Citation
Understanding wastewater effluent impacts on the Clear Creek watershed

Material Information

Title:
Understanding wastewater effluent impacts on the Clear Creek watershed
Creator:
Conway, Kimberly Rae ( author )
Language:
English
Physical Description:
1 electronic file (44 pages). : ;

Subjects

Subjects / Keywords:
Water reuse -- Colorado ( lcsh )
Sewage -- Purification -- Colorado ( lcsh )
Sewage -- Purification ( fast )
Water reuse ( fast )
Colorado ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Visible changes to the water quality of Clear Creek are seen below several point sources located in Golden, CO, introduce wastewater discharge into the aquatic environment. There are several regulations within the United States that set up parameters for the amount of discharge and the type of effluent allowed to be released into a freshwater system, but fluctuations in water quality can still occur. This study was designed to provide a comprehensive comparison between water quality above and below these point sources, specifically a brewery with several prior NPDES permit violations in Golden, Colorado. Testing was conducted at two sites below the point sources and one site above with sampling dates occurring in the July, September, October, and November of 2014. Analysis confirmed the change between test sites cc-1 compared to cc-2 and cc-3 in water quality with drastic differences in several analyte concentrations. Results showed elevated temperatures, higher alkalinity and specific conductance measurements, higher concentrations of chloride, phosphorus, nitrate, nitrite, orthophosphate, total organic carbon, and E. coli downstream. Analyte concentrations also showed a correlation between stream flow with average concentrations being the lowest during the July sampling date when stream flow was the highest. Many of these analytes are directly dependent on each other, so any changes in concentration typically resulted in changes of other analyte concentrations within Clear Creek. E. coli concentrations increased downstream where nutrient concentrations were elevated and temperatures were higher. The October sampling date in particular showed the largest concentrations with the greatest increase between cc-1 and cc-2 for several of the analytes measured. The exact reason for the greater increase in analyte concentrations cannot be confirmed, but reduced flow beyond the point sources or an increase in point source discharge could very well be what caused the differences. All of these changes can pose serious risks to both environmental and human health, and should be addressed by making continual improvements to the treatment process for each point source.
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
Department of Geography and Environmental Sciences
Statement of Responsibility:
by Kimberly Rae Conway

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
921892576 ( OCLC )
ocn921892576
Classification:
LD1193.L547 2015m C66 ( lcc )

Downloads

This item has the following downloads:


Full Text
UNDERSTANDING WASTEWATER EFFLUENT IMPACTS ON THE CLEAR CREEK WATERSHED
By
KIMBERLY RAE CONWAY
B.S., Colorado State University Pueblo, 2010
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Masters of Science
Environmental Science Program
2015


This thesis for the Masters of Science degree by
Kimberly Rae Conway
has been approved for the
Environmental Sciences Program
by
Brian Page, Chair
Fred Chambers
Charles Patterson
ii


Conway, Kimberly Rae (M.S., Environmental Science)
Understanding Brewery Effluent Impacts on the Clear Creek Watershed
Thesis directed by Associate Professor Brian Page
ABSTRACT
Visible changes to the water quality of Clear Creek are seen below several point sources located
in Golden, CO, introduce wastewater discharge into the aquatic environment. There are several
regulations within the United States that set up parameters for the amount of discharge and the type of
effluent allowed to be released into a freshwater system, but fluctuations in water quality can still occur.
This study was designed to provide a comprehensive comparison between water quality above and
below these point sources, specifically a brewery with several prior NPDES permit violations in Golden,
Colorado. Testing was conducted at two sites below the point sources and one site above with sampling
dates occurring in the July, September, October, and November of 2014. Analysis confirmed the change
between test sites cc-1 compared to cc-2 and cc-3 in water quality with drastic differences in several
analyte concentrations. Results showed elevated temperatures, higher alkalinity and specific
conductance measurements, higher concentrations of chloride, phosphorus, nitrate, nitrite,
orthophosphate, total organic carbon, and E. coli downstream. Analyte concentrations also showed a
correlation between stream flow with average concentrations being the lowest during the July sampling
date when stream flow was the highest. Many of these analytes are directly dependent on each other,
so any changes in concentration typically resulted in changes of other analyte concentrations within
Clear Creek. E. coli concentrations increased downstream where nutrient concentrations were elevated
and temperatures were higher. The October sampling date in particular showed the largest
concentrations with the greatest increase between cc-1 and cc-2 for several of the analytes measured.
The exact reason for the greater increase in analyte concentrations cannot be confirmed, but reduced
flow beyond the point sources or an increase in point source discharge could very well be what caused


the differences. All of these changes can pose serious risks to both environmental and human health,
and should be addressed by making continual improvements to the treatment process for each point
source.
The form and content of this abstract are approved. I recommend its publication.
Approved: Brian Page
IV


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION...................................................................1
Research Purpose..............................................................12
II. METHODS.......................................................................13
Field Sampling................................................................13
Laboratory Analysis
Microbiology Testing...................................................16
Chemistry Analysis.....................................................17
III. RESULTS AND DISCUSSION........................................................21
IV. CONCLUSION....................................................................34
REFERENCES...........................................................................35
v


LIST OF TABLES
TABLES
1. A Breakdown of the Areas Where Wastewater is Generated and the Characteristics of the
Wastewater: 2012.........................................................................8
2. Typical Ranges of Brewery Untreated "End-of-Pipe" Wastewater Effluent: 2012.......9
3. Stream flow data in cubic feet per second for the Golden monitoring stations on sampling dates
........................................................................................21
4. Water Quality Concentrations for specific conductance, dissolved oxygen, pH, temperature, and
E. coli at sites cc-1, cc-2, and cc-3 on different sampling dates........................22
5. Nutrient Concentrations at sites cc-1, cc-2, cc-3 for all sampling dates................24
6. Anion Concentrations for sites cc-1, cc-2, cc-3 for all sampling dates..................28
7. Alkalinity and TOC Concentrations for Sites cc-1, cc-2, cc-3 for all Sampling Dates.....28
8. Estimated Total Dissolved Solids at sites cc-1, cc-2, cc-3 for all sampling dates.......31
VI


LIST OF FIGURES
FIGURES
1. Daily Maximum temperature data for Clear Creek in the Outfall 001 mixing zone, August 2006
through June 2012 provided by GEI Consultants, Inc: 2015.............................5
2. Weekly average temperature data for Clear Creek at the Outfall 001 mixing zone, August 2006
through June 2012 provided by GEI Consultants, Inc: 2015.............................5
3. Monitoring Stations with Reported Water Quality Results on MyWaters: 2013.............7
4. Location of sample site CC-1. Black star shows approximate location where water samples were
taken. Images taken during the first sampling date in July...........................13
5. Location of sample site CC-1. Black star shows approximate location where water samples were
taken. Images taken during the first sampling date in July...........................14
6. Location of sample site CC-1. Black star shows approximate location where water samples were
taken. Images taken during the first sampling date in July...........................15
7. Indirect correlation between DO and E. coli concentrations during the October sampling date .23
8. Average nitrate, nitrite, orthophosphate, and phosphorus concentrations compared to Clear
Creek streamflow data taken at the Golden monitoring station................................25
9. Process of Nitrification....................................................................26
10. Correlation between specific conductance and concentrations found for chloride and total
organic carbon..............................................................................30
11. Correlation between phosphorus, nitrate and chloride concentrations compared to total
dissolved solids............................................................................32
VII


CHAPTER I
INTRODUCTION
"Society as a whole, and many freshwater ecologists, have tended to ignore humans and their
use and misuse of fresh waters, as influential factors in the maintenance of lake and river ecosystems."
(Wetzel 2001, pg. 1) If this is the case, then there is a great deal of work that needs to be done to
understand the effects humans have on these ecosystems. Human usage of freshwater systems is
estimated every five years by the US Geological Survey (USGS); they compile county, state, and national
water withdrawal and usage data for a number of water-use categories. The last Total Water Use study,
done in 2010, found that the average amount of water used per day in the United States was 355 billion
gallons (USGS 2014). This rate of withdrawal is actually the lowest since prior to 1970, which shows that
conservation efforts have been made and are providing immediate benefits. A small fraction of this
usage is for industrial purposes; the amount withdrawn for self-supplied industrial use was 15.9 billion
gallons per day (which was mainly supplied from surface-water sources) (USGS 2014). Industrial use
may have a minor influence on water withdrawal, but can have significant impacts on the environment
due to the reintroduction of the water into the source after being used. "The impact of industrial
wastewater on rivers has attracted a lot of attention worldwide because of its overwhelming
environmental significance" (Ipeaiyeda and Onianwa 2009, pg. 189).
Industrial effluent within the United States is regulated through county, state, and federal
guidelines; there are approximately 100 different guidelines set up by the US Environmental Protection
Agency (EPA) that pertain to specific industries and the effluent these industries produce (EPA 2014).
The differentiation of these guidelines is necessary for the purpose of addressing a wide range of
analytes that have the potential to be released into the environment and potentially cause harm. This
pollution can include hazardous and non-hazardous material that can have an adverse effect on flora
and fauna that inhabit regions downstream of where reintroduction of wastewater occurs. Specific
1


analytes within the effluent may even become harmful to human health, if this water downstream is
used for recreation or for drinking water. These concerns are the driving force for the implementation
and enforcement of water treatment standards; all wastewater must go through some form of
treatment process before being reintroduced into a freshwater ecosystem. The amount of treatment
will not only depend on these standards, but the actual treatment process itself; treating industrial
wastewater can be an expensive process, especially if the water was used for multiple purposes and
accumulated a large variety of impurities. It may be difficult to remove all impurities and have the water
meet regulatory standards after treatment. It is extremely important to understand if the waste
treatment process is thorough enough to ensure environmental and human safety; many bodies of
water within the United States remain above pollution levels considered safe for the ecosystem
(Brewers Association 2012). If an ecosystem is showing visible adverse effects from the introduction of
wastewater effluent, an investigation needs to occur in order to profile what analytes are causing
stresses on that ecosystem. For that exact reason, a study was conducted on the Clear Creek Watershed
(which shows visible changes in its ecosystem within a short stretch of the creek traveling through
Golden, Colorado). There are several point and non-point sources of discharge and runoff that could be
causing these changes. The EPA Enforcement and Compliance Online (ECHO) recognizes at least six
point sources with four of these sources having National Pollutant Discharge Elimination System
(NPDES) permits regulating outfalls discharging wastewater into the creek (EPA 2014). The analytes
being monitored vary for each permit since the chemical make-up of the wastewater varies depending
on the business producing the waste.
The effects a point source may have on an ecosystem is more predictable, since there is a
clearer understanding of what chemicals are present within the wastewater compared to non-point
where the source cannot always be identified. Non-point sources include storm water runoff that has
accumulated contaminants such as roadway debris, metals, dirt, or magnesium chloride or runoff
2


coming from nearby farms or suburbs that has been contaminated with fertilizers, manure, or
pesticides/herbicides. Not only is it challenging to know what has accumulated in non-point source
runoff, it can be difficult to predict the effects it will have on the ecosystem and how to address the
impacts. In the case of Clear Creek, the non-point sources may have less of a comparative impact on the
water quality compared to the six point sources discharging into the creek. Four of these sources have
minor or major NPDES permits to monitor for various analytes within the wastewater (EPA 2014). One
source is a metal container production facility that has a minor NPDES permit, which monitors a variety
of analytes in the discharge as well as the total recoverable amount during the treatment process. Four
outfalls are present at this facility that discharge into the creek and are monitored on a monthly basis.
Some of the wastewater parameters monitored for this permit include temperature, dissolved oxygen,
pH, total suspended solids, nutrients such as nitrite and nitrate, anions like fluoride, and metals like iron,
nickel, and lead. This facility has had violations in the past, but were only for reporting issues, not
exceedances in analyte concentrations. Another facility (involved in ceramics production) discharges
wastewater into the creek, which is monitored through two minor NPDES permits. This facility also had
violations with permit reporting, but no analyte concentration violations. The effluent discharged is
monitored for cadmium, chromium, copper, cyanide, lead, nickel, pH, silver, total toxic organics, total
suspended solids, zinc, and flow on a monthly basis.
A main point source introducing wastewater into the creek is a brewing company that treats its
own wastewater and reintroduces that effluent back into the creek. This company has two facilities
located at different locations along Clear Creek that have their own separate NPDES permits (EPA 2014).
Differences in water quality of Clear Creek become apparent traveling downstream of Golden, below the
brewery, into Wheat Ridge. Some variances in the ecosystem that are noticeable without testing
include temperature and flow. Lower stream flow has been observed below the brewery compared to
stream flow above; this change can be attributed to the brewery's water withdrawal of the stream for
3


production purposes. The reduced flow can also be credited towards two canals slightly above the
brewery that divert water to Standley Lake. The overall changes in water quality, as well as
temperature, will inevitably change when less water is traveling downstream of any of the point sources
located along the creek. The more water allowed to stay in the creek, the more the wastewater will be
diluted and analyte concentrations (per liter of water) will be lower. A higher volume of water will also
help equilibrate the wastewater temperature to the natural temperature of the stream. Less volume
can encourage more heat transfer through the water from radiation, causing the entire stream to heat
up. Under the CWA and the NPDES permit program, any point source that discharges wastewater into
water must ensure that the temperature of the effluent matches the temperature of the stream, or
should only influence the overall stream temperature by a few degrees centigrade (EPA 2014).
However, these standards for temperature changes are relatively new, so limitations on temperature
changes have not yet been implemented in the NPDES permit issued to the brewing company. There
are temperature standards set up by the Colorado Department of Public Health and Environment for
certain segments of Clear Creek throughout the year, which can be compared to temperature data
recorded by GEI Consultants, Inc for the annual Aquatic Biological Monitoring of Clear Creek, Colorado
(2015). One particular segment monitored is at one of the brewery's outfalls; the daily maximum and
weekly average temperature standards were both exceeded during the winter seasons between 2006
and 2012 (Figure 1 & 2). There were also frequent exceedances of the coldwater criteria in the summer
for the daily temperature standards seen between 2006 and 2012. The water shows an apparent
warming in temperature, not only to the touch, but steam can be seen rising from the water in the early
morning in areas below the discharge, which is not observed above. The major concern with the
exceeding of temperature standards for the creek is the effect it may have on the aquatic biota within
the creek. A consultant with the brewery did prepare a report on the effects the temperature changes
4


have on aquatic biota in Clear Creek and concluded there were no negative impacts (GEI Consultants, Inc
2015).
Figure 1: Daily Maximum temperature data for Clear Creek in the Outfall 001 mixing zone, August 2006
through June 2012 (GEI Consultants, Inc 2015)
Figure 2: Weekly average temperature data for Clear Creek at the Outfall 001 mixing zone, August 2006
through June 2012 (GEI Consultants, Inc 2015)
5


Water quality monitoring of Clear Creek occurs at several locations by the City of Golden, as well
as the State of Colorado through the Colorado Department of Public Health and Environment (CDPHE),
the EPA, the Clear Creek Watershed Management Agreement Monitoring Program, and the brewery
itself at different times throughout the year. The EPA's Enforcement and Compliance Online webpage
gives information on the particular permits the brewery has and whether it has had any compliance
issues in the past four years (EPA 2015). NPDES permits have been issued to the brewery under the
CWA, which monitors temperature, pH, flow, oil and grease concentrations, and total suspended solids
(TSS) concentrations at five outfalls belonging to the upstream facility. The downstream facility has a
minor NPDES permit that monitors the settling pond where discharge is held. Chlorine, flow, oil and
grease, total suspended solids, and pH are monitored in this settling pond. In the past four years the
brewery has had seven violations regarding TSS concentrations and pH at three of the five monitored
outfalls. The NPDES permit requires the pH of water at the outfalls to be between 6.5 and 9, and the
TSS concentrations to not exceed 45 mg/L for the average weekly limit and 30 mg/L for the average
monthly limit. The largest TSS concentration was 71 mg/L, which was a recorded weekly average for the
last week in June of 2014. This particular violation was found at Outfall 008, which experienced one
other weekly average violation in June of 2013 (61 mg/L) and one monthly average violation in June of
2014 (32 mg/L). The other two outfalls that experienced TSS and pH violations were Outfall 006 and
Oil. Outfall Oil reported a TSS violation for an average weekly concentration of 54 mg/L in June of
2014 and a pH violation of 6.4 at the same time. Outfall 006 also had a TSS and pH violation in June of
2014 with a reported average weekly TSS concentration of 60 mg/L and a pH of 6.3. Only one other
parameter monitored under the NPDES permit has a set maximum limit, which is for oil and grease.
Concentrations cannot exceed 10 mg/L and any visible detection on the surface of the water would be in
violation. Temperature and flow are monitored on a quarterly bases, but set maximum and minimum
standards are not provided within the permit.
6


Data on the entire watershed can be found online on the EPA's Watershed Assessment, Tracking
& Environmental Results webpage. Water impairments, monitoring stations, and other EPA water data
can be accessed through an interactive map called MyWaters Mapper (EPA 2013). From this map, it can
Figure 3: Monitoring Stations with Reported Water Quality Results on MyWaters (Source: EPA 2013)
be determined that there are four monitoring stations directly below the effluent outfall. These specific
monitoring stations are in mixing zone areas and are ran by the Colorado Department of Public Health &
7


Environment. The data for the outfall stations is not provided by MyWaters, but water quality
monitoring results are provided for stations above the brewery and at stations below the mixing zones.
Four stations in particular may provide the best results when determining the effects the brewery has
on the creek, since these stations are collecting water that has either not been influenced by the
brewery or has been influenced, which will provide a better grasp of the overall water quality changes of
the stream. The location of these sites provided by the EPA are shown in the Figure 1. Flowever, the
majority of the data cannot be used for this comparison, since the sampling did not occur in the same
timeframe at their respective sites; this is an issue since any number of factors could have influenced the
water quality between sampling dates, making any direct comparisons between sites difficult.
Table 1: A Breakdown of the Areas where Wastewater is Generated and the
Characteristics of the Wastewater
Source Operation Characteristics
Mash Tun Rinsing Cellulose, sugars, amino acids. ~3,000 ppm Biochemical Oxygen Demand (BOD)
Lauter Tun Rinsing Cellulose, sugars, spent grain. Suspended solids (SS)~3,000 ppm, BOD ~ 10,000 ppm
Spent Grain Last running and washing Cellulose, nitrogenous material. Very high in SS (~30,000 ppm). Up to 100,000 ppm BOD
Boil Kettle Dewatering Nitrogenous residue. BOD ~2,000 ppm
Whirlpool Rinsing spent hops and hot trub Proteins, sludge and wort. High in SS (~35,000 ppm). BOD ~ 85,000 ppm
Fermenters Rinsing spent hops and hot trub Yeast SS ~ 6,000 ppm, BOD up to 100,000 ppm
Storage tanks Rinsing Beer, yeast, protein. High SS (~4,000 ppm). BOD ~80,000ppm
Filtration Cleaning, start up, end of filtration, leaks during filtration Excessive SS (up to 60,000 ppm). Beer, yeast, proteins. BOD up to 135,000 ppm
Beer spills Waste, flushing etc. 1,000 ppm BOD
Bottle washer Discharges from bottle washer operation High pH due to chemical used. Also high SS and BOD, especially thru load of paper pulp.
Keg washer Discharges from keg washing operations Low in SS (~400 ppm). High BOD
Miscellaneous Discharged cleaning and sanitation materials. Floor washing, flushing water, boiler blow-down etc. Relatively low on SS and BOD. Problem is pH due to chemicals being used.
(Chart source: Brewers Association 2012)
8


The water use of the brewing industry is becoming a global concern due to the extensive
amount of water required in the brewing process and the large amount of polluted effluent produced
afterward. On average, breweries within the United States discharge 70% of their incoming water as
effluent (Brewers Association 2012). During the brewing process, large quantities of water are used for
the production of beer itself, as well as for washing, cleaning and sterilizing of various units (tanks,
bottles, machines and floors) during production (Olajire 2012, Ipeaiyeda and Onianwa 2009).
The wastewater should be subjected to an on-site pre-treatment process in order to reduce the
strength of the effluent being discharged. Discharge limits are applied more often to brewers due to the
high organic nature of the wastewater; this organic material can cause a rapid depletion of oxygen in the
surface water the effluent is being introduced into, which can have negative impacts on living species
and biodiversity within that waterway (Brewers Association 2012). Discharge and strength of effluent
are monitored in the United States under the Clean Water Act (CWA) (EPA 2015). Some of the rules
under the CWA (EPA 2015), including Effluent Limitations Guidelines, Pre-treatment Streaming Rule,
NPDES Permit Program, Sewage Sludge Rule, and Total Maximum Daily Load and Impaired Water Rules
(EPA 2014) can have a direct or indirect impact on brewery operations (Brewers Association 2012). If
these regulations were not in place, brewers may be less inclined to treat the wastewater and remove
analytes that could cause harm to the ecosystem.
Table 2: Typical Ranges of Brewery Untreated "End-of-Pipe" Wastewater Effluent
Parameter Typical Range
Biochemical Oxygen Demand (BOD) 600-5,000 ppm
Chemical Oxygen Demand (COD) 1,800-5,500 ppm
Nitrogen 25-80 mg/L
Phosphorus 10-50 mg/L
pH 3-12 (Standard Units)
Total Suspended Solids 200-1,500 ppm
(Chart source: Brewers Association 2012)
9


Research on the effects of untreated brewery wastewater on the ecosystem have been
conducted in countries where water quality regulations are not in place. One such study looked at the
impacts the effluent had on the Olosun River located in Ibadan, Nigeria (Ipeaiyeda and Onianwa 2009).
Quantifying these impacts was done by taking two samples above the brewery and five samples at
varied distances below the brewery. These samples were analyzed for the following parameters
(Ipeaiyeard and Onianwas 2009):
ammonia (NH4) (nesslerisation colorimetric method)
biochemical oxygen demand (BOD) (dilution method with Winkler's titration)
chemical oxygen demand (COD) (potassium dichromate oxidation and titrimetry)
chloride (CL ) (mercurimetric titration)
dissolved oxygen (DO) (Winkler's titration)
nitrate (N03) (phenodisulphonic acid colorimetric method)
sulfate (S042 ) (turbidimetry)
total solids (TS)
total suspended solids (TSS) (gravimetry)
turbidity (turbidimetry)
Samples were also tested for Cadmium (Cd), Calcium (Ca), Chromium (Cr), Cobalt (Co), Copper (Cu), Lead
(Pb), Nickel (Nl), Zinc (Zn) using atomic absorption spectrophotometry (Perkin Elmer model 2380). The
analysis conducted showed the average pH values upstream were much higher than the corresponding
values downstream. "The pH fall is traceable to acidic discharge of effluent into the river" (Ipeaiyeda
and Onianwa 2009, pg. 196). The DO also decreased due to an increased consumption rate from the
higher organic load of matter within the effluent. This consumption rate will increase due to a
stimulated algae growth caused by an increase in phosphate levels (Ipeaiyeda and Onianwa 2009). They
also found "from upstream to downstream, the turbidity, Cl", NH3, Ni, Zn, Cr, Co, Cu, Cd and Pb levels of
10


samples increased considerably and were significantly above the recommended water quality
standards" (pg. 197). A possible reason for the increase in ammonia levels traveling downstream could
be due to "the conversion of organic forms of nitrogen in protein and chitin containing materials in the
effluent to ammonium nitrogen" (pg. 197) as the effluent flows down river. The BOD and COD were also
found to have increased values downstream from the brewery. "This is in accordance with the fact that
high organic load is found in wastewater from ri(n)sing of bottles and mixing tanks, which consequently
affect high bacterial population(s) and very low oxygen level(s) in the river" (pg. 198). There was also
evidence of a build-up of heavy metals, particularly Ni, Zn, Cu and Pb within crops surrounding the river,
which could pose a risk to humans when consumed (Ipeaiyeda and Onianwa 2009).
The presence and/or change in levels of certain chemicals within a body of water caused by
anthropogenic activities can create a cascade of adverse effects towards the aquatic biota. N.S. Oluah
studied the effects of brewery wastewater effluent on the freshwater catfish Clarias albopunctatus
(2007). This species of catfish was tested in a controlled environment thus variations observed in the
catfish were entirely dependent on the concentration of polluted effluent they were exposed to. It was
found that the groups of catfish exposed to increased amounts of effluent showed both a decrease in
haemoglobin levels as well as decreased levels in erythrocyte counts. "The progressive reduction in the
haemoglobin concentration in C. albopunctatus exposed to brewery wastewater is indicative of
impaired oxygen-carrying capacity in the fish" (pg. 159). Studies on other chemicals within the brewery
effluent have shown depressed growth in leafy vegetation from exposure to increased concentration of
nitrate (Chen et al. 2004). "The combination of high S042- and high alkalinity in surface water negatively
affected Lobelia dortmanna plants causing mortality, decreasing biomass and reducing actual
photosynthetic efficiency" (Pulido, et al. 2012, pg. 291). T.D. Davies also found that increased levels of
sulfate caused reduction in final shoot length, reduction in final dry weight, and reduction in chlorophyll
11


a and b in the aquatic moss species Fontinalis antipyretica (2006). Elevated levels of lead were found to
significantly reduce the average lifespan of certain species of Cladocera (Garcia-Garcia 2006).
Research Purpose
This study was designed to provide a comprehensive comparison between water quality above
and below the brewery and other point sources located in Golden. The issues associated with the data
provided by the EPA were addressed by testing the sites on the same day so a more accurate
comparison could be made. The chemical and statistical analysis performed in this study was modeled
after the research done by Ipeaiyeda and Onianwa (2009) on the Olosun River. There are some key
difference between this study and the one on the Olosun River, including the time frame the samples
were taken and the analysis of heavy metals. This study was performed over a period of one year, and
analysis of heavy metals was omitted due to budget restrictions for this study. The EPA Region 8
laboratory assisted in retrieving the samples and sample analysis at their laboratory in Golden, Colorado.
This study also presents the impacts of treated effluent from a four different point sources, not just the
brewery wastewater, on water quality instead of untreated effluent as was seen with the Olosun River
study.
12


CHAPTER II
METHODS
Field Sampling
Field sampling occurred at three different sites on July 1, 2014, September 5, 2014, October 7,
2014, and November 4, 2014. The three sites used were close to three CDPHE monitoring stations that
provide water quality data. The fourth station highlighted in orange was inaccessible, so no sampling
occurred in that area. The first sampling site was located near the monitoring station highlighted in red.
This was near the parking lot at Prospect Park in Wheat Ridge, Colorado; the sampling ID for this
Comcast C/5
EH] W 44th Ave r&e|
A
Rocky Mountain Forest Products i< Prospect RV Park z 2P. n>"


Tabor St v#* <^ c2
rail erf*** Prospect Perk C'.fpar ' Trait
*
Figure 4: Location of sample site CC-3. Black star shows approximate location where water samples
were taken. Images taken during the first sampling date in July.
13


location is designated as "cc-3" (Figure 4). The next water samples were taken close to the monitoring
station highlighted in green under the Youngfield St. Bridge. The sampling ID for this location is "cc-2"
(Figure 5). The last sampling site was near the monitoring station highlighted in blue. This was near the
Figure 5: Location of sample site CC-2. Black star shows approximate location where water samples
were taken. Images taken during the first sampling date in July.
parking lot of the Golden Library and Clear Creek History Park. The sampling ID for this location is "cc-1"
(Figure 5).
Two bulk samples were taken at each site and split into five different containers that were later
tested for alkalinity, anions, total nitrogen (TN), total phosphorus (TP), total organic carbon (TOC),
nutrients, and E. coli. Each container consisted of its own laboratory ID #; alkalinity and anions were
tested using the same 250mL container, which was stored at 4C; TN and TOC were tested using the
14


same 125mL container with HCI preservative; TP was tested using its own 125mL container with H2S04
preservative; nutrients were tested from the 60mL container that was frozen before analysis. The last


%
Parfet Park +
v5.v>


arts Park
Clear Creek History Park %
Cta-r- creek
sN

O^*
%
^fpi



CO W
ft* C-Sl
Figure 6: Location of sample site CC-1. Black star shows approximate location where water samples
were taken. Images taken during the first sampling date in July.
container was a 120mL sealed, plastic bottle for E. coli analysis.
These samples were taken by wading out into a safe part of the stream where a collapsible,
sterile, square container (cubitainer) could be completely immersed under the water. The cubitainer
was triple rinsed with water from the creek before a final sample was collected. This water would then
15


be used to fill four of the five small plastic containers. Water was immediately collected in the sealed
sterile plastic bottle after collecting water in the cubitainer in the same location. Laboratory testing only
required lOOmL; excess water was poured out until the meniscus was at the lOOmL line. Once all the
smaller containers were filled for that location, they were placed into a small bag and put on ice in a
cooler for storage while out in the field.
Temperature, pH, conductivity, and DO were tested on site using an In-Situ TROLL 95000. This
meter was placed a couple of feet upstream from where water samples were taken to ensure that any
disturbances from wading did not affect the meter. The meter was placed at the bottom of the creek
where it could be completely immersed under the water. Measurements were then recorded in field
notes and saved on the meter for every site before removing the meter from the water.
Laboratory Analysis
Microbiology Testing. All water samples were submitted to the EPA region 8 laboratory for
testing. The E. coli samples were tested in the microbiology lab at the region 8 laboratory within 24
hours of submission. These samples were tested using the Colilert Enzyme Substrate method with
Quanti-Trays (IDEXX, Inc.). This process started by homogenizing the samples using a vortex, followed
by pouring off any excess water over the lOOmL mark. One ampoule of Colilert-18 media was then
added to each container and mixed by tightly recapping and vortexing again. The contents of this
container were then poured into a 96 well Quanti-Tray/2000, which is then sealed by placing the tray in
the Quanti-Tray/2000 rubber insert and feeding it into the Quanti-Tray sealer. Once each tray is sealed
and labeled with its proper ID, it is loaded into an incubator (35 + 0.5C) for 18 (+ 4) hours.
Standard Quanti-Tray/2000 comparators are also used as a color reference for the creek
samples in order to determine if coliform bacteria and E. coli are present. The reference samples are
made by filling 4 sterile containers with lOOmL of ultra-pure (~18 MQ-cm) autoclaved water. One
ampoule of Colilert media is then added to each container and mixed by tightly recapping and vigorously
16


shaking 25 times. Four test tube vials containing E. coli, Klebsiella pneumoniae, Pseudomonas
aeruginosa, and a blank are then paired with one of the containers. Each container is inoculated with its
designated test tube by dipping a plastic loop into the test tube and transferring the liquid film formed
across the plastic loop into its respective container. The contents of the loop are spread by swirling the
loop in the ultra-pure water within the container. Once all controls are inoculated, the contents of each
container are then transferred into a Quanti-Tray/2000. These trays are incubated alongside the sample
Quanti-Trays for the same period of time.
If the colors between the sample Quanti-Tray/2000 and the comparator are indistinguishable
after 24 hours, then both must be placed back into the incubator for no more than the maximum 28
hour incubation period and read at a later time. The results of the samples cannot be accepted until the
controls and blank are verified. In this case, the blank should remain colorless and not fluoresce, the E.
coli QC sample should turn yellow and fluoresce under 366-367 nm UV light, the K. pneumonia QC
sample should turn yellow and not fluoresce, and the P. aeruginosa QC sample should not turn yellow or
fluoresce. If the controls perform as expected, then the samples may be pulled from the incubators and
analyzed. After incubation, the color of any well in the sample Quanti-Tray/2000s is compared to the
wells of the comparator. If the sample wells have a yellow color that is equal to or greater than the
color intensity of the comparator wells, then the sample is total coliform positive. All of the small and
large wells are then totaled and recorded as most probable number (MPN) using the 96 well Quanti-Tray
MPN table. E. coli positive is determined by comparing the fluorescence of any of the wells in the
sample trays to the comparator wells. If the sample trays have a fluorescence that is equal to or greater
than the intensity of the comparator wells then the sample is E. coli positive. The total number of wells
that fluoresced are then added and recorded as MPN using the same MPN table.
Chemical Analysis. Alkalinity is a measure of the capacity of water to neutralize acids (EPA
2012). The testing is performed using a DL 50 autotitrator that analyzes 50 mL of sample from the 250
17


mL LDPE sample bottle. The DL 50 autotitrator measures alkalinity by measuring the amount of sulfuric
acid needed to drop the sample pH to 4.2. Analysis can begin after the machine is calibrated and the
sample table is updated on the DL 50 software. Each sample, along with a blank and any controls taken,
can then be loaded onto the sample charger according to the sample table. Once measurements are
complete and the results are recorded in the sample table the alkalinity report can be printed and
manually added into the Laboratory Information Management System (LIMS) for review. Before the
results for alkalinity can be reviewed, an acceptance criteria method must be performed (which should
give a pH greater than 4.0). The quality control samples are also used to verify the accuracy of the
standardization and to identify any matrix effects; if errors are seen with either the quality control
samples or the acceptance criteria method, then the alkalinity results cannot be used.
Water collected in the same 250 mL LDPE sample bottle can be used to test for anions. Testing
must occur within 48 hours of collection for nitrite-N, nitrate-N and orthophosphate-P, and within 28
days of collection for fluoride, chloride, bromide, and sulfate. The water used for this test must be
filtered using a 0.45 micron filter prior to analysis. Dilutions of the sample may be required if anion
concentrations are too high or if the pH is 4.5 or lower. After filtration, 5.9 mL of well-mixed sample can
be added to a 10 mL polyvial and placed into the Dionex ion chromatograph model ICS5000 autosampler
with standards and controls for analysis. If the response for the peak exceeds the working range of the
system, the sample will be diluted with an appropriate amount of reagent water and reanalyzed.
Results are generated automatically through the Chromeleon software and later transferred to the LIMS
for review.
Water collected in the 60 mL LDPE bottle can be analyzed for the presence of Nitrate-N, Nitrite-
N and Orthophosphate-P. This sample should be analyzed within 48 hours of collection. If this is not
possible, then the sample can be frozen and analyzed within 28 days. Prior to analysis, EDTA is added to
eliminate interference from metals and turbid samples must be filtered using a 0.45 pm filter. Analysis
18


begins by powering on all units of the instrument and verifying that the instrument is properly
communicating with the Omnion software. Once equipment calibration is complete and the software is
updated with the new sample table, sample analysis can start by loading the sampler tray with the
standards, QC controls, and samples. Results are automatically generated through the software by
comparing the sample absorbance to the calibration curve (created for the known standards), giving a
value for each sample. If nitrates and nitrites need to be recorded separately, the Cd reduction column
must be disconnected and the nitrite-N can only be measured. Then this value can be subtracted from
the (Nitrate + Nitrite)-N result giving the nitrate-N value. All data can then be uploaded into LIMS and
formatted into an Excel file for review.
Water collected in the 125 mL plastic container, with HCI preservative, was used for the TOC and
TN analysis. The Shimudzu TOC/TN analyzer is used to measure TOC or TN, which can be changed by
changing the furnace temperature (680C for TOC analysis or 720C for TN). The sample must be
transferred into the appropriate glass sample bottle that fits into the Shimudzu sample tray. Once the
equipment's software is updated with the correct sample table the samples can be loaded into the
sample tray and the analysis can begin. Results will be generated automatically by the software.
Samples whose concentrations exceed the calibration curve must be diluted and rerun; after QC
samples have been verified, the results for the samples can be transferred into the LIMS and into an
Excel spreadsheet for review.
The last sample analysis done within the lab is the total phosphorus (TP) test. This uses the
water collected in the 125 mL container with H2S04 preservative. This water will be transferred into a
disposable 16 X 100 mm glass culture tube and digested. Digestion occurs by adding 2.0 mL of digestion
reagent into the glass culture tube and placing it into an All American electric pressure steam sterilizer
for a half hour. The temperature of the pressure steamer should be 250F at 15 psi. Once the sample
has been taken out of the pressure steamer, the Lachat QC8500 Flow Injection Autoanalyzer can be set
19


up with the appropriate detection parameters for phosphate-P analysis including an 880 nm filter and
heating coil at 37C, with the pump tubes in their respective reagent containers. The reagents should be
allowed to flow until a stable baseline is obtained. Analysis can begin once the sample tray is loaded
with the standards, controls and samples. Results are found by comparing sample absorbance to the
standard calibration curve of the known standards. These results are automatically generated by the
software, which are then transferred into LIMS and an Excel spreadsheet for review.
20


CHAPTER III
RESULTS AND DISCUSSION
Changes in water quality of Clear Creek between the sampling locations are apparent. Tables 4
through 7 show that analyte concentrations were affected in the creek along the stretch traveling
through Golden. Whether these changes were caused by the brewery, the other point sources, or even
the non-point sources cannot be established, but some of the differences in analyte levels are great
enough to imply that the changes are from anthropogenic activity. Concentrations for many of the
parameters tested were also greatly affected by the stream flow. Flow was not measured at the specific
sites due to safety issues, but flow data from the USGS was retrieved from one station located in Golden
(Table 3) (USGS 2015).
Table 3: Stream flow data in cubic feet per second for the Golden monitoring stations on sampling dates
USGS 06719505 CLEAR CREEK AT GOLDEN, CO
Date Discharge (cubic feet per second)
7/1/2014 948
9/5/2014 182
10/7/2014 87
11/4/2014 83
Flow rates were the highest during the first testing date in July for both monitoring stations,
with Golden showing an average flow of 948 cubic feet per second (cfs). This high flow (dilution) would
account for why concentrations for many of the parameters tested were lower than concentrations
measured during the other three sampling dates. The prevalence of E. coli was one such parameter
affected by stream flow; having the lowest colony count and changes during the July sampling date.
Unfortunately, there are no flow gaging stations below the point sources so only visual comparisons
could be made on the changes in flow. A basic trend seen for all sampling dates was the decreased flow
at the two testing sites downstream; as mentioned before, the decrease in flow could be from water
being withdrawn from the canals that distribute water from the creek to Standley Lake or the brewery
that uses the water for production. The differences in flow between sampling dates and between
21


sampling sites may have also had an effect on temperature. The downstream sampling sites had
elevated temperatures compared to site cc-1, which could have been due to the reduced flow or the
increase in direct sunlight from being out of the canyon. The increase in temperature could also be
caused by the brewery effluent, since past temperature measurements in mixing zones recorded values
higher than the daily maximum standard for the creek (Figures 1 and 2). Water temperatures tended to
stay elevated downstream, which also could have resulted from the shallow flow. Other point sources
may have added effluent with temperatures higher than the creek temperature, which could also allow
raised temperatures to persist.
Table 4: Water Quality Concentrations for Specific Conductance, DO, pH, Temperature, and E. coli at
Sites cc-1, cc-2, and cc-3 on Different Sampling Dates
Sample Date Site Specific Conductance (uS/cm) Dissolved Oxygen (mg/L) pH(SU) Temperature (C) E coli (MPN/lOOmL)
7/1/2014 CC-1 (Golden) 99 9.1 6.5 10.85 13.2
7/1/2014 CC-2 (Youngfield) 122 9 6.96 12.53 13.2
7/1/2014 CC-3 (Prospect) 128 8.9 7.66 12.8 18.3
9/5/2014 CC-1 224 8.7 7.37 12.55 77.6
9/5/2014 CC-2 382 7.5 7.85 18.21 >2420
9/5/2014 CC-3 371 7.6 8.12 18.33 >2420
10/7/2014 CC-1 245 9.5 7.38 8.69 14.5
10/7/2014 CC-2 815 6.1 7.8 16.22 86
10/7/2014 CC-3 749 7 7.97 14.49 50.4
11/4/2014 CC-1 299 11.1 7.31 2.49 8.4
11/4/2014 CC-2 403 9 7.69 10.76 111
11/4/2014 CC-3 406 9.1 7.89 10.18 210
All four sampling dates did show a pattern in temperature increase between sampling sites, with
the sites below the brewery being higher. The largest difference between two sites was cc-1 (above)
and cc-2 (below) in November, which had approximately an 8C increase after the brewery mixing zone
(Table 4). The smallest increase was in July with a difference of only <2C from cc-1 to cc-2 and cc-3.
These temperature changes were associated with the amount of E. coli found at each site. The presence
of E. coli is an indicator of recent sewage or animal waste contamination (EPA 2013). The increase in E.
coli at the sites downstream could either be caused by an increase in runoff with waste contamination
22


or a combination of higher temperatures and an increase in nutrients. All strains of E. coli are
facultative, aerobic and anaerobic, gram-negative bacteria that grow best in temperatures around 37C
(Todar 2008-2012). The warmest temperatures were recorded in September at 18.33 and 18.21C. The
two locations where these temperatures were recorded were cc-2 and cc-3, which experienced the
highest counts of E. coli with >2420 MPN/lOOmL (TNTC for 96 well Quanti-Tray/2000). The drastic spike
in E. coli between the September sampling date and the July sampling date was not expected, so
dilutions were not performed for a more accurate MPN count. Nutrient concentrations appeared to
trend with the amount of E. coli present between sampling sites, but not between sampling dates. E.
coli concentrations increased downstream where nutrient concentrations were also elevated. Under
anaerobic conditions, E. coli will grow by utilizing nitrates and nitrites as final electron acceptors for
respiratory electron transport processes (Todar 2008-2012). The sampling date with the highest
presence of E. coli did not correlate with overall highest nutrient concentrations, and instead correlated
with the highest temperatures recorded. E. coli can also grow aerobically by respiration using 02, which
could explain the inverse correlation between DO and E. coli concentrations (Unden et al. 1994).
Inverse Correlation of DO against E. coli Concentrations and Temperature
Figure 7: Inverse correlation of DO against E. coli concentrations and Temperature during the October
sampling date
23


The only measurement taken showing an apparent decrease downstream was dissolved oxygen
(DO) (Table 4). This correlates with the increase in E. coli growth, which provides some explanation as to
the reduced levels in DO (Figure 7). Similar to other facultative anaerobic bacteria, E. coli's highest
priority level for growth and energetics is through aerobic metabolism, which can be triggered by the
presence of 02 (Unden 1994). E. coli has the regulator mechanisms to then change its metabolic
pathway to anaerobic once the amount of available 02 decreases, therefore turning to nitrate as the
main nutrient for growth (Unden 1994). If DO concentrations were completely dependent on the
amount of E. coli present, then the sampling date in September with the highest counts of E. coli should
have seen the lowest concentrations of DO, which was not the case. Other aquatic biota, such as algae,
could also have a direct effect on the amount of DO; the increase in nutrients downstream may be
promoting algae growth, which would result in an increase in DO since algae release oxygen during
photosynthesis. This would explain why there was a slight increase in dissolved oxygen between cc-2
and cc-3 where nutrients were high.
Table 5: Nutrient Concentrations at Sites cc-1, cc-2, cc-3 for all Sampling Dates
Sample Date Site Phosphorous (ug/L) Nitrate asN (ug/L) Nitrite asN (ug/L) Orthophosphate as P (ug/L) Total Nitrogen (mg/L)
7/1/2014 CC-1 (Golden) <10.0 102 <5.0 19.1 0.166
7/1/2014 CC-2 (Youngfield) 29.1 105 15.4 37.6 0.247
7/1/2014 CC-3 (Prospect) 36.8 104 15.5 41.2 0.242
9/5/2014 CC-1 19.2 238 <5.0 22.6 0.247
9/5/2014 CC-2 273 308 73.6 203 0.845
9/5/2014 CC-3 285 319 61.1 208 0.772
10/7/2014 CC-1 <10 155 <5.0 27.3 0.228
10/7/2014 CC-2 240 970 315 180 3.75
10/7/2014 CC-3 302 1190 201 258 2.7
11/4/2014 CC-1 <10.0 224 <5.0 19 0.257
11/4/2014 CC-2 55.9 322 81.3 46.8 0.885
11/4/2014 CC-3 54.9 365 70.2 46 0.781
All nutrient concentrations increased at sampling sites below the brewery compared to the
testing site above (Table 5). The nutrient that showed the most drastic increase was nitrate, with the
most significant difference being around 800 ug/L. The nutrient with the most notable change was
nitrite, since levels were actually below detection limits upstream of all the point sources at cc-1 for all
24


sampling dates (this trend was also seen with phosphorus concentrations for three of the sampling
dates). The drastic change in nutrient concentrations could have been caused by any of the point
sources, or a combination of them and non-point sources from runoff. Anthropogenic influences on the
creek can cause a variety of issues or changes that can offset the natural balance of aquatic life. The
direct relationship between E. coli, temperature and all of the nutrients is one such example that shows
how this balance can change if a viable nutrients are provided. Nitrite showed a decrease in
concentration between cc-2 and cc-3 with the largest decrease in concentration seen between cc-2
Nitrate, Nitrite, Orthophosphate, Phosphorus and Dissolved Oxygen
concentrations compared to Clear Creek streamflow
1000
900
800
700
600
500
400
300
200
100
0
7/1/2014
9/5/2014 10/7/2014 11/4/2014
Golden Streamflow (cfs)
Orthophosphate Average (ug/L)
Phosphorus Average (ug/L)
Nitrate Average (ug/L)
Nitrite Average (ug/L)
Dissolved Oxygen (mg/L)
12 5
10
8
6
4
2
0
£
0)
o
£
o
u
£
0)
tXO
£
o
c
>
Figure 8: Average nitrate, nitrite, orthophosphate, phosphorus and dissolved oxygen concentrations
compared to Clear Creek streamflow data taken at the Golden monitoring station
Note: Concentrations for phosphorous and nitrite measured at cc-1 were omitted since values
were below detection limits (phosphorous 9/5/2014for cc-1 was included)
(315ug/L) and cc-3 (201ug/L) in October. This was on the same day as the largest increase in nitrite
between cc-1 (<5.0ug/L) and cc-2. As nitrite concentrations decreased between sites cc-2 and cc-3,
nitrate concentrations increased. The process of nitrification best explains why there is an inverse
correlation between nitrite and nitrate concentrations (Scott 2012). With the assistance of a nitrite-
oxidizer such as Nitrobacter and oxygen as the oxidizing agent, nitrite will be oxidized into nitrate (Figure
25


9). This would explain the decrease in nitrite and increase in nitrate between the two testing sites
below the point sources. When nitrification occurs in a closed system, DO levels decrease and the pH
becomes more acidic since for every two nitrates produced, four hydrogens are produced and one
oxygen is consumed (Figure 9). Instead, the pH became more basic and the DO levels stayed relatively
constant between the downstream sites. Depressed pHs (more acidic) will only occur if there is not a
sufficient alkalinity concentration (Campbell and Ogden 1999) and if nitrite concentrations were
increased from the oxidation of ammonia (Figure 9) (Anthonison et al. 1976). In the case of the testing
sites below the point sources in Clear Creek, alkalinity was relatively high (Table 7) which would explain
why pH was not affected by the nitrification process. The decrease in DO concentrations from
nitrification could have been offset by oxygen producing algae growing in the water.
The nutrient concentrations were much more diluted during the high flow sampling date in July,
which could potentially reduce nutrient loading on the ecosystem (Figure 8). Flow did not correlate
completely with the average nutrient concentrations since the results for October showed the highest
concentrations for all of the nutrients. It should be noted, though, that the difference between cc-1 and
26


cc-2 concentrations for all nutrients was greatest during the October sampling date with relatively large
increases between the two sites. This could be a result from an increase in anthropogenic activity on
October 7th or there may have been an increase in water withdrawal between the two sampling sites,
therefore changing the flow. This makes comparing available flow data with concentration difficult since
Clear Creek's streamflow can drastically change due to water withdrawal. The other nutrients tested
showed an increase in concentration between cc-2 and cc-3, which shows that these analytes tend to
persist in the aquatic ecosystem. Persistence seems to not be the only factor with these nutrients, since
concentrations do in fact increase between the sites that are both downstream of the brewery and the
other point sources. The reasons for this increase in phosphorus, nitrate, and orthophosphate could be
caused by contaminated runoff from non-point sources such as farms that use these nutrients in their
fertilizers. The creek itself may also foster pollutant accumulation or reintroduction, especially if there is
a lack in a sufficient riparian buffer zone. The larger the distance between the creek and the non-point
source runoff (meaning the larger the buffer zone is), the less chance there is for contaminants to make
it to the creek; it will instead be deposited or filtered out of the runoff and into the soil. These nutrients
may also be deposited in soil along the stream banks during low streamflow and later be reintroduced
back into the water when stream discharge increases.
Measurements for orthophosphate and total phosphorus are closely related since total
phosphorus is the sum of orthophosphates, organically bound phosphates, and condensed phosphates
(EPA 2012). Orthophosphate concentrations should be less than all of the total phosphates, which was
not seen for any of the dates sampled for cc-1 and was not seen at either site during the high flow
sampling date. The best explanation for this occurrence is how the phosphorus was analyzed; lower
concentrations of phosphorus compared to orthophosphate could occur if the water sample was filtered
first, which will only allow for total dissolved phosphorus to be measured. If this was the form that was
analyzed, then higher orthophosphate concentrations would be possible. This means that the
27


orthophosphates were the only phosphates detected for the total dissolved phosphorus at the
sites/dates that experienced higher concentrations of orthophosphates.
Table 6: Anion Concentrations for Sites cc-1, cc-2, cc-3 for all Sampling Dates
Sample Date Site Chloride (mg/L) Fluoride (mg/L) Sulfate as S04 (mg/L)
7/1/2014 CC-1 (Golden) 3.88 0.29 15.8
7/1/2014 CC-2 (Youngfield) 6.12 0.3 17.1
7/1/2014 CC-3 (Prospect) 6.78 0.31 17.5
9/5/2014 CC-1 13.3 0.43 45
9/5/2014 CC-2 30.4 0.51 45.7
9/5/2014 CC-3 29.7 0.46 43.3
10/7/2014 CC-1 13.4 0.39 52.4
10/7/2014 CC-2 69 0.76 67.9
10/7/2014 CC-3 69.3 0.63 65.9
11/4/2014 CC-1 19.7 0.49 74.8
11/4/2014 CC-2 31.6 0.61 77.9
11/4/2014 CC-3 32.2 0.62 79.3
Out of the three anions tested, only chloride showed a distinct difference in concentration from
the site upstream of the point sources compared to the two sites below (Table 6). As previously seen
with other analytes tested, the increase in concentration for chloride was less significant during the high
flow sampling date. The largest difference was seen in October; where concentrations went from
13.4mg/L at cc-1 to 69.3mg/L downstream at cc-3. The large increase does not correlate with stream
flow, since the subsequent sample date had less flow and a smaller concentration average for the three
sites. The cause for the large increase in October could be from increased point source discharge.
Table 7: Alkalinity and TOC Concentrations for Sites cc-1, cc-2, cc-3 for all Sampling Dates
Sample Date Site Alkalinity (mg/L) Total Organic Carbon (mg/L)
7/1/2014 CC-1 (Golden) 21.6 2.07
7/1/2014 CC-2 (Youngfield) 27.2 2.41
7/1/2014 CC-3 (Prospect) 29 2.26
9/5/2014 CC-1 34 1.64
9/5/2014 CC-2 84.6 4.03
9/5/2014 CC-3 83.4 3.86
10/7/2014 CC-1 38.5 1.3
10/7/2014 CC-2 222 9.86
10/7/2014 CC-3 195 7.88
11/4/2014 CC-1 45 1.09
11/4/2014 CC-2 78.7 2.95
11/4/2014 CC-3 78.2 2.75
28


Whether one point source was discharging more that day or there were more active outfalls from
several of the point sources cannot be verified, but the increased chloride concentration for all four
sample dates was most likely caused by treated discharge from one or more of these sources. Many
water treatment processes incorporate the use of chloride species as a form of disinfectant, which
would account for the changes in concentrations at both cc-2 and cc-3. The anion with the highest
concentrations overall was sulfate, with one measurement reaching up to 79.3mg/L. The presence of
sulfate and fluoride were most likely not caused by any discharge being released between cc-1 and cc-3,
since concentrations stayed relatively the same for all three sampling sites. Fluctuations in the amount
of sulfate and fluoride could simply be due to the accuracy of the autosampler during sample analysis,
since percent recovery for reference samples can be plus or minus 10%.
Another indication that the water quality is impaired downstream from the combination of
point sources and non-point sources is the amount of TOCs found in the water per liter. All four
sampling dates showed an increase in concentration between cc-1 to cc-2 (Table 7). The sampling date
that experienced the greatest increase was in October, where TOC levels increased around 8.5 mg/L
between cc-1 and cc-2. This day showed the largest concentrations and the greatest increase between
cc-1 and cc-2 for several of the analytes measured. Once again, this could be a result of an increase in
anthropogenic activity from either one or more of the point sources discharging into the stream. TOC
concentrations were relatively stable for the other three sampling dates with only minor increases
between the sampling sites. This is not the case in regions receiving untreated point source discharge as
seen with the Olosun River, which showed large analyte concentration increases downstream of the
brewery (Ipeaiyeda and Onianwa 2009). The small TOC concentration increases seen below the brewery
and the other point sources discharging in Clear Creek may be due to the outfalls being inactive at that
specific time or the wastewater treatment process is efficient at filtering out organic matter. These
minor changes may not even be associated with the brewery or other point sources, and could actually
29


be caused by a natural sources like decaying plants (Schumacher 2002). Analysis determining the
quantity of decaying plants and leaf litter at each site would be required to prove that the concentration
spike was from a natural source. Another interesting pattern observed was the drop in the amount of
TOC present after cc-2 for all sampling dates; this could be from the larger organic particles settling into
the sediment at the bottom of the creek or micro and macro organisms consuming the particles.
The other water quality parameter that seemed to correlate with TOC and chloride
concentrations was specific conductance (Table 4 and Figure 10) (USGS 2014). Specific conductance is a
measure that shows how well water can conduct electrical current, which can increase with the
introduction of dissolved solids such as salt (USGS 2014). This would explain why conductance increased
at sites below the brewery where concentrations of chloride and TOC were higher as well (Figure 10).
fC
+-
o
T3
C
LT)
3
CuO 600
O 400
tn
~o
a) 300
>
Q.
LH
Correlation between Specific Conductance and
Chloride and TOC concentrations
CC-l CC-2 CC-3
Specific Conductance (uS/cm)^^Total Dissolved Solids (mg/L)
Chloride (mg/L) --Total Organic Carbon (mg/L)
80
70
00
E
c
a)
40 £
o
u
30 0
o
20 £
c
ro
10 T3
U
Figure 10: Correlation between specific conductance and concentrations found for chloride and total
organic carbon
The measurements for specific conductance correlated with chloride concentrations for the different
sampling dates with low measurements seen during the high flow date and high measurements found in
October. The smallest conductance measured downstream during the high flow sampling date was 122
30


uS/cm; whereas the highest measured conductance was almost seven times that at 815 uS/cm in
October.
The measurements for specific conductance can also be used to estimate total dissolved solids
(mg/L) (Hem 1985). Total dissolved solids can be calculated by multiplying the specific conductance by a
constant derived from data collected by the USGS on the Gila River in 1947 (Hem 1985). The Gila River
study found a well -defined relationship between the two parameters with an uncertainty of + 100
mg/L. The equation KA = S (K: specific conductance, A: constant, and S: total dissolved solids) was
analyzed with Excel to determine the approximate total dissolved solids present per liter in Clear Creek
for each sample site and date (Table 8). The constant used was 0.59, which was the slope of the
regression line (r2 of 0.9604) for the entire Gila River data set. Since these parameters are directly
correlated, total dissolved solids also increased below the point sources and directly correlated with
chloride and TOC concentrations (Figure 10). Dissolved solids within a stream consist of any ion particles
that can pass through a filter with pores around 2 microns in size, which include calcium, chlorides,
nitrate, phosphorus, iron, and sulfur (EPA 2012). The average concentrations for chloride, nitrate, and
Table 8: Estimated Total Dissolved Solids at sites CC-1, CC-2, and CC-3 for all sample dates
Sample Date Site Specific Conductance (uS/cm) Total Dissolved Solids (mg/L)
7/1/2014 CC-1 (Golden) 99 58.41
7/1/2014 CC-2 (Youngfield) 122 71.98
7/1/2014 CC-3 (Prospect) 128 75.52
9/5/2014 CC-1 224 132.16
9/5/2014 CC-2 382 225.38
9/5/2014 CC-3 371 218.89
10/7/2014 CC-1 245 144.55
10/7/2014 CC-2 815 480.85
10/7/2014 CC-3 749 441.91
11/4/2014 CC-1 299 176.41
11/4/2014 CC-2 403 237.77
11/4/2014 CC-3 406 239.54
phosphorus during each sample event at Clear Creek did directly correlation with the estimated
dissolved solids concentration (Figure 11). As discussed previously, the lowest concentrations for all
31


four parameters shown in Figure 11 were during the high flow sampling date in July and the highest
concentrations were seen in October.
The buffering capacity (alkalinity) of the water drastically changed after the point sources, with
the greatest difference seen in October (Table 7). Once again, this could be due to the increased
amount of dissolved material in the water since dissolved organic anions derived from large amounts of
dissolved organic carbon can add alkalinity (Wetzel 2001). The introduction of wastewater from the
point sources also explains why the alkalinity is high at cc-2 and cc-3, since typical wastewater has high
concentrations of nutrients that have acid buffering properties, such as phosphorus (MECC no date).
1200
tXQ
£ 1000
£ 800
o
u
600
TJ
O)
_>
o
400
£ 200
Correlation between Phosphorus, Nitrate and Chloride Concentrations
compared to Total Dissolved Solids
7/1/2014 9/5/2014 10/7/2014 11/4/2014
Sample Date
Total Dissolved Solids Average (mg/L) -*-Nitrate Average (ug/L)
Phosphorus Average (ug/L) -^Chloride Average (ug/L)
3"
- 900 M 3 IS)
- 800 C o fO
700 +J c O) u
- 600 c o u O)
- 500 ;o 'Z o
400 u T5 c fO
- 300 oT +j f0 +J
- 200 iz vT 3
100 O o
0 a i/) o £ Q.
Figure 11: Correlation between phosphorus, nitrate and chloride concentrations compared to total
dissolved solids
This would explain the 183.5mg/L difference between sites cc-1 and cc-2 in October, which also had the
highest levels in total phosphorus (Table 5), total nitrates (Table 5), chloride (Table 6), and TOCs (Table
7). The changes in pH could potentially be affected by the alkalinity especially if carbonate is involved
32


(Table 4). The carbonate species will react with water, which results in the release of the hydroxide ion
making the pH more basic.
CaCCU (s) ^Ca2++ C032
{}
C03z'+H20 ^HC03'+OH-
All sampling dates experienced a gradual increase in pH (more basic) between cc-1 to cc-3. The
pH measurements may have only changed gradually due to the high buffering capacity of the water. If
alkalinity was lower, the water could have been more acidic since dissolved organic matter (a form of
organic carbon that can pass through a 0.45 micron filter) is acidic. The gradual change shows that
whatever reactions may be occurring, are taking place outside of the treatment facility. It is in the best
interest of the treatment facility to keep the pH as neutral as possible, since higher pH values suppress
the effectiveness of disinfection with chloride species and a lower pH can corrode or even dissolve
metals and other substances (USGS, 2014).
33


CHAPTER IV
CONCLUSION
This study highlights the impacts point source discharge can have on an aquatic ecosystem even
after treatment. Unfortunately, this data cannot determine which sources influenced the creek's water
chemistry the most without testing at outfall mixing zones, but an overall understanding of how the
creek reacts to all of the point sources was determined. Analysis confirmed that the water quality
changed and several analytes tested had drastic differences between cc-1 (above the point sources)
compared to cc-2 and cc-3 (below the point sources). Many of these analytes are directly dependent on
each other, so any changes in concentration typically resulted in changes of other analyte
concentrations within Clear Creek. An example of this dependence was seen with the increase in E. coli
concentrations at sites with an observed increase in temperature and nutrient concentrations. The E.
coli present within the creek probably performed both aerobic and anaerobic metabolic processes since
DO was seen to decrease between cc-1 and cc-2. The oxygen was also used for nitrification; this was
seen between sites cc-2 and cc-3 where nitrite concentrations showed an apparent decrease, while
nitrate concentrations showed an increase. Dissolved oxygen stayed relatively stable between cc-2 and
cc-3 even though there was the combination of microbial and chemical processes consuming it, which
could be due to algae producing oxygen during photosynthesis. Flow also played a major role with all of
the analytes measured; where the lowest measurements were found during the high flow sampling date
in July. The October sampling date in particular showed the largest concentrations and the greatest
increase between cc-1 and cc-2 for several of the analytes measured. The exact reason for the greater
increase in analyte concentrations cannot be confirmed, but reduced flow beyond the point sources are
an increase in point source discharge could very well be what caused the differences. All of these
changes can pose serious risks to both environmental and human health, and should be addressed by
making continual improvements to the treatment process for each point source.
34


REFERENCES
Anthonisen, A.C., Loehr, R.C., Prakasam, T.B.S., Srinath, E.G. (1976) Inhibition of nitrification by ammonia
and nitrous acid. Journal Water Pollution Control Federation 48: 835-852
Brewers Association. (2012) Water and Wastewater: Treatment/Volume Reduction Manual.
Campbell, C.S., Ogden, M. (1999) Constructed Wetlands in the Sustainable Landscape (1st ed.) New York,
New York: John Wiley & Sons, Inc.
Chen, Bao-Ming, Wang, Zhao-Hui, Li, Sheng-Xiu, Wang, Gen-Xuan, Song, Hai-Xing, Wang, Xi-Na (2004)
Effects of nitrate supply on plant growth, nitrate accumulation, metabolic nitrate concentration and
nitrate reductase activity in three leafy vegetables. Plant Science 167: 635-643
Crittenden, J.C., Trussell, R.R., Hand, D.W., Howe, K.J., Tchobanoglous, G. (2012). Physical and Chemical
Quality of Water. MWH's Water Treatment: Principles and Design (3rd ed., pp. 60-61). Hoboken, New
Jersey: John Wiley & Sons.
Davies, Trevor D (2006) Sulphate toxicity to the aquatic moss, Fontinalis antipyretica. Chemosphere 66:
444-451
Garcia-Garcia, Gerardo, Nandini, S. (2006) Turbidity mitigates lead toxicity to cladocerans (Cladocera)
Ecotoxicology 15: 425-436
GEI Consultants, Inc. (2015, February) Aquatic Biological Monitoring in clear Creek, Colorado. Retrieved
April 14, 2015.
ftp://ft.dphe.state.co.us/wqc/wqcc/38TriennialReviewRMH_2015/Proponents/MillerCoors%20Exhibits/
MillerCoors%20Exhibit%20E_GEI%20Data%20Report.pdf
Habte Lemji, Haimanot, Eckstadt, Hartmut (2013) A pilot scale trickling filter with pebble gravel as media
and its performance to remove chemical oxygen demand from synthetic brewery wastewater. Jounal of
Zhejiang University-SCIENCE B Biomedicine & Biotechnology 14: 924-933
Hem, J.C. (1985) Specific electrical conductance. Study and interpretation of chemical characteristics of
natural water. (3rd ed., pg. 67). Alexandria, Virginia: U.S. Geological Survey.
Ipeaiyeda, A. R., Onianwa, P. C. (2009) Impact of brewery effluent on water quality of the Olosun River in
Ibadan, Nigeria. Chemistry and Ecology 25: 189-204
Linge, K., Blythe, J., Busetti, F., Blair, P., Rodriguez, C., & Heitz, A. (2013). Formation of halogenated
disinfection by-products during microfiltration and reverse osmosis treatment: Implications for water
recycling. Separation and Purification Technology 104: 221-228.
Janhom, Tansiphorn, Wattanachira, Suraphong, Pavasant, Prasert (2008) Characterization of brewery
wastewater with spectrofluorometry analysis. Journal of Environmental Management 90: 1184-1190
35


Janhom, Tansiphorn, Pavasant, Prasert, Wattanachira, Suraphong, (2010) Profiling and monitoring of
DOM in brewery wastewater and treated wastewater. Environmental Monitoring Assessment 176: 403-
418
Maupin, M.A., Kenny, J.F., Hutson, S.S., Lovelace, J.K., Barber, N.L., Linsey, K.S. (2014, November 5).
Estimated Use of Water in the United States in 2010. United States Geological Survey. Retrieved January
11, 2015, from http://pubs.usgs.gov/circ/1405/.
Mountain Empire Community College (no date) Alkalinity. Water/Wastewater distance learning website.
Retrieved April 14, 2015. http://water.me.vccs.edu/exam_prep/alkalinity.html
Olajire, Abass A. (2012) The brewing industry and environmental challenges. Journal of Cleaner
Production 1-21
Oluah, NS (2007) Anaemia in the freshwater catfish Clarias albopunctatus (Teleostei: Clariidae) exposed
to brewery wastewater effluent. African Journal of Aquatic Science 33: 157-160
Pulido, Cristina, Keijsers, Danny J. H. (2012) Elevated alkalinity and sulfate adversely affect the aquatic
macrophyte Lobelia dortmanna. Aquatic Ecology 46: 283-295
Schumacher, B. (2002). Methods for the determination of total organic carbon (TOC) in soils and
sediments. Ecological Risk Assessment Support Center, NCEA-C- 1282(EMASC-001), 1-23. Retrieved
November 27, 2014, from http://epa.gov/esd/cmb/research/papers/bsll6.pdf
Scott, R. (2012, February 29) Understanding the basic principles of nitrogen. Connecticut Water
Pollution Abatement Association. Retrieved June 22, 2015. https://ctwpaa.org/understanding-the-basic-
principles-of-nitrogen/
Todar, K. (2008-2012) Pathogenic E. coll. Online Textbook of Bacteriology. Madison, Wisconsin.
http://textbookofbacteriology.net/e. coli.html
Unden, G., Becher, S., Bongaets, J., Schirawski, J., Six, S. (1994) Oxygen regulated gene expression in
facultatively anaerobic bacteria. Antonie van Leeuwenhoek 66: 3-22
U.S. Environmental Protection Agency. (2014, December 9). Centralized Waste Treatment Effluent
Guidelines. Retrieved January 11, 2015. water.epa.gov/scitech/wastetech/guide/cwt/index.cfm
U.S. Environmental Protection Agency (2014, September 2). Enforcement and Compliance History
Online. Facility Search Results. Retrieved April 14, 2015. echo.epa.gov/detailed-facility-
report?fid:110033145736
U.S. Environmental Protection Agency (2014, September 2). Enforcement and Compliance History
Online. Detailed Facility Report. Retrieved May 29, 2015. http://echo.epa.gov/detailed-facility-
report?fid=110037263742
U.S. Environmental Protection Agency. (2013, June 21). My Waters Mapper. Retrieved January 11, 2015.
http://watersgeo.epa.gov/mwm/
36


U.S. Environmental Protection Agency. (2015, March 13). Summary of the Clean Water Act. Laws and
Regulations. Retrieved June 23, 2015. http://www2.epa.gov/laws-regulations/summary-clean-water-act
U.S. Environmental Protection Agency. (2014, December 29). Water: Industry Effluent Guidelines.
Industrial Regulations. Retrieved January 11, 2015.
water.epa.gov/scitech/wastetech/guide/industry.cfm#studies.
U.S. Environmental Protection Agency. (2012, March 6) 5.6 Phosphorus. Retrieved June 24, 2015.
http://water.epa.gov/type/rsl/monitoring/vms56.cfm.
U.S. Environmental Protection Agency. (2012, March 6) 5.8 Total Solids. Retrieved June 24, 2015.
http://water.epa.gov/type/rsl/monitoring/vms58.cfm.
U.S. Environmental Protection Agency. (2012, March 6). 5.10 Total Alkalinity. Retrieved May 23, 2015.
http://water.epa.gov/type/rsl/monitoring/vms510.cfm.
U.S. Geological Survey. (2015, May 16). National Water Information System: Web Interface. Retrieved
May 16, 2015. http://waterdata.usgs.gov/co/nwis/inventory/?site_no=06719505.
U.S. Geological Survey. (2015, May 16). National Water Information System: Web Interface. Retrieved
May 16, 2015. http://waterdata.usgs.gov/co/nwis/inventory/?site_no=06718550.
U.S. Geological Survey. (2014, April 8). Water Properties and Measurements. Retrieved January 11,
2015. http://water.usgs.gov/edu/characteristics.html.
U.S. Geological Survey. (2014, March 17). pH Water Properties. Retrieved January 11, 2015.
http://water.usgs.gov/edu/ph.html.
Wetzel, R. (2001). Prologue. In Limnology Lake and River Ecosystems (3rd ed., p. 1). San Diego,
California: Elsevier Academic Press.
37


Full Text

PAGE 1

UNDERSTANDING WASTEWATER EFFLUENT IMPACTS ON THE CLEAR CREEK WATERSHED By KIMBERLY RAE CONWAY B.S., Colorado State University Pueblo, 2010 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment o f the requirements for the degree of Masters of Science Environmental Science Program 2015

PAGE 2

ii This thesis for the Ma s ters of Science degree by Kimberly Rae Conway h as been approved for the Environmental Sciences Program b y Brian Page Chair Fred Chambers Charles Patterson

PAGE 3

iii Conway, Kimberly Rae (M.S., Environmental Science) Understanding Brewery Effluent Impacts on the Clear Creek Watershed Thesis directed by Associate Professor Brian P age ABSTRACT Visible changes to the water quality of Clear Creek are seen below several point sources located in Golden, CO, introduce wastewater discharge into the aquatic environment. There are several regulations wi thin the United States that set up paramet ers for the amount of discharge and the type of effluent allowed to be released into a freshwater system, but fluctuations in water quality can still occur. This study was designed to provide a comprehensive comparison between water quality above a nd below these point sources, specifically a brewery with several prior NPDES permit violations in Golden, Colorado. Testing was conducted at two sites below the point sources and one site above with sampling dates occurring in the July, September, October, and November o f 2014. Analysis confirmed the change between test sites cc 1 compared to cc 2 and cc 3 in water quality with drastic differences in several analyte concentrations Results showed elevated temperatures, higher alkalinity and specific conductan ce measurements, higher concentrations of chloride, phosphor us, nitrate, nitrite, orthophosphate, total organic carbon, and E. coli downstream Analyte concentrations also showed a correlation between stream flow with average concentrations being the lowe st during the July sampling date when stream flow was the highest. Many of these analytes are directly dependent on each other, so any changes in concentration typically resulted in changes of other analyte concentrations within Clear Creek E. coli conc entrations increased downstream where n utrient concentrations were elevated and temperatures were higher. The October sampling date in particular showed the larges t concentrations with th e greatest increase between cc 1 and cc 2 for several of the analyte s measured. The exact reason for the greater increase in analyte concentrations cannot be confirmed, but reduced f low beyond the point sources or an increase in point source dischar ge could very well be what caused

PAGE 4

iv the differences All of these changes c an pose serious risks to b oth environmental and human health and should be addressed by making continual improvements to the treatment process for each point source The form and content of this abstract are approved. I recommend its publication. App roved: Brian Page

PAGE 5

v TABLE OF CONTENTS CHAPTER I. 1 ..12 II. METHODS Field Laboratory Analysis Microbiology 16 17 III. RESULTS AND DISCUSSION ................... ...21 IV. CONCLUSION 35

PAGE 6

vi LIST OF TABLES TABLES 1. A Breakdown of the Areas Where Wastewater is Generated an d the Characteristics of the Wastewater : 2012 ............................................................................................................. ...........8 2. of : 2012 3. Stream flow data in cubic feet per second for the Golden monitoring stations on sampling dates 4. Wa ter Quality Concentrations for specific conductance, dissolved oxygen, pH, t emperature, and E. coli at s ites cc 1, cc 2, and cc 3 on different sampling d ates 5 Nutrient Co ncentrations at s ites cc 1, cc 2, cc 3 for all sampling d ates 6 Anion Con centrat ions for s ites cc 1, cc 2, cc 3 for all sampling d 7 Alkalinity and TOC Con centrations for Sites cc 1, cc 2, cc 3 for all Sampling Da 8. Estimated Total Dissolved Solids at sites cc 1, cc 2, cc 3 for all s ..... .31

PAGE 7

vii LIST OF FIGURES FIGURES 1. Daily Maximum temperature data for Clear Creek in the Outfall 001 mixing zone, August 2006 through June 2012 provided by GEI Consultants, Inc: 2015 2. Weekly average temperature data for Clear Creek at the Outfall 001 mixing zone, August 2006 through June 2012 provided by GEI Consultants, Inc: 2015 3. Monitoring Stations with Reported Water Quality R esults on MyWaters: 2013 4 Location of sample site CC 1. Black star shows approximate location where water samples were taken. Images taken during the first sampling date in July 5. Location of sample site CC 1. Black star shows approximate location where water samples were taken. Images taken during the first sampling date in July 6. Location of sample site CC 1. Black star shows approximate location where water samples were taken. Ima ges taken during the first sampling date in July 7. Indirect correlation between DO and E. coli concentrations during the October sampling d ate .23 8 Average nitrate, nitrit e, orthophosphate, and phosphor us concentration s compared to Clear Creek streamflow data taken at the Golden monitoring station 9 10 Correlation between specific conductance and concentrations found for chlor ide and total 11. Correlation between phosphor us, nitrate and chloride concentrations compared to total 32

PAGE 8

1 CHAPTER l INTRODUCTION s, have tended to ignore humans and their (Wetzel 2001, pg. 1) If this is the case then there is a great deal of work that needs to be done to understand the effects humans have on these ecosystems. Human usage of freshwater systems is estimated every five years by t he US Geological S urvey (USGS) ; t hey compile county, state, and n atio nal water withdrawal and usage data for a number of water use categories The last Total Water Use study don e in 2010 found that the average amount of water used per day in the United States was 355 billion gallons (USGS 2014) This rate of withdrawal i s actually the lowest since prior to 1970, which shows that conservation efforts have been made and are providing immediate benefits A small fraction of this usage is for industrial purposes ; the amount withdrawn for self supplied industrial use wa s 15. 9 billion gallons per day ( which was mainly supplied from surface water sources ) (USGS 2014) Industrial use may have a mino r influence on water withdrawal but can have significant impacts on the environment due to the reintroduction of the water into th e source after being used. wastewater on rivers has attracted a lot of attention worldwide because of its overwhelming Industrial effluent within the United Sta tes is regulated through county, s tate, and federal guidelines; t here are approximately 100 different guidelines set up by the US Environmental Protection Agency (EPA) that pertain to specific industries and the effluent these industries produce (EPA 2014 ) The differentiation of these guidelines is necessary for the purpose of addressing a wide range of analytes that have the potential to be released into the environment and potentially cause harm. This pollution can include hazardous and non hazardous m aterial that can have an adverse effect on flora and fauna that inhabit regions downstream of where reintroduction of wa stewater occurs. Specific

PAGE 9

2 analytes within the effluent may even become harmful to human health if this water downstream is used for re creation or for drinking water. These concerns are the driving force for the implementation and enforceme nt of water treatment standards; a ll wast ewater must go through some form of treatment process before being reintroduced into a freshwater ecosystem. The amount of treatment will not only depend on these standards but the actual treatment process itself; t reating industrial wastewa ter can be an expensive process, especially if the water was used for multiple purposes and accumulated a large variet y of impurities. It may be difficult to remove all impu rities and have the water meet regulatory standards after treatment I t is extremely important to understand if the waste treatment process is thorough enough to ensure e nvironmental and human safety; m a ny bodies of water within the United States remain above pollution levels co nsidered safe for the ecosystem (Brewers Association 2012 ) If an ecosystem is showing visible adverse effects from the introduction of wastewater effluent, an investigation needs to occur in order to profile what analytes are causing stresses on that ecosystem. For that exact reason, a study was conducted on the Cle ar Creek Watershed ( which shows visible changes in its ecosystem within a short stretch of the creek traveling throu gh Golden, Colorado ) There are several point and non point sources of discharge and runoff that could be causing these changes. The EPA Enforcement and Compliance Online (ECHO) recognizes at least six point sources with four of these sources having Nat ional Pollutant Discharge Elimination System (NPDES) permits regulating outfalls discharging wastewater into the creek (EPA 2014) The analytes being monitored vary for each permit since the chemical make up of the wastewater varies depending on the busin ess producing the waste. The effects a point source may have on an ecosystem is more predictable since there is a clearer understanding of what chemicals are present within the wastewater compared to non point where the source cannot always be identifie d. Non point sources include storm water runoff that has accumulated contaminants such as roadway debri s metals, dirt, or magnesium c hloride or runoff

PAGE 10

3 coming from nearby farms or suburbs that has been contaminated with fertilizers, manure, or pesticides/h erbicides. Not only is it challenging to know what has accumulated in non point source runoff it can be difficult to predict the effects it will have on the ecosystem and how to address the impacts. In the case of Clear Creek the non point sources may have less of a comparative impact on the water quality compared to the six point sources discharging into the creek F our of these sources have minor or major NPDES permits to monitor for various analytes within the wastewater (EPA 2014 ) One source is a metal container production facility that has a minor NPDES permit, which monitors a variety of analytes in the discharge as well as the total recoverable amount during the treatment process. Four outfalls are present at this facility that discharge into the creek and are monitored on a monthly basis Some of the wastewater parameters monitored for this permit include temperature, dissolved oxygen, pH, total suspended solids, nutrien ts such as nitrite and nitrate, anions like fluoride, and metals like iron, nickel, and lead. This facility has had violations in the past, but were only for reporting issues not exceedances in analyte concentrations. Another facility ( involved in ceram ics production) discharges wastewater into the creek, which is monitored through two minor NPDES permits. This facility also had violations with permit reporting but no analyte concentration violations. The effluent discharged is monitored for cadmium, chr omium, copper, cyanide lead, nickel, pH, silver, total toxic organics, total suspended solids, zinc and flow o n a monthly basis. A main point source introducing waste water into the creek is a brewing company that treats its own wastewa ter and rei ntroduces that effluent back into the creek This company has two facilities located at different locations along Clear Creek that have their own separate NPDES permits (EPA 2014) Differences in water quality of Clear Creek become apparent tra veling downstream of Golden, below the brewery, into Wheat Ridge. Some variances in the ecosystem that are noticeable without testing include temperature and flow L ower stream flow has been observed below the brewery co mpared to stream flow above; this

PAGE 11

4 production purposes. The reduced flow can also be credited towards two canals slightly above the brewery that divert water to Standley Lake. The overall changes in water qualit y, as well as temperature, will inevitably change when less water is traveling downstream of any of the point sources located along the creek The more water allowed to stay in the creek, the more the wastewater will be diluted and analyte concentrations ( per liter of water ) will be lower. A higher volume of water will also help equilibrate the wastewater temperature to the natural temperature of the stream. Less volume can encourage more heat transfer through the water from radiation, causing the entire stream to heat up. U nder the CWA and the NPDES permit program, any point s ource that discharges wastewater into water must ensure that the temperature of the effluent matches the temperature of the stream, or should only influence the overall stream temp erature by a few degrees centigrade (EPA 2014). However, these standards for temperature changes are relatively new, so limitations on temperature changes have not yet been implemented in the NPDES permit issued to the brewing company. There are temperat ure standards set up by the Colorado Department of Public Health and Environment for certain segments of Clear Creek throughout the year, which can be compared to temperature data recorded by GEI Consultants, Inc for the annual Aquatic Biological Monitorin g of Clear Creek, Colorado (2015). One particular segment mon itored is at one of the brewery s outfalls; the daily maximum and weekly average temperature standards were both exceeded during the winter seasons between 2006 and 2012 (Figure 1 & 2). There w ere also frequent exceedances of the coldwater criteria in the summer for the daily temperature standards seen between 2006 and 2012. The water shows an apparent warming in temperature, not only to the touch, but steam can be seen rising from the water in the early morning in areas below the discharge, which is not observed above. The major concern with the exceeding of temperature standards for the creek is the effect it may have on the aquatic biota within the creek. A consultant with the brewery did p repare a report on the effects the temperature changes

PAGE 12

5 have on aquatic biota in Clear Creek and concluded there were no negativ e impacts (GEI Consultants, Inc 2015). Figure 1: Daily Maximum temperature data for Clear Creek in the Outfall 001 mixing zon e, August 2006 through June 2012 (GEI Consultants, Inc 2015) Figure 2: Weekly average temperature data for Clear Creek at the Outfall 001 mixing zone, August 2006 through June 2012 (GEI Consultants, Inc 2015)

PAGE 13

6 Water quality monitoring of Clear Creek occur s at several locations by the C ity of Golden, as well as the S tate of Colorado through the Colorado Department of Public Health and Environment (CDPHE) the EPA, the Clear Creek Watershed Management Agreement Monitoring Program, and the brewery itself at d gives information on the particular permits the brewery has and whether it has had any compliance iss ues in the past four years (EPA 20 15). NPDES permit s have been is sued to the brewery under the CWA, which monitors temperature, pH, flow, oil and grease concentrations, and total suspended solids (TSS) concentrations at five outfalls belonging to the upstream facility. The downstream facility has a minor NPDES permit t hat monitors the settling pond where discharge is held Chlorine, flow, oil and grease, total suspended solids, and pH are monitored in this settling pond. In the past four years the brewery has had seven violations regarding TSS concentrations and pH at three of the five monitored outfalls. The NPDES permit requires the pH of water at the outfalls to be between 6.5 and 9 and the TSS concentrations to not exceed 45 mg/L for the average weekly limit and 30 mg/L for the average monthly limit. The largest TSS concentration was 71 mg/L, which was a recorded weekly average for the last week in June of 2014. This particular violation was found at Outfall 008, which experienced one other weekly average violation in June of 2013 (61 mg/L) and one monthly avera ge violation in June of 2014 (32 mg/L). The other two outfalls that experienced TSS and pH violations were Outfall 006 and 011. Outfall 011 reported a TSS violation for an average weekly concentration of 54 mg/L in June of 2014 and a pH violation of 6.4 at the same time. Outfall 006 also had a TSS and pH violation in June of 2014 with a reported average weekly TSS concentration of 60 mg/L and a pH of 6.3. Only one other parameter monitored under the NPDES permit has a set maximum limit, which is for oil and grease. Concentrations cannot exceed 10 mg/L and any visible detection on the surface of the water would be in violation. Temperature and flow are monitored on a quarterly bases, but set maximum and minimum standards are not provided within the per mit.

PAGE 14

7 Data on the entire watershed can be found online & E nvironmental Results web page Water impairments, monitoring stations, and other EPA water data can be accessed through an interactive map called MyWater s Mapper (EPA 2013). From this map, it can be determined that there are four monitoring stations directly below the effluent outfall. These specific monitoring stations are in mixing zone areas and are ran by the Colorado Department of Public Health &

PAGE 15

8 Environment. The d ata for the outfall stations is not provided by MyWaters, but water quality monitoring results are provided for stations above the brewery and at stations below the mixing zones. Four stations in particular may provide the best results when determining the effects the brewery has on the creek, since these stations are collecting water that has either not been influenced by the brewery or has been influenced, which will provide a better grasp of the overall water quality changes of the s tream. The location of these sites provided by the EPA are shown in the Figure 1. Howe ver, the majority of the data cannot be used for this comparison, since the sampling did not occur in the same time frame at their respective sites; this is an issue sin ce any number of factors could have influenced the water quality between sampling dates, making any direct comparisons between sites difficult. (Chart source: Brewers Association 2012) Source Operation Characteristics Mash T un Rinsing Cellulose, sug ars, amino acids. ~3,000 ppm Biochemical Oxygen Demand (BOD) Lauter Tun Rinsing Cellulose, sugars, spent grain. Suspended solids (SS)~3,000 ppm, BOD ~ 10,000 ppm Spent Grain Last running and washing Cellulose, nitrogenous material. Very high in SS (~30,000 ppm). Up to 100,000 ppm BOD Boil Kettle De watering Nitrogenous residue. BOD ~2,000 ppm Whirlpool Rinsing spent hops and hot trub Proteins, sludge and wort. High in SS (~35,000 ppm). BOD ~ 85,000 ppm Fermenters Rinsing spent hops and hot trub Yeast SS ~ 6,000 ppm, BOD up to 100,000 ppm Storage t anks Rinsing Beer, yeast, protein. High SS (~4,000 ppm). BOD ~80,000ppm Filtration C l eaning, start up, end of filtration, leaks during filtration Excessive SS (up to 60,000 ppm). Beer, yeast, proteins. BOD up to 135,000 ppm Beer spills Waste, flushing et c 1,000 ppm BOD Bottle washer Discharges from bottle washer operation High pH due to chemical used. Also high SS and BOD, especially thru load of paper pulp. Keg washer Discharges from keg washing operations Low in SS (~400 ppm). High BOD Miscellaneous Discharged cleaning and sanitation materials. Floor washing, flushing water, boiler blow down etc. Relatively low on SS and BOD. Problem is pH due to chemicals being used.

PAGE 16

9 T he water use of the brewing industry is becoming a global concern due to the extensive amount of water required in the brewing process and the large amount of polluted effluent produced afterward. On average, b re weries within the United States discharge 70% of their inco ming water a s effluen t (Brewers Association 2012 ) During the brewing process, large quantities of water are used for the production of beer itself, as well as for washing, cleaning and sterilizing of various units (tanks, bottles, machines and floor s) during production (Olaj ire 2012, Ipeaiyeda and Onianwa 2009). The wastewate r should be subjected to a n on site pre treatment process in order to reduce the strength of the effluent being dischar ged. Discharge limits are applied more often to brewers due to the high organic nat ure of the wastewater; t his organic material can cause a rapid depletion of oxygen in the surface water the effluent is being introduced into which can have negative impacts on living species and biodiversity within that waterway (Brewers Association 2012 ) Discharge and strength of effluent are monitored in the United States under the Clean Water Act (CWA) (EPA 2015 ) Some of the rules under the CWA (EPA 2015) including Effluent Limitations Guidelines, Pre treatment Streaming Rule, NPDES Permit Program Sewage Sludge Rule, and Total Maximum Daily Load and Impaired Water Rules (EPA 2014) can have a direct or indirect impact on brewery operations (Brewers Association 2012 ) If these regulations were not in place, brewers may be less inclined to treat the wastewater and remove analytes that could cause harm to the ecosystem. Table 2: of (Chart source: Brewers Association 2012) Parameter Typical Range Biochemical Oxygen Demand (BOD) 600 5,000 ppm Chemical Oxygen Demand (COD) 1,80 0 5,500 ppm Nitrogen 25 80 mg/L Phosphor us 10 50 mg/L pH 3 12 (Standard Units ) Total Suspended Solids 200 1,500 ppm

PAGE 17

10 Research on the effects of untreated brewery wastewater on the ecosystem have been c onducted in countries where water quality regulations are not in place. One such study looked at the impacts the effluent had on the Olosun R iver located in Ibadan, Nigeria ( Ipeaiyeda and Onianwa 2009). Quantifying these impacts was done by taking two s amples above the brewery and five samples at varied distances below the brewery. These samples were analyzed for the following parameters ( Ipeaiyeard and Onianwas 2009) : ammonia (NH 4 ) (nesslerisation colorimetric method) biochemical oxygen demand (BOD) ( chemical oxygen demand (COD) (potassium dichromate oxidation and titrimetry) chloride (CL ) (mercurimetric titration) nitrate (NO 3 ) (phenodisulphonic acid colorimetric m ethod) sulfate (SO 4 2 ) (turbidimetry) total solids (TS) total suspended solids (TSS) (gravimetry) turbidity (turbidimetry) S ample s were also tested for Cadmium (Cd), Calcium ( Ca), Chromium (Cr), Cobalt (Co), Copper (Cu), Lead (Pb), Nickel (NI), Zinc (Zn) u sing atomic absorption spectrophotometry (Perkin Elmer model 2380). The analysis conducted showed the average pH values upstream were much higher than the corresponding river (Ipeaiy eda and Onianwa 2009, pg. 196). The DO also decreased due to an increased consumption rate from the higher organic load of matter within the effluent. This con sumption rate will increase due to a stimulated algae growth caused by an increa se in phosphate levels (Ipeaiyeda and Onianwa 2009). They NH 3 Ni, Zn, Cr, Co, Cu, Cd and Pb levels of

PAGE 18

11 samples increased considerably and were significantly above the recommended water quality s ff luent flows down river. The BOD and COD were also found to have increase d high organic load is found in wastewater from ri(n)sing of bottles and mixing tanks, which consequen tly There was also evidence of a build up of heavy metals, particularly Ni, Zn, Cu and Pb within crops surrounding the river which could pose a risk to humans when consumed (Ipeaiyeda and Onianwa 2009). The presence and/or change in levels of certain chemicals within a body of water caused by anthropogeni c activities can create a cascade of adverse effects towards the aquatic biota. N S Oluah studied the effects of brewery wastewater effluent on the freshwater catfish Clarias albopunctatus (2007). This species of catfish was tested in a c ontrolled environment thus variations observed in the catfish were entirely dependent on the concentration of polluted effluent they were exposed to. It was found that the groups of catfish exposed to increased amounts of effluent showed both a decrease in haemoglobin concentr ation in C. albopunctatus exposed to brewery wastewater is indicative of impaired oxygen carrying capacity in the fish (pg. 159). Studies on other chemicals within the brewery effluent have shown depressed growth in leafy vegetation from exposure to incr eased concentration of 4 2 and high alkalinity in surface water negatively affected L obelia dortmanna plants causing mortality, decreasing biomass and reducing actual photosynthetic efficiency (Pulid o et al. 2012, pg. 291). T.D. Davies also found that incre ased levels of sulf ate caused reduction in final shoot length, reduction in final dry weight, and reduction in chlorophyll

PAGE 19

12 a and b in the aquatic moss species Fontinalis antipyretica (2006). Elev ated levels of lead were found to significantly reduce the average lifespan of certain species of Cladocera (Garcia Garcia 2006). Research Purpose This study was designed to provide a comprehensive c omparison between water qualit y above and below the br ewery and other point sources located in Golden The issues associated with the data provided by the EPA were addressed by testing the sites on the same day s o a more accurate comparison could be made. The chemical and statistical analysis performed in t his study was modeled after the research done by Ipeaiyeda and Onianwa (2009) on the Olosun River. The re are some key difference between this study and the one on the Olos un River, including the time frame the samples were taken and the analysis of heavy metals. This study was performed over a period of one year and analysis of heavy metals was omitte d due to budget restrictions for this study. The EPA Region 8 laboratory assisted in retrieving the samples and sample analysis at their laboratory in Golden, Colorado. This study also presents the impa cts of treated effluent from a four different point sources, not just the brewery wastewater, on water quality instead of untreated effluent as was seen with the Olosun River study.

PAGE 20

13 CH APTER ll METHODS Field Sampling Field sampling occurred at three different sites on July 1, 2014, September 5, 2014, October 7, 2014, and November 4, 2014 The three sites used were close to thre e CDPHE monitoring stations that provide water quality data. The fourth station highlighted in orange was inaccessible so no sampling occurred in that area. The first sampling site was located near the monitoring station highlighted in red. This was near the parking lot at Prospect Park in Wheat Ridge, Co lorado ; t he sampling ID for this

PAGE 21

14 location is designated a s cc 3 (Figure 4) Th e next water samples were taken close to the monitoring station highlighted in green under the Youngfield St. Bridge. The sampling ID for this location i s cc 2 (Figure 5) The last sampling site was near the monitoring station h ighlighted in blue. This was near the parking lot of the Golden Library and Clear Creek History Park. The sampling ID for this location i s cc 1 (Figure 5) Two bulk sample s were taken at eac h site and split into five different containers that were later tested for alkalinity, anions, total nitrogen (TN) total phosphor us (TP) total organic carbon (TOC) nutrients, and E. coli Each container consisted of its own laboratory ID # ; a lkalinity and anions were tested using the same 250mL container which was stored at 4C ; TN and TOC were tested using the

PAGE 22

15 same 125mL container with HCl preservative ; TP was tested using its own 125mL container with H 2 SO 4 preservative ; nutrients were tested from the 60mL container that was frozen before analysis The last container was a 120mL sealed, plastic bottle for E. coli analysis. These samples were taken by wading out into a safe part of the stream where a collapsible, sterile, square container (cubitai ner) could be completely immersed under the water. The cubitainer was triple rinsed with water from the creek before a final sample was collected This water would then

PAGE 23

16 be used to fill four of the five small plastic containers. Water was immediately col lected in the sea led, sterile plastic bottle after collecting water in the cubitainer in the same location Laboratory testing only required 100mL ; excess water was poured out until the meniscus was at the 100mL line. Once all the smaller containers were filled for that location, they were placed into a small bag and put on ice in a cooler for storage while out in the field. Temperature, pH, c onductivity, and DO were teste d on site using a n In Situ TROLL 95000 This meter was placed a couple of feet up stream from where water samples were taken to ensure that any disturbances from wading did not affect the meter. The meter was placed at the bottom of the creek where it could be completely immersed under the water. Mea surements were then recorded in field notes and saved on the meter for every site b efore removing the meter from the water. Laboratory Analysis Microbiology Testing All water samples were submitted to the EPA region 8 laboratory for testing The E. coli samples were tested in the microbiology lab at the region 8 laboratory within 24 hours of submission. These samples were te sted using the Colilert Enzyme Substrate method with Quanti Trays (IDEXX, Inc.). This process starte d by homogenizing the samples using a vortex, foll owed by pouring off any excess water over the 100mL mark. One ampoule of Colilert 18 media was then added to each container and mixed by tightly recapping and vortexing again Th e contents of this container were then poured into a 96 well Quanti Tray/200 0, which is then sealed by placing the tray in the Quanti Tray/2000 rubber insert and feeding it into the Quanti Tray sealer. Once each tray is sealed and labeled with its proper ID, it is loaded into an incubator (35 + 0.5C) for 18 ( + 4) hours. Standa rd Quanti Tray/2000 comparators ar e also used as a color reference for the creek samples in order to determine if coliform bacteria and E. coli are present. The reference samples are made by filling 4 sterile containers with 100mL of ultra pure (~18 ) autoclaved water. One ampoule of Colilert medi a is then added to each container and mixed by tightly recapping and vigorously

PAGE 24

17 shaking 25 times. Four test tube vials containing E. coli Klebsiella pneumo niae P seudomonas aeru gi nosa and a blank are then paired with one of the containers. Each container is inoculated with its designated test tube by dipping a plastic loop into the test tube and transferring the liquid film formed across the plastic loop into its respective container. The contents of the loop are spread by swirling the loop in the ultra pure water within the container. Once all controls are inoculated, the contents o f each container are then transferred into a Quanti Tray/2000. These trays are incubated alongside the sample Quanti Trays for the same period of time. I f the colors between the sample Quanti Tray/2000 and the comparator are indistinguishable after 24 hours then both must be placed back into the incubator for no more than the maximum 28 hour incubation period and read at a later time The res ults of the samples cannot be ac cepted until the controls and blank are verified. In this case the blank should remain colorless and not fluoresce, the E. coli QC sample should turn yellow and fluoresce under 366 367 nm UV light, the K. pneumonia QC sample should turn yellow and not fluoresce, and the P. aeruginosa QC sample should not turn yellow or fluoresce. If the controls perform as expected then the samples may be pulled from the incubators and analyzed A fter incubation, the color of any well in the sample Quanti Tray/2000 s is compared to the wells of the comparator. If the sample wells have a yellow color that is equal to or greater than the color intensity of the comparator wells, then the sample is total coliform positive A ll of the small and large wells are then totaled and recorded as most probable number (MPN) using the 96 well Quanti Tray MPN table E. coli positive is determined by comparing the fluorescence of any of the wells in the sample trays to the comparator wells. If the sample trays have a fluorescence that is equal to or greater than the intensity of the comparator wells then the sample is E. coli positive. The total number of wells t hat fluoresced are then added and recorded as MPN using the same MPN ta ble Chemical Analysis Alkalinity is a measure of the capacity of water to neutralize acids (EPA 2012). The testing is performed using a DL 50 autotitrator that analyzes 50 mL of sample from the 250

PAGE 25

18 mL LDP E sample bottle. The DL 50 autotitrator mea sures alkalinity by measuring the amount of sulfuric acid needed to drop the sample pH to 4.2. Analysis can begin after the machine is calibrated and the sample table is updated on the DL 50 software. Each s ample along with a blank and any controls take n, can then be loaded onto the sample charger according to the sample table Once measurements are complete and the results are recorded in the sample table t he alkalinity re port can be printed and man ually added into the Laboratory Information Management System ( LIMS ) for review. Before the r esults for alkalinity can be reviewed an acceptance criteria m ethod must be performed ( which should give a pH greater than 4.0 ) The quality control samples are also used to verify the accuracy of the standardizati on and to identify any matrix effects; i f errors are seen with either the quality control samples or the acceptance criteria method then the alkalinity results cannot be used. Water collected in the same 250 mL LDPE sample bottle can be used to test for anions. Testing must occur within 48 hours of collection for nitrite N, nitrate N and orthophosphate P, and within 28 days of collection for fluoride, chloride, bromide, and sulfate. The water used for this test must be filtered using a 0.45 micron filter prior to analysis. Dilutions of the sample may be required if anion concentration s are too high or if the pH is 4.5 or lower. After filtration, 5.9 mL of well mixed sample can be added to a 10 mL polyvial and placed into the Dionex ion chromatogra ph model ICS5000 autosampler with standards and controls for analysis. If the response for the peak exceeds the working range of the system, the sample will be diluted with an appropriate amount of reagent water and reanalyzed. Results are generated auto matically through the Chromeleon software and later transferred to the LIMS for review. Water collected in the 60 mL LDPE bottle can be analyzed for the presence of Nitrate N, Nitrite N and Orthophosphate P. This sample should be analyzed within 48 hours of collection. If this is not possible then the sample can be frozen and analyzed within 28 days. Prior to analysis, EDTA is added to eliminate interference from metals and turbid samples must be filtered using a 0.45 m filter. Analysis

PAGE 26

19 begins by pow ering on all u nits of the instrument and verifying that the instrument is properly communicating with the Omnion software. Once equipment calibration is complete and the software is updated with the new sample table, sample analysis can start by loading t he sampler tr a y with the standards, QC controls, and samples. Results are automatically generated through the software by comparing the sample absorbance to the calibration curve (created for the known standards), giving a value for each sample If nitra tes and nitrites need to be recorded separately the Cd reduction column must be disconnected and the nitrite N can only be measured. Then this value can be subtracted from the (Nitrate + Nitrite) N result giving the nitrate N value. All data can then be uploaded into LIMS and formatted into an Excel file for review. Water collected in the 125 mL plastic container with HCl preservative was used for the TOC and TN analysis. The Shimudzu TOC/TN analyzer is used to measure TOC or TN, which can be chan ged by changing the furnace temperature (680C for TOC analysis or 720C for TN) The sample must be transferred into the appropriate glass sample bottle that fits into the Shimudzu sample tray. Once the is updated with the correct s ample table the samples can be loaded into the sample tray and the analysis can begin. Results will be generated automatically by the software. Samples whose concentrations e xceed the calibration curve must be diluted and rerun; a fter QC samples have bee n verified, the results for the samples can be transferred into the LIMS and into an Excel spreadsheet for review. The last sample analysis done withi n the lab is the total phosphor us (TP) test. This uses the water collected in the 125 mL containe r with H 2 SO 4 preservative. This water will be transferred into a disposable 16 X 100 mm glass culture tube and digested Digestion occurs by adding 2.0 m L of digestion reagent into the glass culture tube and placing it into an All American electric press ure steam sterilizer for a half hour The temperature of the pressure steamer should be 250 F at 15 psi. Once the sample has been taken out of the pressure ste amer the Lachat QC8500 Flow Injection Autoanalyzer can be set

PAGE 27

20 up with the appropriate detectio n parameters for phosphate P analysis including a n 880 nm filter and heating coil at 37 C, with the pump tubes in their respective reagent containers. The reagents should be allowed to flow until a stable baseline is obtained. Analysis can begin once the sample tray is loaded with the standards, controls and samples. Results are found by comparing sample absorbance to the standard calibration curve of the known standards. These results are automatically generated by the software, which are the n transfer red into LIMS and an E xcel spreadsheet for review.

PAGE 28

21 CHAPTER lll RESULTS AND DISCUSSION C hanges in wa ter quality of Clear Creek between the sampling l ocations are apparent. Tables 4 through 7 show that analyte concen trations were a ffected in the creek along the stretch traveling through Golden. Whether these changes were caused by the brewery, the other point sources, or even the non point sources cannot be established, but some of the differences in analyte levels a re great enough to imply that the changes are from anthropogenic activity. Concentrations for many of the parameters tested were also greatly affected by the stream flow. Flow was not measured at the specific sites due to safety issues, but flow data fro m the USGS w as retrieved from one statio n located in Golden (Table 3) (USGS 2015) Table 3: Stream flow data in cubic fe et per second for the Golden monitoring stations on sampling dates USGS 06719505 CLEAR CREEK AT GOLDEN, CO Date Discharge (cubic fee t per second) 7/1/2014 948 9/5/2014 182 10/7/2014 87 11/4/2014 83 Flow rates were the highest during the first testing date in July for both monitoring stations, with Golden showing an average flow of 948 cubic feet per second (cfs). This high flow (dilution) would account for why concentrations for many of the parameters tested were lower than concentrations measured during the other three sampling dates. The prevalence of E. c oli was one such parameter affected by stream flow; having the lowest colony count and changes during the July sampling date. Unfortunately there are no f low gaging stations below the point sources so only visual comparisons could be made on the changes in flow. A basic trend seen for all sampling dates was the decreased flow at the two test ing sites downstream; a s mentioned before, the decrease in flow could be from water being withdrawn from the canals that distribute water from the creek to Standley Lake or the brewery that uses the water for production. The differences in flow between sa mpling dates and between

PAGE 29

22 sampling sites may have also had an effect on temperature. The downstream sampling sites had elevated temperatures compared to site cc 1 which could have been due to the reduced flow or the increase in direct sunlight from being out of the canyon The increase in temperature could also be ca used by the brewery effluent, since past temperature measurements in mixing zones recorded values higher than the daily maximum standard for the creek (Figures 1 and 2) Water temperatures tended to stay elevated downstream, which also could have resulted from the shallow flow. Other point sources may have added effluent with temperatures higher than the creek temperature, which could also allow raised temperatures to persist. Table 4 : Water Quality Concentrations for Specif ic Conductance, DO pH, Temperature, a nd E. coli at Sites cc 1, cc 2, and cc 3 on Different Sampling D ates All four sampling dates did show a pattern in temperature increase between sampling sites, with the sites below the brewery being higher. The largest diffe rence between two sites was cc 1 (above) and cc 2 (belo w) in November, which had approximately an 8C increase after the brewery mixing zone (Table 4 ) The smallest increase was in July with a d ifference of only <2C from cc 1 to cc 2 and cc 3 These temperature changes were associ ated with the amount of E. coli found at each site The presence of E. coli is an indicator of recent sewage or animal waste contamination (EPA 2013). The increase in E. coli at the sites downstream could either be caused by an increase in runoff wi th waste contamination

PAGE 30

23 or a combination of higher temperatures and an increase in nutrients. All strains of E. coli are facultative, aerobic and anaerobic, gram negative bacteria that grow best in temperatures around 37C (Todar 2008 2012 ) The warmest t emperatures were recorded in September at 18.33 and 18.21C. The two locations where these temperatures were recorded were cc 2 and cc 3 which experience d the highest counts of E. coli with >2420 MPN/100mL (TNTC for 96 well Quanti Tray/2000). The drasti c spike in E. coli between the September sampling date and the July sampling date was not expected so dilutions were not performed for a more accurate MPN count. Nutrient concentrations appeared to trend with the amount of E. coli present between samplin g sites, but not between sampling dates. E. coli concentrations increased downstream wh ere nutrient concentrations were also elevated. Under anaerobic conditio ns, E. coli will grow by utilizing nitrates and nitrites as final electron acceptors for respir atory electron transport processes (Todar 2008 2012). The sampling date with the highest presence of E. coli did not correlate with overall highest nutrient concentrations and instead correlated with the highest temperatures recorded. E. coli can also g row aerobically by respiration using O 2 which could explain the inverse correlation between DO and E. coli concentrations (Unden et al. 1994) Figure 7 : Inverse correlation of DO a gainst E. coli concentrations and Temperature during the October sampli ng date 0 10 20 30 40 50 60 70 80 90 100 0 2 4 6 8 10 12 14 16 18 CC-1 CC-2 CC-3 E. coli Concentration (MPN/100mL) Dissolved Oxygen Concentration (mg/L) and Temperature ( C) Sample Sites Inverse Correlation of DO against E. coli Concentrations and Temperature Dissolved Oxygen (mg/L) Temperature (C) E. coli (MPN/100 mL)

PAGE 31

24 The only measurement taken showing an ap parent decrease downstream was dissolved oxygen (DO) (Table 4 ) This correlates with the increase in E. coli growth which provides some explanation as to the reduced levels in DO (Figure 7 ) Similar to o s highest priority level for growth and energetics is through aerobic metabolism, which can be triggered by the presence of O 2 (Unden 1994). E. coli has the regulator mechanisms to then change its metabolic pat hway to anaerobic once the amount of available O 2 decreases, therefore turning to nitrate as the main nutrient for growth (Unden 1994). If DO concentrations were completely dependent on the amount of E. coli present, the n the sampling date in Sept ember with the highest counts of E. coli should have seen the lowest concentrations of DO, which was not the c ase. Other aquatic biota such as algae, could also have a direct effect on the amount of DO ; t he increase in nutrients downstream may be promoti ng algae growth, which would result in an increase in DO since algae release oxygen during photosynthesis. This w ould explain why there was a slight increase in dissol ved oxygen between cc 2 and cc 3 where nutrients were high. Table 5: Nutri ent Concentra tions at Sites cc 1, cc 2, cc 3 for all Sampling Dates All nutrient concentrations increased at sampling sites below the brewery compared to the testing site above (Table 5 ) The nutrient that showed th e most drastic increase was nitra te, with the most significant difference being around 800 ug/L. The nutrient with the most notable change was nitrite since levels were actually be low detection limits upstream of all the point sources at cc 1 for all

PAGE 32

25 sampling dates (t his trend was also seen with pho sphor us concentrations for three of the sampling dates ) The drastic change in nutrient concentrations could have been caused by any of the point sources or a combination of them and non point sources from runoff A nthropogenic influence s on the cre ek can cause a variety of issues or changes that can offset the natural balance of aquatic life. The direct relationship between E. coli temperature and all of the nutrients is one such example that shows how this balance can change if a viable nutrients are provided. Nitrite showed a decrease in con centration between cc 2 and cc 3 with t he largest decrease in concentration seen between cc 2 (315ug/L) and cc 3 (201ug/L) in October. This was on the same day as the largest increase in nitrite between c c 1 (<5.0ug/L) and cc 2 As nitrite concentrations decreased between sites cc 2 and cc 3, nitrate concentrations increased. The process of nitrification best explains why there is an inverse correlation between nitrite and nitrate concentrations (Scott 2 012) With the assistance of a nitrite oxidizer such as Nitrobacter and oxygen as the oxidizing agent, nitrite will be oxidized into nitrate (Figure 0 2 4 6 8 10 12 0 100 200 300 400 500 600 700 800 900 1000 7/1/2014 9/5/2014 10/7/2014 11/4/2014 Dissolved Oxygen Concentration (mg/L) Nitrate, Nitrite, Orthophosphate, Phosphorus and Dissolved Oxygen concentrations compared to Clear Creek streamflow Golden Streamflow (cfs) Nitrate Average (ug/L) Orthophosphate Average (ug/L) Nitrite Average (ug/L) Phosphorus Average (ug/L) Dissolved Oxygen (mg/L)

PAGE 33

26 9 ) This would explain the decrease in nitrite and increase in nitrate between the two testing sites belo w the point sources. When nitrification occurs in a closed system, DO levels decrease and the pH become s more acidic since for every two nitrate s produced, four hydrogens are produced and one oxygen is consumed (Figure 9 ) Instead, the pH became more bas ic and the DO levels stayed relatively constant between the downstream sites. Depressed pH s ( more acidic) will only occur if there is n ot a sufficient alkalini ty concentration (Campbell and Ogden 1999) and if nitr ite concentrations were increased from th e oxidation of ammonia (Figure 9 ) (Anthonison et al. 1976). In the case of the testing sites below the point sou rces in Clear Creek, alkalinity was relatively high (Table 7) which would explain why pH was not affected by the nitrification process. The d ecrease in DO concentrations from nitrification could have been offset by oxygen producing algae growing in the water. Figure 9 : Process of Nitrification The nutrient concentrations were much more diluted during the high flow sampling date in July whic h could potentially reduce nutrient loading on the ecosystem (Figure 8 ) Flow did not correlate completely with the average nutrient concentrations since the results for October showed the highest concentrations for all of the nutrients. It should be not ed though, that the difference between cc 1 and

PAGE 34

27 cc 2 concentrations for all nutrients was greatest during the October sampling date with relatively large increases between the two sites. This could be a result from an increase in anthropogenic activity o n October 7 th or there may have been an increase in water withdrawal between the two sampling sites, therefore changing the flow. This makes comparing available flow data with concentration difficult since ue to water withdrawal. The other nutrients tested showed an increase in con centration between cc 2 and cc 3 which shows that these analytes tend to persist in the aquatic ecosystem. Persistence seems to not be the only factor with these nutrients sinc e concentrations do in fact increase between the sites that are b oth downstream of the brewery and the other point sources The reasons for this i ncrease in phosphor us, nitrate, and orthophosphate could be caused by contaminated runoff from non point sour ces such as farms that use these nutrients in their fertilizers. The creek itself may also foster pollutant accumulation or reintroduction especially if there is a lack in a sufficient riparian buffer zone. The larger the distance between the creek and t he non point source runoff (meaning the larger the buffer zone is), the less chance there is for contaminants to make it to the creek ; it will instead be deposited or filtered out of the runoff and into the soil. These nutrient s may also be deposited in s oil along the stream banks during low stream flow and later be reintroduced back into the water when stream discharge increases Measurements for orthophosphate and total phosphorus are closely related since total phosphorus is the sum of orthophosphates organically bound phosphates, and condensed phosphates (EPA 2012) Orthophosphate concentrations should be less than all of the total phosphates, which was not seen for any of the dates sampled for cc 1 and was not seen at either site during the high flow sampling date The best explanation for this occurrence is how the phosphorus was analyzed; l ow er concentrations of phosphorus compared to orthophosphate could occur if the water samp le was filtered first, which will only allow for total dissolved phosphorus to be measure d. If this was the form that was analyzed, then higher orthophosphate concentrations would be possible This means that the

PAGE 35

28 orthophosphates were the only phosphates detected for the total dissolved phosphorus at the sites/dates that experienced higher concentrations of orthophosphates Table 6: Ani on Concentrations for Sites cc 1, cc 2, cc 3 for all Sampling Dates Out of the three anions tested, only chlorid e showed a distinct dif ference in concentration from the site upstream of the point sources compared to the two sites below (Table 6 ) As previously seen with other analytes tested, the increase in concentration for chloride was less significant dur ing th e high flow sampling date. The largest difference was seen in October; where concentrations went from 13.4mg/L at cc 1 to 69.3mg/L downstream at cc 3 The large increase does not correlate with stream flow since the subsequent sample date had less flow and a smaller concentration average for the three sites. The cause for the large increase in October could be from increased point source discharge.

PAGE 36

29 Whether one point source was discharging more that day o r there were more active outfalls from several of the point sources cannot be verified but the increased chloride concentration for all four sample dates was most likely caused by treated discharge from one or more of these sources. M any water treatment processes incorporate the use of chloride species as a form of disinfectant which would account for the changes in conc entrations at both cc 2 and cc 3 The anion with the highest concentrations overall was sulfate with one measurement reaching up to 79.3mg/L. The presence of sulfate and fluoride we re most likely not caused by any discharge bein g released between cc 1 and cc 3 since concentrations stayed relatively the same for all three sampling sites. Fluctuations in the amount of sulfate and fluoride could simply be d ue to the accuracy of the au tosampler during sample analysis since percent recovery for reference samples can be plus or minus 10% Another indica tion that the wa ter quality is impaired downstream from the combination of point sources and non point sources is the amount of TOCs foun d in the water per liter. All four sampling dates showed an increase in con centration between cc 1 to cc 2 (Table 7 ) The sampling date that experienced the greatest increase was in October where TOC levels increa sed around 8.5 mg/L between cc 1 and cc 2 This day showed the largest concentrations and th e greatest increase between cc 1 and cc 2 for several of the analytes measured. Once again, this could be a result of an increase in anthropogenic activity from either one or more of the point sources d ischarging into the stream. TOC concentrations were relatively stable for the other three sampling dates with only minor increases between the sampling sites. This is not the case in regions receiving untreated point source discharge as seen with the Olo sun River, which showed large analyte concentration increases downstream of the brewery ( Ipeaiyeda and Onianwa 2009) The small TOC concentration increases seen below the brewery and the other point sources discharging in Clear Creek may be due to the out falls being inactive at that specific time or the wastewater treatment process is efficient at filtering out organic matter. These minor changes may not even be associated with the brewery or other point sources and could actually

PAGE 37

30 be caused by a natural sources like decaying plants (Schumacher 2002 ) Analysis determining the quantity of decaying plants and leaf l itter at each site would be requir ed to prove that the concentration spike was from a natural source. Another interesting pattern observed was the drop in the amount of TOC present after cc 2 for all sampling dates; t his could be from the larger organic particles settling into the sediment at the bottom of the creek or micro and macro organisms consuming the particles. The other water quality par ameter that seemed to correlate with TOC and chloride concentrations was specific conductance (Table 4 and Figure 10 ) (USGS 2014) Specific conductance is a measure that show s how well water can conduct electrical c urrent, which can increase with the intr oduction of diss olved solids such as salt (USGS 2 014). This would explain why conductance increased at sites below the brewery where concentrations of chloride and TOC were higher as well (Figure 10 ) Figu re 10 : Correlation between specific conductance and concentrations found for chloride and total organic carbon The measurements for specific conductance correlated with chloride concentrations for the different sampling dates with low measurements seen during the high fl ow date and high measurements found in October. T h e smallest conductance measured downstream during the high flow sampling date was 122

PAGE 38

31 uS/cm; where as the highest me asured conductance was almost seven times that at 815 uS/cm in October. The measurements for s pecific conductance can a lso be used to estimate total dissolved solids (mg/L) (Hem 1985) Total dissolved solids can be calculated by multiplying the specific conductance b y a constant derived from data collected by the USGS on the Gila River in 1947 (Hem 1985 ). The Gila River study found a well defined relationship between the two parameters with an uncertainty of + 100 mg/L The equation KA = S (K: specific conductance, A: constant, and S : total dissolved solids ) was analyzed with E xcel to determine the approximate total diss olved solids present per liter in Clear Creek for each sample site and date (Table 8). The constant used was 0.59, which was the slope of the regression line (r 2 of 0.9604 ) for the entire Gila River data set. Since these parameters are directly correlate d, total dissolved solids also increased below the point sources and directly correlated with chloride and TOC concentrations (Figure 10). Diss olved solids within a stream consist of any ion particles that can pass through a filter with pores around 2 mic rons in size, which include calcium, chlorides, nitrate, phosphorus, iron, and sulfur (EPA 2012). The average concentrations for chloride, nitrate, and Table 8: Estimated Total Dissolved Solids at sites CC 1, CC 2, and CC 3 for all sample dates phosphor u s during each sample event at Clear Creek did directly correlation with the estimated dissolved solids concentration (Figure 11) As discussed previously the lowest concentrations for all

PAGE 39

32 four parameters shown in Figure 11 were during the high flow sampli ng date in July and the highest concentrations were seen in October. The buffering capacity (alkalinity) of the water dra stically changed after the point sources with the greatest difference seen in October (Table 7 ) Once again this could be due to t he increased amount of dissolved material in the water since dissolved organic anions derived from large amounts of dissolved organic carbon can add alkalinity (Wetzel 2001) The introduction of wastewater from the point sources also explains why the alk alinity is high at cc 2 and cc 3 since typical wastewater has high concentrations of nutrients that have acid bufferi ng properties, such as phosphor us (MECC no date). Figure 1 1: Correlation between phosphor us, nitrate and chloride concentrations compare d to total dissolved solids This would explain the 183.5mg /L difference between sites cc 1 and cc 2 in October, which also had the h ighest levels in total phosphor us (Table 5 ) total nitrates (Table 5 ) chloride (Table 6 ) and TOCs (Table 7 ) The changes in pH could potentially be affected by the alkalinity especially if carbonate is involved

PAGE 40

33 (Table 4 ) The carbonate species will react with water, which results in the release of the hydroxide ion, making the pH more basic. A ll sampling dates experien ce d a grad ual increase in pH ( more basic) between cc 1 to cc 3 The pH measurements may have only changed gradually due to the high buffering capacity of the water. If alkalinity was lower, the water could have been more acidic since dissolved organic ma tter (a form of organic carbon that can pass through a 0.45 micron filter) is acidic The gradual change shows that whatever reactions may be occurring, are taking place outside of the treatment facility. It is in the best interest of the treatment facil ity to keep the pH as neutral as possible since higher pH values sup press the effectiveness of disinfection with chloride species and a lower pH can corrode or even dissolve metals and other substances (USGS, 2014).

PAGE 41

34 CHAPTE R IV CONCLUSION This study highlights the impacts point source discharge can have on an aquatic ecosystem even after treatment. chemistry the most without testing at o utfall mixing zones, but an overall understanding of how the creek reacts to all of the point sources was determined. Analysis confirmed that the water quality changed and several analytes tested had drastic differences between cc 1 ( above the poi nt sources ) compared to cc 2 and cc 3 ( below the point sources) Many of these analytes are directly dependent on each other, so any changes in concentration typically resulted in changes of other analyte concentrations within Clear Creek. An example of this dependence was seen with the increase in E. coli concentrations at sites with an observed increase in temperature and nutrient concentrations The E. coli present within the creek probably performed both aerobic and anaerobic metabolic processes since DO was seen to decrease betw een cc 1 and cc 2. The oxygen was also used for nitrification; this was seen between sites cc 2 and cc 3 where nitrite concentrations showed an appare nt decrease while nitrate concentrations showed an increase Dissolved o xygen stayed relativel y stable between cc 2 and cc 3 even though there was the combination of microbial and chemical processes consuming it, which could be due to algae producing oxygen during photosynthesis. Flow also played a major role with all of the analytes measured; whe re the lowest measurements were found during the high flow sampling date in July. The October sampling date in particular showed the largest concentrations and th e greatest increase between cc 1 and cc 2 for several of the analytes measured. The exact re ason for the greater increase in analyte concentrations cannot be confirmed, but reduced flow beyond the point sources are an increase in point source discharge could very well be what caused the differences. All of these changes can pose serious risks to b oth environmental and human health, and should be addressed by making continual improvements to the treatment process for each point source.

PAGE 42

35 REFERENCES Anthonisen, A.C., Loehr, R.C., Prakasam, T.B.S., Srinath, E.G. (1976) Inhibition of nitrification by ammonia and nitrous acid. Journal Water Pollution Control Federation 48: 835 852 Brewers Association. (2012) Water and Wastewater: Treatment/Volume Reduction Manual. Campbell, C.S., Ogden, M. (1999) Constructed Wetlands in the Sustainable Landscape (1 st ed.) New York, New York: John Wiley & Sons, Inc. Chen, Bao Ming, Wang, Zhao Hui, Li, Sheng Xiu, Wang, Gen Xuan, Song, Hai Xing, Wang, Xi Na (2004) Effects of nitrate supply on plant growth, nitrate accumulation, metabolic nitrate concentration and nitrat e reductase activity in three leafy vegetables. Plant Science 167: 635 643 Crittenden, J.C., Trussell, R.R., Hand, D.W., Howe, K.J., Tchobanoglous, G. (2012). Physical and Chemical Quality of Water. MWH's Water Treatment: Principles and Design (3rd ed., p p. 60 61). Hoboken, New Jersey: John Wiley & Sons. Davies, Trevor D (2006) Sulphate toxicity to the aquatic moss, Fontinalis antipyretica Chemosphere 66: 444 451 Garcia Garcia, Gerardo, Nandini, S. (2006) Turbidity mitigates lead toxicity to cladoceran s (Cladocera) Ecotoxicology 15: 425 436 GEI Consultants, Inc. (2015, February) Aquatic Biological Monitoring in clear Creek, Colorado. Retrieved April 14, 2015. ftp://ft.dphe.state.co.us/wqc/wqcc/38TriennialReviewRMH_2015/Proponents/MillerCoors%20Exhibits/ MillerCoors%20Exhibit%20E_GEI%20Data%20Report.pdf Habte Lemji, Haimanot, Eckstadt, Hartmut (2 013) A pilot scale trickling filter with pebble gravel as media and its performance to remove chemical oxygen demand from synthetic brewery wastewater Jounal of Zhejiang University SCIENCE B Biomedicine & Biotechnology 14: 924 933 Hem, J.C. (1985) Specif ic electrical conductance. Study and interpretation of chemical characteristics of natural water (3rd ed., pg. 67). Alexandria, Virginia: U.S. Geological Survey. Ipeaiyeda, A. R., Onianwa, P. C. (2009) Impact of brewery ef fluent on water quality of the Olosun R iver in Ibadan, Nigeria. Chemistry and Ecology 25: 189 204 Linge, K., Blythe, J., Busetti, F., Blair, P., Rodriguez, C., & Heitz, A. (2013). Formation of halogenated disinfection by products during microfiltration and reverse osmosis treatment: Im plications for water recycling. Separat ion and Purification Technology 104 : 221 228. Janhom, Tansiphorn, Wattanachira, Suraphong, Pavasant, Prasert (2008) Characterization of brewery wastewater with spectrofluorometry analysis. Journal of Environmental Ma nagement 90: 1184 1190

PAGE 43

36 Janhom, Tansiphorn, Pavasant, Prasert, Wattanachira, Suraphong, (2010) Profiling and monitoring of DOM in brewery wastewater and treated wastewater. Environmental Monitoring Assessment 176: 403 418 Maupin, M.A., Kenny, J.F., Hutso n, S.S., Lovelace, J.K., Barber, N.L., Linsey, K.S. (2014, November 5). Estimated Use of Water in the United States in 2010. United States Geological Survey. Retrieved January 11, 2015, from http://pubs.usgs.gov/circ/1405/. Mountain Empire Community Colle ge (no date) Alkalinity. Water/Wastewater distance learning website. Retrieved April 14, 2015. http://water.me.vccs.edu/exam_prep/alkalinity.html Olajire, Abass A. (2012) The brewing indus try and environmental challenges. Journal of Cleaner Production 1 21 Oluah, NS (2007) Anaemia in the freshwater catfish Clarias albopunctatus (Teleostei: Clariidae) exposed to brewery wastewater effluent. African Journal of Aquatic Science 33: 157 160 Pu lido, Cristina, Keijsers, Danny J. H. (2012) Elevated alkalinity and sulfate adversely affect the aquatic macrophyte Lobelia dortmanna. Aquatic Ecology 46: 283 295 Schumacher, B. (2002). M ethods for the determination of total organic carbon (TOC) in soils and sediments. Ecological Risk Assessment Support Center, NCEA C 1282 (EMASC 001), 1 23. Retrieved November 27, 2014, from http://epa.gov/esd/cmb/research/papers/bs116.pdf Scott, R. (2012, February 29) Understanding the basic principles of nitrogen. Connecticut Water Pollution Abatement Association. Retrieved June 22, 2015. https://ctwpaa.org/understanding the basic principles of nitrogen/ Todar, K. (2008 2012) Pathogenic E. coli Online Tex tbook of Bacteriology. Madison, Wisconsin. http://textbookofbacteriology.net/e.coli.html Unden, G., Becher, S., Bongaets, J., Schirawski, J., Six, S. (1994) Oxygen regulated gene expression in facultatively anaerobic bacteria. Antonie van Leeuwenhoek 66: 3 22 U.S. Environmental Protection Agency. (2014, December 9). Centralized Waste Treatment Effluent Guidelines. Retrieved January 11, 2015. water.epa.gov/scitech/wastetech/guide/cwt/index.cfm U.S. Environmental Protection Agency (2014, September 2). Enforcement and Compliance History Online. Facility Search Results. Retrieved April 14, 2015. echo.epa.gov/detailed facility report?fid:110033145736 U.S. Environmental Protection Agency (2014, Sep tember 2). Enforcement and Compliance History Online. Detailed Facility Report. Retrieved May 29, 2015. http://echo.epa.gov/detailed facility report?fid=110037263742 U.S. Environmental Protection Agency. (2013, June 21). My Waters Mapper. Retrieved Janua ry 11, 2015. http://watersgeo.epa.gov/mwm/

PAGE 44

37 U.S. Environmental Protection Agency. (2015, March 13). Summary of the Clean Water Act. Laws and Regulations. Retrieved June 23, 2015. http://www2.epa.gov/laws regul ations/summary clean water act U.S. Environmental Protection Agency. (2014, December 29). Water: Industry Effluent Guidelines. Industrial Regulations. Retrieved January 11, 2015. water.epa.gov/scitech/wastetech/guide/industry.cfm#studies. U.S. Environmental Protection Agency. (2012, March 6) 5.6 Phosphorus. Retrieved June 24, 2015 http://water.epa.gov/type/ rsl/monitoring/vms56.cfm U.S. Environmen tal Protection Agency. (2012, March 6) 5.8 Total Solids. Retrieved June 24, 2015. http://water.epa.gov/type/rsl/monitoring/vms58.cfm U.S. Environmental Protection Agency. (2012, March 6). 5.10 Total Alkalinity. Retrieved May 23, 2015. http://water.epa.gov/type/rsl/monitoring/vms510.cfm U.S. Geological Survey. (2015, May 16). National Water Information System: Web Interfa ce Retrieved May 16, 2015. http://waterdata.usgs.gov/co/nwis/inventory/?site_no=06719505 U.S. Geological Survey. (2015, May 16). National Water Information System: Web Interfa ce. Retrieved May 16, 2015. http://waterdata.usgs.gov/co/nwis/inventory/?site_no=06718550 U.S. Geological Survey. (2014, April 8). Water Properties and Measurements. Retrieved January 11, 2015 http://water.usgs.gov/edu/characteristics.html U.S. Geological Survey. (2014, March 17). pH Water Properties. Retrieved January 11, 2015. http://water.usgs.gov/edu/ph.html Wetzel, R. (2001). Prologue. In Limnology Lake and River Ecosystems (3rd ed., p. 1). San Diego, California: Elsevier Academic Press.