, push start button, and
the cableway is ready for operation. The speed is controlled by the rheostat in the control
box. Start with the speed control at mid range and increase or decrease as desired.
2. Controls for operating the cableway system for positioning the samplers are as follows:
a. toggle switch located on the support post arm:
i. reverse position takes the sampler out and lowers the sampler (down/out).
ii. forward position raises the sampler and brings the sampler into the platform (up/in).
b. lever on the winch assembly shifts the double drum from one drum to two drum
operation:
i. up position raises and lowers the sampler, direction being controlled by the toggle
92
-------
Clouds Creek SFBR60
EAB-006.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 5
ii. switch located on the support post arm. Forward position of the toggle raises the
sampler and reverse lowers the sampler.
iii. down position controls the transverse motion of the sampler, direction also being
controlled by the toggle switch. Forward position of the toggle brings the sampler in
and reverse takes the sampler out.
iv. speeds of the directions are controlled by the rheostat on the control box.
v. The winch is equipped with an automatic Weston brake system that will allow the
sampler to be held in any position even with the loss of generator power. A loud
clicking sound will be heard as the sampler is raised. Do not adjust the brake
control knobs.
vi. It is very important at the beginning of each shift for the mechanism of the Weston
brake system to be cleaned. Place Kimwipes under the mechanism and then apply a
liberal amount of WD-40 to the brake while the winch system is in motion. At the
termination of the sampling event, run the winch and tighten the cables so that the
sampler attachment (on end of cable) is pulled to the traveler block such that the
small cables have minimum slack over the river and the traveler is within the
confines of the fence. Do not overtighten the cables.
The winch has two counters to provide the distance that the sampler has traveled. The
counter on the right side facing the river registers in feet and the counter on the left side registers
in tenth's of feet. These are the counters to be used in positioning the samplers at selected
distances from the platform for sampling.
Cableway Calibration Check
DH-59
Zero the cableway using the DH-59 sampler. Attach the DH-59 to the cableway using the
shackle pins and clips. Suspend the sampler using the lifting mode of the mechanism and allow it
to travel to the edge of the platform. Lower the sampler to a point slightly below the platform
then bring the vertical cable back even with the platform. Once this position has been achieved,
bring the sampler up to the shackle mechanism on the traveler cable. This is point "0" and should
be indicated on both counters located on opposite sides of the winch system. If not at "0",
manually set both counters to "0" at this time.
BEDLOAD
Calibrate the cableway for the bedload sampler by lowering the sampler to the edge of the
platform floor and position it in the center of the 6-inch sample intake. Once this position has
been achieved, bring the sampler up to the shackle mechanism on the traveler cable. This is point
"0" and should be indicated on both counters.
93
-------
Clouds Creek SFBR60
EAB-006.0
Rev.#: 0
Date: September 08, 2003
Page 3 of 5
Sediment Sampling with Cableway
Depth Integrated Sampling using DH-59
The DH-59 depth-integrating sampler accumulates a water-sediment mixture throughout the
depth of the water column. If the 1-pint bottle becomes completely filled during a sampling
operation, it is not representative and must be discarded. The time it takes to lower and raise the
sampler in the water column shall be sufficient to produce a sample volume between 300 and
400 ml.
1. Install the intake nozzle in the sampler. Start with the large (1/4 inch) nozzle. Check each
nozzle and sampler intake hole to make sure it is free of debris and not clogged. There are
five nozzles for use, two of the 1/4 inch, two of the 3/16 inch, and one 1/8 inch diameter
bore, threaded for assembly into the sampler head.
2. Place a 1-pint bottle into the sampler by pulling the end handle to open, make sure gasket is
in position for the sample bottle opening.
3. Beginning on the far-side (right side facing downstream) of the stream from the platform,
sampling points are designated A through E and are located at counter readings of 88, 76, 64,
52 and 40 foot, respectively (12-foot increments between sites with 5-foot from each side of
the stream bank). In the event of exceedingly high water levels above bank full, three
additional sample sites are located at 100, 28 and 16 feet designated as AA, El and E2,
respectively. Sample only when the water depth at these sites is 6-inches or greater. The
sampling site locations are provided on a chart located on the wall of the cableway platform.
Beginning at the79 foot interval, position the sampler consecutively at each of these
locations on the cableway.
4. After positioning the sampler at the designated locations, allow the sampler to descend to
the bottom of the river under controlled speed. First lower the sampler until the rear fins are
just in the water, allow the sampler to orient itself with current flow and then lower the unit
to the stream bottom. As the unit decends into the water, start counting from 1001- 1030
depending on the depth of water. Bubbles will appear on the water surface indicating that
the sample is entering the bottle. Lower the sampler at a steady speed until the rear of the
sampler hits the stream bottom, (look for a slight upward deflection of the traveler) then raise
the sampler at the same steady speed upwards to obtain the sample. Retrieve the sampler and
check the volume collected. If necessary, repeat this process until the collected volume is
between 300 and 400 ml. DO NOT COMPOSITE THE SAMPLES. The DH-59 are
individual samples of each stream cross section.
5. Record the sampling time, date, and cross stream location on sample bottles and in field log
book See numbering SOP.
94
-------
Clouds Creek SFBR60
EAB-006.0
Rev.#: 0
Date: September 08, 2003
Page 4 of 5
6. Empty the sample bottle by swirling the contents and then transfer to the labeled sample
bottles.
7. Store samples on ice in ice chest.
8. Move to next stream cross section, and continue to collect samples at each location for that
time interval.
Helley Smith Bedload Sampler
Bedload sampling is collecting sediment that is being transported along the stream bottom
during a storm event. The sampler weighs 170 pounds, consists of a 6-inch opening and a nitex
bag for the collection of sediment.
Attach the bedload sampler to the cableway and raise to an upright position so that it is
standing on its fins.
1. Install nitex sampling bag on the sampler by placing the rubber ring into groove with the two
pleats at bottom of sampler for alignment, tighten stainless steel bar clamp with wing nuts.
Secure other end of bag to chain.
2. Beginning at the 88 foot interval, lower the sampler to the bottom of the river until the rear
fins are just in the water. Allow the sampler to orient itself with current flow and lower to
stream bottom, being careful not to disturb the stream bottom by dropping the sampler too
fast. When the sampler makes contact with the stream bottom there is a slight upward
deflection of the traveler. Allow the sampler to remain at that stream cross sectional location
for 10 minute (if the water depth is 6-inches or greater). Use stop watch to determine the 10
minute sampling interval. After the 10 minutes is complete, move to the next cross section by
raising the sampler above the water and determining the distance to travel to the next
location.
3. Retrieve the sampler and lower it to an upright position on the platform. Keep the cable
secured to the sampler. Use the garden sprayer filled with tap water to wash the sediment
from the inside surface of the sampler and the side walls of the bag so that the sediment is
concentrated in the lower corner of the nitex bag.
4. Remove the nitex bag from the sampler.
5. Using a 5-gallon bucket with about 3-gallons of tap water, immerse the sample bag several
times to clean the bag and to concentrate the sample in the bottom of the bag for ease of
transfer to sample bottles.
6. Transfer the bedload sediment sample into a one-liter widemouth Nalgene bottle by using a
large stem funnel.
7. Label the sample bottle and record in the field log book the sampling time, date, and cross
section locations that were sampled. See numbering SOP.
95
-------
Clouds Creek SFBR60
EAB-006.0
Rev.#: 0
Date: September 08, 2003
Page 5 of 5
8. Place the composite sample on ice for transport to the laboratory.
9. Reattach the bag to the sampler for the next sampling time interval.
10. Remove the bedload sampler from the cableway and attach the DH-59 sampler for the next
sampling time interval.
96
-------
Cableway SFBR70
EAB-007.0
Rev.#: 0
Date: September 08, 2003
Page 1 of 4
Field Standard Operating Procedures
Cableway System
Carlton Site
PURPOSE
To collect storm event stream samples for developing sampling and modeling protocols for
determining sediment and nutrient TMDL's.
When the sampling team arrives at the site, unlock entrance gate, and turn on light switch
(raise lid cover) located on outside wall of building as you start up the access ramp. This switch
will light flood lights going up ramp and back of building. Unlock main entrance doors and turn
on first light switch for the overhead light (look for rodents and snakes). The second switch turns
on the sports light facing the river for cableway sampling at night. This site has two power
sources, one being utility power (Rayle EMC) and the other a Kohler 11KW generator. In the
event of power loss from utility power, the transfer switch will automatically switch to generator
power. It will take about 2 minutes until full power is restored to the light. For emergency
lighting, a battery powered light will come on as soon as power is up from the generator. This
light will turn off automatically. The sports light that is being used for cableway sampling is a
400 watt halide lamp. When power is turned off manually or due to power failure, it will not
come back on for about 10 minutes since it has to cool off before igniting again. Do not cut off
switches thinking something is wrong, be patient and wait. The generator is fueled by propane
gas (100 pound tank) and a second tank will be available for replacement. The propane fuel tank
will remain on at all times. In case the propane gas tank is empty, the generator will try to start
itself 3 times before shutting down. In the event it does not start call Charlie Smith for
directions. The generator system is all automatic and does not require assistance from the
sampling teams IMPORTANT: DO NOT CUT OFF ANY SWITCHES WITH THE
GENERATOR SYSTEM INCLUDING THE TRANSFER SWITCH ( GE Zenith Control
Box) ON BACK WALL. Every two weeks the generator will run automatically 15 minutes for
warmup to keep operational.
Once sampling is complete, reverse the steps for securing the sampling site and turning off
lights. Be sure to disconnect the 9-V battery in the green power control box along back
wall.
Located on the inside right wall there is a telephone for sampling team use. The phone number
is797-3206.
The installation of a new OTT cableway system and support equipment for securing the ISCO
strainers and the YSI multi-probe in the center of a 180 foot wide river was a difficult challenge
97
-------
Cableway SFBR70
EAB-007.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 4
since the stream bed consisted of loose sand (2-feet deep) overlying bedrock. The site was made
operational with the ISCO and YSI multi-probe May 16, 2002.
Operating the cableway system
1. Open the green power control box along the back wall of the cableway and install the 9-V
battery that powers the digital read out on the remote control. Close door.
2. Turn on cableway by the yellow switch key and then press green "ON" button.
3. Connect the DH-59 sampler to cableway for first stream sampling.
4. Remove the cableway remote control from wall and lift the DH-59 sampler from the floor.
5. Zero the cableway by bringing the sampler to the stop switch (located at top of cableway
system) and press the distance measurement to zero the system.
6. Collect 5-samples located at 10 meter intervals apart including 10, 20, 30, 40 and 50. The
edge of the river's right bank facing downstream is 7 meters from zero and the far side bank
facing downstream is 59 meters.
Depth Integrated Sampling using DH-59
The DH-59 depth-integrating sampler accumulates a water-sediment mixture throughout the
depth of the water column. If the 1-pint bottle becomes completely filled during a sampling
operation, it is not representative and must be discarded. The time it takes to lower and raise the
sampler in the water column shall be sufficient to produce a sample volume between 300 and
400 ml.
1. Install the intake nozzle in the sampler. Start with the large (1/4 inch) nozzle. Check each
nozzle and sampler intake hole to make sure it is free of debris and not clogged. There are
five nozzles for use, two of the 1/4 inch, two of the 3/16 inch, and one 1/8 inch diameter
bore, threaded for assembly into the sampler head.
2. Place a 1-pint bottle into the sampler by pulling the end handle to open, make sure gasket is
in position for the sample bottle opening.
3. Beginning on the far-side (left side facing downstream) of the stream from the platform, on a
chart located on the wall of the cableway platform. Beginning at the 50 meter interval,
position the sampler consecutively at each of these locations on the cableway.
4. After positioning the sampler at the designated locations, allow the sampler to descend to
the bottom of the river under controlled speed. First lower the sampler until the rear fins are
just in the water, allow the sampler to orient itself with current flow and then lower the unit
to the stream bottom. As the unit decends into the water, start counting from 1001- 1030
depending on the depth of water. Bubbles will appear on the water surface indicating that
98
-------
Cableway SFBR70
EAB-007.0
Rev.#: 0
Date: September 08, 2003
Page 3 of 4
the sample is entering the bottle. Lower the sampler at a steady speed until the rear of the
sampler hits the stream bottom, (look for a slight upward deflection of the traveler) then raise
the sampler at the same steady speed upwards to obtain the sample. Retrieve the sampler and
check the volume collected. If necessary, repeat this process until the collected volume is
between 300 and 400 ml. DO NOT COMPOSITE THE SAMPLES. The DH-59 are
individual samples of each stream cross section.
5. Record the sampling time, date, and cross stream location on sample bottles and in field log
book See numbering SOP.
6. Empty the sample bottle by swirling the contents and then transfer to the labeled sample
bottles.
7. Store samples on ice in ice chest.
8. Move to next stream cross section, and continue to collect samples at each location for that
time interval.
Helley Smith Bedload Sampler
Bedload sampling is collecting sediment that is being transported along the stream bottom
during a storm event. The sampler weighs 170 pounds, consists of a 6-inch opening and a nitex
bag for the collection of sediment.
Attach the bedload sampler to the cableway and raise to an upright position so that it is
standing on its fins.
1. Install nitex sampling bag on the sampler by placing the rubber ring into groove with the two
pleats at bottom of sampler for alignment, tighten stainless steel bar clamp with wing nuts.
Secure other end of bag to chain.
2. Beginning at the 50 meter interval, lower the sampler to the bottom of the river until the rear
fins are just in the water. Allow the sampler to orient itself with current flow and lower to
stream bottom, being careful not to disturb the stream bottom by dropping the sampler too
fast. When the sampler makes contact with the stream bottom there is a slight upward
deflection of the traveler. Allow the sampler to remain at that stream cross sectional location
for 10 minute (if the water depth is 6-inches or greater). Use stop watch to determine the 10
minute sampling interval. After the 10 minutes is complete, move to the next cross section by
raising the sampler above the water and determining the distance to travel to the next
location.
3. Retrieve the sampler and lower it to an upright position on the platform. Keep the cable
secured to the sampler. Use the garden sprayer filled with tap water to wash the sediment
from the inside surface of the sampler and the side walls of the bag so that the sediment is
concentrated in the lower corner of the nitex bag.
99
-------
Cableway SFBR70
EAB-007.0
Rev.#: 0
Date: September 08, 2003
Page 4 of 4
4. Remove the nitex bag from the sampler.
5. Using a 5-gallon bucket with about 3-gallons of tap water, immerse the sample bag several
times to clean the bag and to concentrate the sample in the bottom of the bag for ease of
transfer to sample bottles.
6. Transfer the bedload sediment sample into a one-liter widemouth Nalgene bottle by using a
large stem funnel.
7. Label the sample bottle and record in the field log book the sampling time, date, and cross
section locations that were sampled. See numbering SOP.
8. Place the composite sample on ice for transport to the laboratory.
9. Reattach the bag to the sampler for the next sampling time interval.
10. Remove the bedload sampler from the cableway and attach the DH-59 sampler for the next
sampling time interval.
100
-------
Sample Numbering
EAB-008.0
Rev.#: 0
Date: September 08, 2003
Page 1 of 4
Standard Operating Procedures
Field Sample Numbering System
PURPOSE
To label storm event stream samples, sediment cores collected from stream cross sections and
monthly grab samples for analysis. Data will be used to develop sampling and modeling
protocols for determining sediment and nutrient TMDL's.
Each sampling team shall have a team leader or designee who will be responsible for making
sure all samples are collected, properly labeled, recorded and taken to the Field Research Annex
(FRA) for analysis.
As Isco samples are logged in at the FRA, approximately, every 10th sample shall be
identified with red tape to indicate a QA sample for duplicate analysis in the laboratory. The QA
designated samples shall have a Q located at the end of the sample ID. For field and laboratory
duplicate samples, a Q shall be located at the end of the sample ID. For all other samples, an X
shall be placed at the end of the sample ID as a place holder.
SUMMARY OF PROCEDURE
Record Keeping:
For record keeping use "rite in the rain" notepads and record the site number, sample type
(bedload, depth integrated, grab), stream cross section location (if appropriate) or Isco sampler
letter (if sample is from one of the Iscos), the date, and time (military).
Label the depth integrated and bedload sample bottles using a Sharpie permanent marking
pen using the following instructions:
Each container shall be labeled with the following numbering system:
Site Number (first number of label, 1 digit):
1 =Ila
2 = Double Branch
3 = Highway 172 (cableway)
4 = Brush Creek
5 = Highway 22 (cableway proposed)
6 = Clouds Creek (cableway)
7 = Carlton (cableway proposed)
101
-------
Sample Numbering
EAB-008.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 4
Sample Type (next three letters of label):
DI = vertically water depth-integrated samples, segregated by cross-section (collected
with the DH48 hand-held sampler or line/cable-delivered DH59 for non-wadeable
streams) with the third place designated as the cross-sectional position (see below).
Bd = bedload (taken with 6-inch bedload samplers (whether taken by wading or withcrane
and largest sampler) will be sampled at appropriate cross-sectional locations
but all cross-sectional sub-samples will be combined into a composite sample
with the third space occupied by the letter X
IsG = baseline grab samples taken with Isco
GRB = baseline grab samples taken by wading into stream
Cross section (third sample type letter):
(designated with depth integrated suspended sediment samples (DI only):
A = first sampling site located on right side facing downstream
B = next interval equally spaced facing downstream
C = next interval equally spaced facing downstream
D = next interval equally spaced facing downstream
E = next interval equally spaced facing downstream
F = next interval equally spaced facing downstream
G = next interval equally spaced facing downstream
H = next interval equally spaced facing downstream
Date (next six numbers of sample label): = month, day, year
Time (next four numbers in the label): = military time
Last letter in the label: = X is a place holder indicating a regular sample for analysis
Q represents a field or laboratory duplicate sample for Quality
Assurance (QA) analysis.
EXAMPLE LABELS USING:
Site Number = 1-7
Sample Type = bedload (BdX)
Depth integrated [DI (sections A-H)]
Stream cross section location = A through H (for depth integrated suspended sediments)
102
-------
Sample Numbering
EAB-008.0
Rev.#: 0
Date: September 08, 2003
Page 3 of 4
Sample Type = Grab, baseline samples only; IsG, or GRB, (if taken manually)
Isco (appropriately labeled from the respective Isco sampler as IsA or IsB)
Date = month, day, year without slashes (020201)
Time = military time in hours and minutes without punctuation (2330)
Q = QA sample analysis
X = regular sample analysis
EXAMPLE LABELING
Background Samples for suspended sediment and nutrients prior to event:
HsG0125010900X
Ila site, grab sample collected using the Isco sampler, on Jan. 25, 2001, at 9:00 AM
HsG0125010900Q
Ila site, grab sample collected using the Isco sampler, on Jan, 25, 2001, at 9:00 AM
(QA sample)
1GRB0202011630X
Ila site, manually acquired grab sample, collected in sample bottle on February 2, 2001, at
4:30 PM
Bedload Samples:
3BdX0125010100X
Hwy 172 site, bedload sediment sample composite of all cross-sections sampled, on Jan. 25,
2001, at 1:00 AM
3BdX0125010100Q
Hwy 172 site, bedload sediment sample composite of all cross-sections sampled, on Jan. 25,
2001, at 1:00 AM (QA sample)
Isco Samples:
2IsB0125011215X
Double Branch site, Isco sampler B, Jan. 25, 2001, at 12:15 PM
2IsB0125011215Q
Double Branch site, Isco sampler B, Jan. 25, 2001, at 12:15 PM (QA sample)
Depth Integrated Samples:
2DIC0125011445X
103
-------
Sample Numbering
EAB-008.0
Rev.#: 0
Date: September 08, 2003
Page 4 of 4
Double Branch site, depth integrated water sample @ cross-section position C, Jan. 25,
2001, at 2:45 PM
2DIC0125011445Q
Double Branch site, depth integrated water sample @ cross-section position C, Jan. 25,
2001, at 2:45 PM (QA sample)
3DIA0607012400X
Highway 172, depth integrated water sample @ cross-section position A, June 7, 2001 @ 12
mid-night
104
-------
Sample Numbering
EAB-009.0
Rev.#: 0
Date: September 08, 2003
Pagel of 2
Standard Operating Procedures
Stream Cross Section Sample Numbering System
PURPOSE
To label stream cross section samples collected for analysis. Data will be used to develop
sampling and modeling protocols for determining sediment TMDL's.
SUMMARY OF PROCEDURE
Record Keeping:
For record keeping use "rite in the rain" notepads and record the site number, stream cross
section location, sample depth, the date, and time (military).
Label the sample bottles using a Sharpie permanent marking pen using the following
instructions:
Each container shall be labeled with the following numbering system:
Storm Event Sampling Site Number (first number of label, 1 digit):
1 =Ila
2 = Double Branch
3 = Highway 172 (cableway)
4 = Brush Creek
5 = Highway 22 (cableway proposed)
6 = Clouds Creek (cableway)
7 = Carlton (cableway proposed)
Continue the numbering system for the remainder of the selected stream cross section sites (total
is about 300)
Stream Cross section (second sample type letter):
A = first sampling site located on right side facing downstream (i.e. right bank)
B = next interval equally spaced facing downstream
C = next interval equally spaced facing downstream
D = next interval equally spaced facing downstream
E = next interval equally spaced facing downstream (i.e. left bank)
Date (next six numbers of sample label): = month, day, year
105
-------
Sample Numbering
EAB-009.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 2
Time Oast four numbers in the label): = military time
EXAMPLE LABELS USING:
Site Number = 1-300
Stream cross section location = A through E
Date = month, day, year without slashes (020201)
Time = military time in hours and minutes without punctuation (2330)
106
-------
Particle Size Analysis
EAB-010.0
Rev.#: 0
Date: September 08, 2003
Page 1 of 5
PURPOSE
The data of the particle size analysis will be applied to the modeling of Total Maximum Daily
Loads (TMDL).
SUMMARY OF PROCEDURE
Particle size analysis measures the size distribution of particles suspended in a liquid or
suspended in air. Particle size analysis at the Environmental Research Division requires analysis
of particles suspended in water only.
About one gram of dispersing agent is added to the soil sample and mixed (stirred) for several
minutes to enhance separation or dispersion of aggregates. The sample is then poured and
washed into a stack of sieves for separation: 8mm, 4mm, and 2mm. Each one of the four size
groups is dried in a drying oven at 103°C to 105°C to get the Dry Weight followed by ignition in
a muffle furnace at 550°C to get the weight of Loss on Ignition (LOT).
Before drying, a subsample of the less than 2-mm size sample is injected into a particle size
analyzer to get the distribution of sand, silt, and clay. An identical size sample is pipetted into an
empty preweighed 50-mL crucible to obtain the equivalent weight injected into the particle size
analyzer.
INTERFERENCES
There should not be any major interferences since each of the phases of separation is simple and
straight forward.
EQUIPMENT
The particle size analysis is divided into two major parts. Part one consists of the manual
separation (sieving) and part two consists of the particle size analysis of the less than 2-mm
fraction using a particle size analyzer.
Part One: Three sieves, 2 inches high and 8 inches in diameter, with sieve openings of 2mm,
4mm, and 8mm. One sieve collection pan.
Pint wash bottle.
Sodium metaphosphate.
Sink with water faucet sprayer.
250-mL porcelain crucibles.
Spoon scraper.
Balance (top loader) weighing to one decimal place in grams.
Drying oven (103°C to 105°C).
Muffle furnace (550°C).
107
-------
Particle Size Analysis
EAB-010.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 5
Part Two: One five-inch diameter sieve with 2-mm sieve openings.
Particle Size Analyzer, Coulter LS200 Series instrument with a Fluid Module and
Printer.
Most of the equipment in Part One.
Three sizes of plastic pipets: ImL, 5mL, and lOmL. Each of the pipets has the tip cut
off.
50-mL porcelain crucibles.
Balance (top loader) weighing to three decimal places in grams.
METHODOLOGY
The sample is submitted to the laboratory usually in a one-quart plastic bottle. Add one gram of
dispersing agent (sodium metaphosphate) and stir for several minutes to enhance separation of
the aggregates. Pour and rinse the bottle clean in a stack of sieves with the 8-mm openings on the
top followed by the 4-mm and 2-mm sieves. All three sieves sit in a sieve receiver pan. Use the
pint wash bottle and the spoon scraper to wash the sample through the 8-mm sieve. Pour the
content of the 8-mm sieve (larger than 8mm) into a preweighed empty and marked 250-mL
porcelain crucible. Do the same with the 4-mm and 2-mm sieves. Pour the content of the
receiver pan (less than 2mm) back into the original sample bottle. Dry the 250-mL crucible plus
content overnight in a drying oven at 103°C to 105°C. Cool in a desiccator til conditioned
(several hours). Reweigh crucible. The weight difference is the Dry Weight. Put the crucible in a
muffle furnace and heat at 550°C for one hour. Cool and condition in a desiccator for several
hours. Reweigh crucible. The loss in weight from the Dry Weight is the Loss on Ignition (LOT).
Calculate the LOT in percent by:
LOT (%) = (LOT weight x 100)/Dry Weight
The less than 2-mm sample should settle over night in the refrigerator before analyzing with the
Coulter LS200. The analyst must estimate which size pipet to use for sampling. If the sample
appears to contain much clay, use the small pipet. If the sample appears to contain much sand,
use the larger pipets. The analyst should be familiar with the general operation of the Coulter
LS200. The following details the operation of the LS200 for the TMDL Project's Bedload and
Core samples:
PARTICLE SIZE ANALYZER OPERATION
COULTER LS200
Turn water on and let air out (both gauges should have 30 psi or less).
The instrument stays on all the time. If it is turned off, it needs to be on for about four hours
before use. The main switch is on the lower left.
108
-------
Particle Size Analysis
EAB-010.0
Rev.#: 0
Date: September 08, 2003
Page 3 of 5
Turn the computer on, click: Cancel.
Click: Use Optical Module at start of screen (this is mandatory).
Click: ok.
Click: Control. Click: Turn pump on.
Many things can be done under Control.
Click: File. Under "Change Directory", click either bedload samples or core samples or create
directory. The selected directory should appear in the lower left corner of the screen.
Click: Preferences and load Preferences.
Click: Run.
Click: Run Cycle.
Click: New Sample.
Click: Start (on run cycle).
The instrument will go through its routine checks. Have the small 2-mm sieve sitting on top of
the water reservoir.
When zero obscuration is shown, add sample to between 8 and 12 % obscuration using the
proper size pipette. Add an equivalent amount of sample into a 50-mL preweighed crucible.
Click: Done.
Type in required information (sample ID, etc.).
Click:ok. Sample will be analyzed and the result printed out.
Shut down:
Click: Control. Click: Turn pump off.
Click: Shut off Optical Module under Run.
Click: File. Click Exit (always exit under File).
Turn water off and bleed pressure off. Turn the computer off.
Dry the 50-mL crucible in the oven at!03°C to 105°C over night. Cool in the desiccator til
conditioned and reweigh.
Pour and wash the less than 2-mm sample from the original sample bottle into a preweighed
250-mL porcelain crucible after having poured off most of the supernatant water. Continue
weighing and heating as explained under Part One.
METHOD CALIBRATION
Generally, the LS200 instrument performs its own calibration before each analysis. A sand
standard is prepared in the laboratory from sand that is sieved through a 500-um sieve followed
by a 250-um sieve. The standard is called Sand Standard 250-500um and is run every 10 to 15
samples. A statistical analysis was done with the Sand Standard as follows:
109
-------
Particle Size Analysis
EAB-010.0
Rev.#: 0
Date: September 08, 2003
Page 4 of 5
n = 15
mean = 482. Sum
standard deviation = 4.92um
3 x standard deviation = 14.Sum
mean = 482.8 +/- 14.Sum
When the Sand Standard falls within the range of the mean, the instrument's performance is
assumed to be acceptable.
QUALITY CONTROL
Quality control measures should include:
a. Analysis of a duplicate sample every 10 to 15 samples if sampling permits.
b. Analysis of a standard every 10 to 15 samples.
Samples for duplicate analysis will be marked with red tape. If QA samples are not
identified, check with log-in personnel. The duplicate sample for QA shall be labeled
with a Q in back of the label. The following is an example label: 2IsB0125011215Q
c. Calibration of the analytical balances once a year and weekly checks of the accuracy of the
balances by the analyst.
Each run's data are stored on the instrument's chem-station. However, back-up copies will be
made to floppy or zip disk at regular intervals.
SAMPLE COLLECTION
Sample collection is discussed in a separate SOP.
SAMPLE HANDLING AND PRESERVATION
When samples are collected they should be put on ice, delivered to the laboratory as soon as
possible and stored in a freezer until analysis if storage requires more than seven days.
Otherwise store under refrigeration. Preservation of the sample with acid should be avoided to
minimize the handling and preserve the integrity of the sample.
SAMPLE PREPARATION FOR ANALYSIS
The preparation of the sample for analysis is discussed under Methodology.
The frozen or refrigerated sample should be brought up close to room temperature before
processing.
110
-------
Particle Size Analysis
EAB-010.0
Rev.#: 0
Date: September 08, 2003
Page 5 of 5
SOURCES OF ERROR AND VARIABILITY
Because of the crude way of processing the sample, there will be a larger variability in the
reproducibility. However, the Sand Standard should not be affected.
PERSONNEL QUALIFICATIONS
The sieving process is a straight forward one and can be learned quickly by a technician. The
operation of the LS200 instrument requires primarily computer literacy. It can be operated by an
experienced technician who has gone through a hands-on training period until he/she is
thoroughly familiar with the many different aspects of the LS200 instrument.
REFERENCES
Coulter Corporation. 1994. Coulter LS Series Product Manual. Coulter Corporation, Miami,
Florida.
Standard Methods for the Examination of Water and Wastewater, 20th Edition 1998, pp. 2-61 to
2-69.
Ill
-------
Total Organic Carbon
EAB-011.0
Rev.#: 0
Date: Sept. 08, 2003
Pagel of 2
PURPOSE
The data of the carbon analyses will be applied to the modeling of Total Maximum Daily Loads
(TMDL).
SUMMARY OF PROCEDURES
The total carbon (TC) combustion tube is filled with oxidation catalyst and heated to 680°C.
The carrier gas is high purity air, TOC grade, and flows through the tube at a rate of 150
mL/min. The component in the sample is decomposed to become CO2 and flows through an
Inorganic Carbon (1C) reaction vessel and through a halogen scrubber into the Non-Dispersive
Infrared gas analyzer (NDIR) where CO2 is detected. The measurement of non-purgeable
organic carbon (NPOC) is performed by pretreating the sample with acid, 2 N hydrochloric, and
sparging with carrier gas before sending sample to the detector.
INTERFERENCES
Carrier gas containing more than Ippm of CO2, CO and HC would interfere in accurate
measurements.
INSTRUMENTATION
The instrument is a Shimadzu Total Organic Carbon Analyzer model TOC-5050A with an ASI-
5000A autosampler. These are controlled by a Windows based TOC Control software.
METHODOLOGY
The previously refrigerated sample is poured into a 15-mL centrifuge tube and centrifuged for 20
minutes at 4000 rpm. About 5 mL of sample is poured into a 10-mL sample vial and the
remainder of the sample is used for other analyses. The sample is automatically acidified with
25 uL of 2 N HC1 to give a pH of about 2 and is then sparged for five minutes with high quality,
TOC grade, compressed air. The instrument automatically injects with a 250-uL syringe the
amount of sample needed for correct peak size and concentration. The measurement is of Non-
purgeable Organic Carbon (NPOC) and is substituted for the TOC/TC method which requires
each sample to be run twice. The NPOC method is approved in TOC-related standard methods
and referred to in water quality-related test methods.
CALIBRATION METHOD
Stock solution of standards are made according to instructions on page 54 in the instruction
manual. Desired concentrations will be made from this standard and new standard curves will
be run before each sampling event. The output signal of the NDIR is linearized for all ranges.
Neither the combustion system nor the reaction system provides factors causing the
concentration-output characteristics to be deviated from linearity. Therefore one or two point
calibration curve is satisfactory for measurement according to the Shimadzu manual. A three
point curve will be used in our laboratory.
112
-------
Total Organic Carbon
EAB-011.0
Rev.#: 0
Date: Sept. 08, 2003
Page 2 of 2
SAMPLE COLLECTION
Sample Collection is discussed in a separate SOP.
SAMPLE HANDLING AND PRESERVATION
When samples are collected, they should be put on ice, delivered to the laboratory as soon as
possible and stored under refrigeration until analysis. The time between collection and analysis
should be less than 48 hours.
TROUBLESHOOTING AND MAINTENANCE
Troubleshooting and maintenance will be performed according to the Instruction Manual and
the TOC Control Software Manual.
PERSONNEL QUALIFICATIONS
The Total Organic Carbon analyzer can be operated by experienced technicians who have gone
through a hands-on training period until they are thoroughly familiar with the different aspects of
the analysis and the computer programming handling of the system and automatic sampler.
113
-------
Fecal Coliform
EAB-041.0
Rev. #0
Date: September 08, 2003
Page 1 of 4
SCOPE AND APPLICATION
The membrane filter technique is a basic procedure used in the detection of coliform
bacteria. The fecal coliform test may be applicable to investigations of drinking water, stream
pollution, raw water sources, wastewater treatment systems, bathing waters, sea waters, and
general water quality monitoring.
SUMMARY OF METHOD
The procedure involves filtering three different volumes of a sample, which is determined by
referring to table 9222:111 of SM 9222 D, plating the filter on m-FC medium, and then
incubating the sample at the desired temperature of 44.5 ± 0.2°C for 24 ± 2 hours. The standard
volume of drinking water to be analyzed is 100 mL. This may be distributed among multiple
membranes.
The best readable plate is one that contains 20-60 BLUE colonies per plate.
INTERFERENCES
Water samples containing colloidal or suspended particulate material can clog the membrane
filter, thereby preventing filtration, or cause spreading bacterial colonies which could interfere
with the identification of target colonies.
Turbidity caused by the prescence of algae may not permit testing of a sample volume
sufficient to yield significant results.
Low coliform estimates may be caused by the presence of high numbers of noncoliforms or
toxic substances.
HEALTH AND SAFETY PROCEDURES
Adherence to laboratory safety procedures described in the SESD Safety, Health and
Environmental Management Program (SHEM) Procedures and Policy Manual, Section 2.5 is
required.
SPECIAL PROCEDURES
Residual chlorine in chlorinated samples should be neutralized with 0.1 mL of 10% sodium
thiosulfate (Na2S2O3).
When less than 10 mL of a sample is used, add approximately 10 mL of sterile dilution water
to the funnel before filtration to aid in dispersing the bacterial suspension over the entire filtering
surface.
ANALYST TRAINING
An Initial Demonstration of Capability/Performance (IDC or IDP) shall be performed prior
to the analysis of any samples, and with a significant change in instrument type, personnel,
matrix, or test method where applicable.
114
-------
Fecal Coliform
EAB-041.0
Rev. #0
Date: September 08, 2003
Page 2 of 4
Analysts must demonstrate the ability to generate acceptable test results with this method by
preparing and analyzing a minimum of four aliquots of a quality control sample either
concurrently or over a period of days.
The capability of the analyst to produce acceptable results will be determined by their ability
to obtain a percent recovery of 80-120% and to obtain a relative standard deviation of <20%.
REAGENTS AND STANDARDS
Dehydrated, and commercially prepared medium may be used.
Laboratory prepared media should be batched tested for performance with positive and
negative culture controls before it is used for analysis and should be prepared in accordance
with SM 9050.
Dilution water should be prepared in accordance with SM9050 B and C.
APPARATUS AND MATERIALS
Culture dishes, pre-sterilized disposable plastic dishes with tight fitting lids.
Filtration units- Wrap the assembly in heavy aluminum foil, sterilize by autoclaving, and
store until use. Alternately, expose all surfaces of the previously cleaned assembly to ultraviolet
radiation (2 minutes) for the initial sanitization before use in the test procedure, or before reusing
units between successive filtration series.
Pre-sterilized and certified membrane filters and absorbent pads.
Incubator, set at 44.5 ± 0.2°C.
Sample and dilution bottles.
Pipets and graduated cylinders.
Smooth flat forceps, without corrugations on the inner sides of the tips. Sterilize before use
by dipping in 95% ethyl alcohol and then pass through a flame.
Apparatus should be maintained in accordance with SM9030.
SAMPLE COLLECTION AND PRESERVATION
Collect samples in clean wide-mouth plastic bottles with non-leaking caps and non-toxic
liners containing sodium thiosulfate.
Hold samples at <10°C during a maximum transport time of 6 hours.
Refrigerate samples upon arrival in the laboratory and be processed within 2 hours of arrival.
When transport conditions necessitate delays in delivery of samples longer than 6 hours,
consider using field laboratory facilities located near the site of collection.
Samples should be collected in accordance with SM 9060 A andB.
115
-------
Fecal Coliform
EAB-041.0
Rev. #0
Date: September 08, 2003
Page 3 of 4
SAMPLE HOLDING TIME
Source samples should be not be held more than 6 hours from time of collection to the time
analyses are initiated.
Drinking water samples should be analyzed within 30 hours of collection.
METHOD CALIBRATION
There are no calibrations associated with this method.
Check the temperatures in the incubator twice daily to insure that it is functioning properly.
Maintain sterility with equipment media and technique.
SAMPLE ANALYSIS AND PROCEDURE
Place a pad in the culture dish and saturate with at least 2.0 mL m-FC medium. Carefully
remove excess medium by decanting the plate.
Place the prepared filter directly on the pad, filter the appropriate volumes of sample, place
the top on the dish, invert the dish and incubate for 24 ± 2 hours at 44.5 ± 0.2°C.
After filtering a series of 10 samples, filter 100 mL of sterile rinse water to check for
possible cross-contamination or contaminated rinse water, and incubate under the same
conditions as the samples.
QUALITY CONTROL
Laboratory Reagent Blanks (LRB) will be performed at a frequency of at least one per batch.
The batch may or may not consist of 20 or more samples which will be analyzed together as a
group.
Laboratory control samples shall be performed at a frequency of one per batch.
At least one sample must be analyzed in duplicate at a frequency of one in 10 samples.
DATA ANALYSIS AND CALCULATION
Direct plating methods such as the membrane filter procedure permit a direct count of
coliform colonies.
The best readable plate is counted for its colonies, the colonies are verified, and the density is
calculated using the count and the volume of sample filtered. The coliform density is reported
conventionally as membrane filter count per 100 mL.
POLLUTION PREVENTION
See SESD Safety, Health and Environmental Management Program (SHEM) Procedures and
Policy Manual, Section 5.8.
116
-------
Fecal Coliform
EAB-041.0
Rev. #0
Date: September 08, 2003
Page 4 of 4
WASTE MANAGEMENT
Waste management and disposal procedures are described in the SESD Safety, Health and
Environmental Management Program (SHEM) Procedures and Policy Manual, Section 2.5.
REFERENCES
American Public Health Association. Standard Methods for the Examination of Water and
Wastewater, 20th Edition, 1998.
Science and Ecosystem Support Division, Region 4, U.S. Environmental Protection Agency.
September, 2000. Ecological Assessment Branch Laboratory Operations and Quality
Assurance Manual.
Science and Ecosystem Support Division, Region 4, U.S. Environmental Protection Agency.
May, 1998. Safety, Health and Environmental Management Program Procedures and Policy
Manual.
117
-------
Escherichia coli
EAB-042.0
Rev.#: 0
Date: September 08, 2003
Page 1 of 4
Procedural Section
Scope and Application
This enzyme substrate test utilizes hydrolyzable substrates for the simultaneous detection of
total coliform bacteria and Escherichia coli enzymes. The MI Broth used in this analysis,
contains a nutritive lactose-based medium containing inhibitors to eliminate the growth of non-
coliform bacteria. This method is recommended for the analysis of drinking and source water
samples, chemical processing and pharmaceutical manufacturing waters.
Summary of Method
The procedure involves filtering three different volumes of a sample, which may be
determined by referring to table 9222:111 of SM 9222 D, plating the filter on a selective and
differential medium, and then incubating the sample at the desired temperature of
35 ± 0.5°C for 22 to 24 hours.
The standard volume of drinking water to be analyzed is 100 milliliters. This may be
distributed among multiple membranes.
Total coliform bacteria present will produce fluorescent colonies upon exposure to longwave
ultraviolet light (366 nm) after primary culturing on MI agar and E. coli will produce blue
colonies under ambient light after primary culturing on MI agar.
Interferences
Water samples containing colloidal or suspended paniculate material can clog the membrane
filter, thereby preventing filtration, or cause spreading bacterial colonies which could interfere
with the identification of target colonies.
Turbidity caused by the prescence of algae may not permit testing of a sample volume
sufficient to yield significant results.
Low coliform estimates may be caused by the presence of high numbers of non coliforms or
toxic substances.
Health and Safety Procedures
Adherence to laboratory safety procedures described in the SESD Safety, Health and
Environmental Management Program (SHEM) Procedures and Policy Manual, Section 2.5
is required.
Special Procedures
Residual chlorine in chlorinated samples should be neutralized with 0.1 mL of 10% sodium
thiosulfate (Na2S2O3).
When less than 10 mL of a sample is used, add approximately 10 mL of sterile dilution water
to the funnel before filtration to aid in dispersing the bacterial suspension over the entire filtering
surface.
118
-------
Escherichia coli
EAB-042.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 4
Analyst Training
An Initial Demonstration of Capability/Performance (IDC or IDP) shall be performed prior
to the analysis of any samples, and with a significant change in instrument type, personnel,
matrix, or test method where applicable.
Analysts must demonstrate the ability to generate acceptable test results with this method by
preparing and analyzing a minimum of four aliquots of a quality control sample either
concurrently or over a period of days.
The capability of the analyst to produce acceptable results will be determined by their ability
to obtain a percent recovery of 80-120% and to obtain a relative standard deviation of <20%.
Reagents and Standards
Commercially prepared media in liquid form (MI Broth) may be used.
Laboratory prepared media should be batched tested for performance with positive and
negative culture controls before it is used for analysis and should be prepared in accordance with
SM9050.
Dilution water should be prepared in accordance with SM 9050 B and C.
Apparatus and Materials
Culture dishes, pre-sterilized disposable plastic dishes with tight fitting lids.
Filtration units- Wrap the assembly in heavy aluminum foil, sterilize by autoclaving, and
store until use. Alternately, expose all surfaces of the previously cleaned assembly to ultraviolet
radiation (2 minutes) for the initial sanitization before use in the test procedure, or before reusing
units between successive filtration series.
Presterilized and certified membrane filters and absorbent pads.
Incubator, set at 35 ± 0.5°C.
Sample and dilution bottles.
Pipets and graduated cylinders.
Smooth flat forceps, without corrugations on the inner sides of the tips. Sterilize before
use by dipping in 95% ethyl alcohol and then pass through a flame.
Apparatus should be maintained in accordance with SM 9030.
Sample Collection and Preservation
Collect samples in clean wide-mouth plastic bottles with non-leaking caps and non-toxic
liners containing sodium thiosulfate
Hold samples at <10°C during a maximum transport time of 6 hours.
Refrigerate samples upon arrival in the laboratory and be processed within 2 hours of arrival.
119
-------
Escherichia coli
EAB-042.0
Rev.#: 0
Date: September 08, 2003
Page 3 of 4
When transport conditions necessitate delays in delivery of samples longer than 6 hours,
consider using field laboratory facilities located near the site of collection.
Samples should be collected in accordance with SM 9060 A andB.
Sample Holding Time
Source samples should not be held more than 6 hours from time of collection to the time
analyses are initiated.
Drinking water samples should be analyzed within 30 hours of collection.
Method Calibration
There are no calibrations associated with this method.
Check the temperatures in the incubator twice daily to insure that it is functioning properly.
Maintain sterility with equipment media and technique.
Sample Analysis and Procedure
Place a pad in the culture dish and saturate with at least 2.0 mL of medium. Carefully
remove excess medium by decanting the plate.
Place the prepared filter directly on the pad, filter the appropriate volumes of sample, place
the top on the dish, invert the dish and incubate for 22 to 24 hours at 35 ± 0.5°C.
After filtering a series of 10 samples, filter 100 mL of sterile rinse water to check for
possible cross-contamination or contaminated rinse water, and incubate under the same
conditions as the samples.
Quality Control
Laboratory Reagent Blanks (LRB) will be performed at a frequency of at least one per batch.
The batch may or may not consist of 20 or more samples which will be analyzed together as a
group.
Laboratory control samples shall be performed at a frequency of one per batch.
At least one sample must be analyzed in duplicate at a frequency of one in 10 samples.
Intralaboratory quality assurance and control should be in accordance with SM9020.
Data Analysis and Calculation
Direct plating methods such as the membrane filter procedure permit a direct count of
coliform colonies.
The best readable plate is counted for its colonies, the colonies are verified, and the density is
calculated using the count and the volume of sample filtered. The coliform density is reported
conventionally as membrane filter count per 100 mL.
120
-------
Escherichia coli
EAB-042.0
Rev.#: 0
Date: September 08, 2003
Page 4 of 4
Pollution Prevention
See SESD Safety, Health and Environmental Management Program (SHEM) Procedures and
Policy Manual, Section 5.8.
Waste Management
Waste management and disposal procedures are described in the SESD Safety, Health and
Environmental Management Program (SHEM) Procedures and Policy Manual, Section 2.5.
References
American Public Health Association. Standard Methods for the Examination of Water and
Wastewater, 20th Edition, 1998.
Science and Ecosystem Support Division, Region 4, U.S. Environmental Protection Agency.
September, 2000. Ecological Assessment Branch Laboratory Operations and Quality
Assurance Manual.
Science and Ecosystem Support Division, Region 4, U.S. Environmental Protection Agency.
May, 1998. Safety, Health and Environmental Management Program Procedures and Policy
Manual.
121
-------
Enterococci
EAB-043.0
Rev.#: 0
Date: September 08, 2003
Page 1 of 3
Procedural Section
Scope and Application
This test utilizes Enterolert to detect and quantify enterococci in both marine and recreational
fresh water with results in 24 hours.
Summary of Method
The procedure involves aseptically adding the reagent to the sample, pouring the sample into
the Quanti-Tray (counts from 1-200) or Quanti-Tray 2000 (counts from 1-2,419), sealing in the
Quanti-Tray Sealer, and then incubating at 41° C for 24 hours.
The standard volume of water to be analyzed is 100 milliliters.
Wells containing enterococci will appear flourescent, and the Most Probable Number (MPN)
table will be used to determine the count.
Interferences
Turbidity caused by the presence of algae may not permit testing of a sample volume
sufficient to yield significant results.
Low estimates of enterococci may be caused by the presence of toxic substances.
Health and Safety Procedures
Adherence to laboratory safety procedures described in the SESD Safety, Health and
Environmental Management Program (SHEM) Procedures and Policy Manual., Section 2.5 is
required.
Special Procedures
Residual chlorine in chlorinated samples should be neutralized with 0.1 mL of 10% sodium
thiosulfate (Na2S2O3).
Analyst Training
An Initial Demonstration of Capability/Performance (IDC or IDP) shall be performed prior
to the analysis of any samples, and with a significant change in instrument type, personnel,
matrix, or test method where applicable.
Analysts must demonstrate the ability to generate acceptable test results with this method by
preparing and analyzing a minimum of four aliquots of a quality control sample either
concurrently or over a period of days.
The capability of the analyst to produce acceptable results will be determined by their ability
to obtain a percent recovery of 80-120% and to obtain a relative standard deviation of <20%.
122
-------
Enterococci
EAB-043.0
Rev.#: 0
Date: September 08, 2003
Page 2 of 3
Reagents and Standards
Commercially prepared media (Enterolert) may be used.
Dilution water should be prepared in accordance with SM9050 B and C.
8.0 Apparatus and Materials
Quanti-Tray or Quanti-Tray 2000, and Quanti-Tray Sealer
Incubator, set at 41° C
Sample and dilution bottles.
Pipets and graduated cylinders.
9.0 Sample Collection and Preservation
Collect samples in clean wide-mouth plastic bottles with non-leaking caps and non-toxic
liners containing sodium thiosulfate.
Hold samples at <10°C during a maximum transport time of 6 hours.
Refrigerate samples upon arrival in the laboratory and be processed within 2 hours of arrival.
When transport conditions necessitate delays in delivery of samples longer than 6 hours,
consider using field laboratory facilities located near the site of collection.
Samples should be collected in accordance with SM 9060 A andB.
Sample Holding Time
Source samples should not be held more than 6 hours from time of collection to the time
analyses are initiated.
Drinking water samples should be analyzed within 30 hours of collection.
Method Calibration
There are no calibrations associated with this method.
Check the temperatures in the incubator twice daily to insure that it is functioning properly.
Maintain sterility with equipment, media, and technique.
Sample Analysis and Procedure
Aseptically add pre-weighed reagent to sample.
Pour into Quanti-Tray or Quanti-Tray 2000.
Seal in Quanti-Tray Sealer.
Incubate for 24 hours.
Quality Control
Laboratory Reagent Blanks (LRB) will be performed at a frequency of at least one per batch.
The batch may or may not consist of 20 or more samples which will be analyzed together as a
group.
123
-------
Enterococci
EAB-043.0
Rev.#: 0
Date: September 08, 2003
Page 3 of 3
Laboratory control samples shall be performed at a frequency of one per batch.
At least one sample must be analyzed in duplicate at a frequency of one in 10 samples.
Intralaboratory quality assurance and control should be in accordance with SM9020.
Data Analysis and Calculation
Detecting and quantifying enterococci is achieved by observing a change in fluorescence in
the well of the Quanti-Tray. Colorless results indicate a negative test.
A most probable number (MPN) table is used to determine the enterococci count.
Pollution Prevention
See SESD Safety, Health and Environmental Management Program (SHEM) Procedures and
Policy Manual, Section 5.8.
Waste Management
Waste management and disposal procedures are described in the SESD Safety, Health and
Environmental Management Program (SHEM) Procedures and Policy Manual, Section 2.5.
References
American Public Health Association. Standard Methods for the Examination of Water and
Wastewater, 20th Edition, 1998.
Science and Ecosystem Support Division, Region 4, U.S. Environmental Protection Agency.
September, 2000. Ecological Assessment Branch Laboratory Operations and Quality
Assurance Manual.
Science and Ecosystem Support Division, Region 4, U.S. Environmental Protection Agency.
May, 1998. Safety, Health and Environmental Management Program Procedures and Policy
Manual.
IDEXX. IDEXX Water Testing Method, 2001.
124
-------
Appendix 2: Storm Event Sampling Data Analysis
Data collected by EPA Region 4 contractors during the storm events of April 7th - April 11th,
2003 were analyzed to answer the following questions:
1. How similar are Total Suspended Solids (TSS) measurements gathered using the DH-59 and
ISCO samplers? Figures A2.1 and A2.2 address this issue.
2. How much information is lost by adopting a two-hour ISCO sampling interval versus a one-
hour interval? Figures A2.3 and A2.4 address this issue.
3. Given that multiple measures are taken across a stream reach when sampling stream TSS
using the DH-59, is there a relationship between the mean and standard deviation of these
multiple measures. Also, is there a relationship between the Coefficient of Variation
(StDev/Mean) and the mean of these measures. The first issue is addressed by Figure A2.5;
the second by Figure A2.6.
4. Are there significant differences between DH-59 TSS measurements taken at the edges of the
stream versus the center of the stream, versus locations intermediate to the stream edge and
center. This issue is addressed in Figure A2.7.
125
-------
Figure A2.1 shows two example plots of DH-59 versus ISCO TSS measurements. Depth
integrated TSS values at both sites tended to be smaller than ISCO TSS measures, and the
overall scatterplot showing paired points for all sites across this sampling effort reiterates this
point (Figure A2.2). The value of the intercept (11.3) and the overall slope of the relationship
(0.8739) indicate that small ISCO TSS measures tended to be lesser than depth integrated
measures, but as ISCO TSS values increase, they become greater than corresponding depth
integrated TSS measures.
Figures A2.3 and A2.4 indicate a high correlation between actual and estimated ISCO TSS
measures. Actual values were taken every hour, and estimated values were derived from these
actual values in the following manner: an estimate for hour 2 was calculated by taking the mean
of hour 1 and hour 3 values. In the same way, an estimate of the TSS value for hour 4 was
derived using the mean of the values at hour 3 and hour 5. In Figure A2.4, estimates for hours 2
and 4 are paired with the actual measured values for hours 2 and 4. The 90% confidence interval
on the slope of Figure A2.4 includes the value 1.0, but this fact is less relevant than the
percentage of variance explained by the regression line (76%). If ISCO TSS values were taken
every two hours instead of every hour, taking a mean to estimate the intermediate hour's TSS
value would be unbiased (because the intercept is zero and the slope of the regression line is not
significantly different from 1.0), but the accuracy of the estimate would be less than perfect
(because R2 is much less than 1.0).
Figures A2.5 and A2.6 indicate that as in-stream sediment increases, the standard deviation
of multiple DH-59 measures of TSS across a stream section increase, but the coefficient of
variation (StDev/Mean) of these measures declines. Basically, the standard deviation is
increasing as the mean increases, but at a lesser rate, so their ratio declines as the mean
increases.
The three plots of Figure A2.7 indicate no significant bias (90% confidence interval on the
intercept includes the value 0.0) for TSS measures taken mid-stream, at the stream edge, and at
locations between these two. The slope of the relationship between center and intermediate
measurements is significantly different than 1.0 at the 90% confidence level, but this is entirely
due to two observations with very large negative residuals (shown as triangles in the top-right
plot). Without these outliers, the resulting slope is not significantly different from 1.0. All of
this information leads us to include that TSS values for the sampled streams are uniform across a
stream section.
126
-------
CO
190
180
170
160
150
140
130
120
110
100
ISCO vs. DH-59 TSS values
Site 6, 4/10/2003
1400
1600
1800 2000
Time of Day
2200
2400
120 -i
110 -
100 -
o>
•§- 90 H
CO
80 -
70 -
60
ISCO vs. DH-59 TSS Values
Site7,4/11/2003
200
400
600
Time of Day
800
1000
1200
Figure A2.1. Comparison of Total Suspended Solid measurements using Depth-
Integrated and ISCO samplers at two sampling sites.
127
-------
300 -,
250 -
=5. 200 -
150 -
100 -
50 -
Comparison of ISCO and DH-59
Measures of TSS (mg/l)
y = 0.9723X
R2 = 0.859
50
100 150
ISCO TSS Values (mg/l)
200
250
Figure A2.2. Estimated linear relationship between Depth Integrated and ISCO Total
Suspended Solid measurements.
128
-------
250
200 -
D)
150 -
Site 1,4/10/03-4/11/03
-One Hour Sampling
Two Hour Sampling
W 100 -
50 -
— V
o
-------
o
IB
ro
I
Q
ro
•c
c
s
(0
60 n
50 -
40 -
30 -
20 -
10 -
y = 0.0711x +2.63
R2 = 0.1895
0.0
50.0
100.0 150.0
Mean
200.0
250.0
Figure A2.5. Relationship between the standard deviation and the mean of Depth
Integrated Total Suspended Solids measurements.
1.4 -,
1.2 -
o
•j=
.5
•g 0.8 H
o
4-1
I 0.6 H
o
O
0.4-
0.2 -
04217
= 0.5371X
R2 = 0.2969
0.0
50.0
100.0 150.0
Mean
200.0
250.0
Figure A2.6. Relationship between the coefficient of variation and the mean of Depth
Integrated Total Suspended Solids measurements.
130
-------
y=1.0082x+0.8283
R2 = 0.9687
0.0 50.0 100.0 150.0 200.0
Intermediate TSSValues (mg/l)
y = 0.9398x+0.9453
R2 = 0.9586
50.0 100.0 150.0 200.0
Center TSS Values (mg/l)
y = 0.9654x+ 1.0902
0.0 50.0 100.0 150.0 200.0 250.0
Center TSSValues (mg/l)
Figure A2.7. Relationships between Depth Integrated Total Suspended Solid
measurements taken at the edge, center, and intermediate locations of a stream cross-
section.
131
------- |