Citation
Geospatial evaluation of non-point stormwater runoff for developed residential and commercial land uses

Material Information

Title:
Geospatial evaluation of non-point stormwater runoff for developed residential and commercial land uses
Creator:
Zivkovich, Brik R. ( author )
Language:
English
Physical Description:
1 electronic file (113 pages). : ;

Subjects

Subjects / Keywords:
Water -- Pollution -- Colorado -- Denver ( lcsh )
Urban runoff -- Colorado -- Denver ( lcsh )
Urban runoff ( fast )
Water -- Pollution ( fast )
Colorado -- Denver ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Understanding and evaluating urban impacts on natural ecosystem processes has become an increasingly complex task for engineers, planners and environmental scientists. As built environments continue to grow, increased human activity and large-scale development drastically stress receiving urban streams and lakes resulting in the current impaired and degraded state of surface waters. In response, integrated water quality management programs have been adopted to address these unregulated non-point sources by utilizing best management practices to treat this runoff as close to the source as possible. The following study provides a detailed statistical and geospatial analysis process to analyze how different land uses affect urban stream systems. Using metropolitan Denver, Colorado as a case study, this paper presents a general evaluation method to identify critical non-point pollutant source watersheds and associated sub-basins. The two phase analysis led to the development of a non-point stormwater assessment matrix that can be used to aid stormwater professionals to evaluate and specify retrofits of water quality features within urban areas. The selected water quality features can be used reduce, capture and treat stormwater runoff specific to pollutant loading prior to entering nearby urban surface waters as location is vital for optimizing pollutant reduction found in non-point stormwater runoff.
Thesis:
Thesis (M.S.)--University of Colorado Denver.
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
Department of Civil Engineering
Statement of Responsibility:
by Brik R. Zivkovich.

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:
922008527 ( OCLC )
ocn922008527
Classification:
LD1193.E53 2015m Z59 ( lcc )

Downloads

This item has the following downloads:


Full Text
GEOSPATIAL EVALUATION OF NON-POINT STORMWATER RUNOFF FOR
DEVELOPED RESIDENTIAL AND COMMERCIAL LAND USES: CASE STUDY
OF DENVER, COLORADO
by
BRIK R. ZIVKOVICH
B.S., University of Pittsburgh, 2013
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
Master of Science
Civil Engineering
2015


2015
BRIK R. ZIVKOVICH
ALL RIGHTS RESERVED


This thesis for the Master of Science degree by
Brik R. Zivkovich
has been approved for the
Department of Civil Engineering
by
James C.Y. Guo, Chair
David C. Mays
Wesley E. Marshall
August 1st, 2015
n


Zivkovich, Brik R. (M.S., Civil Engineering)
Geospatial Evaluation of Non-Point Stormwater Runoff for Developed Residential and
Commercial Land Uses: Case Study of Denver, Colorado
Thesis directed by Associate Professor David C. Mays.
ABSTRACT
Understanding and evaluating urban impacts on natural ecosystem processes has
become an increasingly complex task for engineers, planners and environmental
scientists. As built environments continue to grow, increased human activity and large-
scale development drastically stress receiving urban streams and lakes resulting in the
current impaired and degraded state of surface waters. In response, integrated water
quality management programs have been adopted to address these unregulated non-point
sources by utilizing best management practices to treat this runoff as close to the source
as possible. The following study provides a detailed statistical and geospatial analysis
process to analyze how different land uses affect urban stream systems. Using
metropolitan Denver, Colorado as a case study, this paper presents a general evaluation
method to identify critical non-point pollutant source watersheds and associated sub-
basins. The two phase analysis led to the development of a non-point stormwater
assessment matrix that can be used to aid stormwater professionals to evaluate and
specify retrofits of water quality features within urban areas. The selected water quality
features can be used reduce, capture and treat stormwater runoff specific to pollutant
loading prior to entering nearby urban surface waters as location is vital for optimizing
pollutant reduction found in non-point stormwater runoff.
The form and content of this abstract are approved. I recommend its publication.
Approved: David C. Mays
m


ACKNOWLEDGEMENTS
Over the past two years, I have been truly overwhelmed and grateful for all the
support and guidance I have received since I have begun my graduate studies at the
University of Colorado Denver. Not only has this university helped enhance and improve
my engineering skillset, but it has become a second home away from family. There are
many people I would like to thank that have helped me along the way as I would not be
in the seat I am today without their support and guidance.
First, I would like to thank is my advisor, Dr. David Mays, for all of his support
and guidance throughout my graduate studies at here at the University of Colorado
Denver. He has been a great mentor and professor that has provided endless support and
advice in relation to my coursework as well as my masters thesis report. He has truly
pushed me to reach my full ability as the past two years of classes and thesis meetings
have enabled great education experience for me.
Next, I would like to thank my two committee members, Dr. James Guo and Dr.
Wesley Marshall. Both have been influential professors who through their classes
enabled this thesis topic to be performed. Additionally, both provided great advice and
guidance when I had reached an obstacle during my masters studies. Their advice
pressed me to look at different ways to address these obstacles in order to find an
alternative way to analyze or think about a problem.
Next, I would like to send out an extremely gracious thank you to all employees at
Urban Drainage and Flood Control District. After acquiring a graduate internship in the
master planning program at the start of my graduate studies, my skillset has exponentially
grown as I have been exposed to real-world engineering problems that coincide perfectly
with my masters studies. A special thanks goes out to my bosses Ken MacKenzie, Shea
IV


Thomas and Holly Piza, as well as Julia Bailey, as they have helped provide advice and
guidance whenever I had a question relating to my thesis work. Additionally, a special
thanks also goes out to contacts met through this internship program including Jane Clary
at Wright Water Engineers, Inc. and David Delagarza at RESPEC Engineering.
Last and most important, an unmeasurable thank you goes out to my family and
friends who have continue to support me as I get one step closer to my career dreams.
They have always been there to care and provide support for me in every decision that I
make and I am truly grateful them in my life. Thank you again for everything, I wouldnt
be here without you.


TABLE OF CONTENTS
Chapter
1. Introduction.................................................................1
1.1 Purpose of the Study....................................................1
1.2 Scope of the Study......................................................2
2. Literature Review............................................................3
2.1 Overview................................................................3
2.2 Regulations and Standards...............................................3
2.3 Stormwater Quality Databases............................................5
2.4 Urbanization Impacts....................................................6
2.5 Objectives of Case Study for Denver, Colorado...........................8
3. Methods.....................................................................10
3.1 Overview...............................................................10
3.2 Site Selection.........................................................12
3.3 Non-Point Stormwater Quality Analysis..................................12
3.3.1 Data Selection.....................................................12
3.3.2 Selection of Constituent Data......................................16
3.4 Geospatial Land Cover Analysis.........................................17
3.4.1 Data Collection....................................................17
3.4.2 Geospatial Processing Methods......................................19
3.5 Geospatial Non-Point Stormwater Assessment Matrix......................21
3.5.1 Development of Geospatial Assessment Matrix........................21
3.5.2 Applications of Matrix using Lakewood Gulch Example................23
4. Results and Discussion......................................................25
4.1 Non-Point Stormwater Quality Results...................................25
4.1.1 Statistical Analysis...............................................25
vi


5.1.1 Correlations
54
4.2 Geospatial Land Cover Results...........................................72
4.3 Geospatial Non-Point Stormwater Assessment Matrix Results...............77
5 Conclusion.....................................................................84
References.......................................................................86
Appendix.........................................................................89
A. 1 Raw Data..................................................................89
A.2 Calculations..............................................................101
vii


LIST OF TABLES
Table
1: Final Site Locations and Descriptions from Selection Criteria..........................14
2: Common Urban Stormwater Pollutants and Potential Sources...............................16
3: NLCD Classifications for Developed Land Use............................................19
4: Summary of Non-Point Runoff Statistics (TSS)...........................................28
5: Summary of t-Test (TSS)................................................................28
6: Individual Site Statistics for TSS Analysis............................................30
7: Summary of Non-Point Runoff Statistics (TKN and NO2+NO3)...............................33
8: Summary of One-Tailed and Two-Tailed t-Test (TKN).......................................33
9: Summary of One-Tailed and Two-Tailed t-Test (NO2+NO3)...................................34
10: Individual Site Statistics for TKN Analysis............................................37
11: Individual Site Statistics forNCL+NCL Analysis.........................................37
12: Summary of Non-Point Stormwater Statistics (TP and DP)................................41
13: Summary of One-Tailed and Two-Tailed t-Test (TP).......................................41
14: Summary of One-Tailed and Two-Tailed t-Test (DP).......................................41
15: Individual Site Statistics for TP Analysis.............................................44
16: Individual Site Statistics for DP Analysis.............................................44
17: Summary of Non-Point Stormwater Statistical Analysis (Cu and Zn)......................48
18: Summary of One-Tailed and Two-Tailed t-Test (Total Copper)...........................48
19: Summary of One-Tailed and Two-Tailed t-Test (Total Zinc).............................49
20: Individual Site Statistics for Total Copper Analysis...................................52
21: Individual Site Statistics for Total Zinc Analysis.....................................52
22: Comparison to 1983 NURP Study (Residential)............................................53
23: Comparison to 1983 NURP Study (Commercial).............................................53
24: I.I\LS I Results for TKN vs. N02+N03...................................................56
25: LIN LS I Results for TP vs. DP.........................................................56
26: LINEST Results for Total Zinc vs. Total Copper.........................................57
27: LINEST Results for TSS vs. TKN......................................................63
28: LINEST Results for TSS vs. N02+N03..................................................63
29: LINEST Results for TSS vs. TP.......................................................64
30: LINEST Results for TSS vs. Total Copper.............................................69
31: LINEST Results for TSS vs. Total Zinc...............................................69
32: NLCD Analysis of Developed Areas within 25-km Radial Boundary.........................73
33: Lakewood Gulch Geospatial Analysis Example using Polygons.............................78
34: Geospatial Summary for Sub-Basins for Lakewood Gulch Matrix Application...............79
viii


LIST OF FIGURES
Figure
1: Residential and Commercial Sampling Locations for Denver, Colorado....................15
2: Box-and-Whisker Plots for Residential and Commercial Locations (TSS)..................29
3: Comparison of TKN for Residential and Commercial Sampling Locations....................35
4: Comparison of NO2+NO3 for Residential and Commercial Sampling Locations................36
5: Comparison of TP for Residential and Commercial Sampling Locations.....................42
6: Comparison of DP for Residential and Commercial Sampling Locations.....................43
7: Comparison of Total Copper for Residential and Commercial Sampling.....................50
8: Comparison of Total Zinc for Residential and Commercial Sampling Locations.............51
9: Constituent Correlations (TKN and NO2+NO3)............................................58
10: Constituent Correlations (TP vs. DP).................................................59
11: Constituent Correlations (Total Zinc vs. Total Copper)...............................60
12: TSS vs. Nutrients Correlations (TKN).................................................65
13: TSS vs. Nutrients Correlations (NO2+NO3).............................................66
14: TSS vs. Nutrients Correlation (Total Phosphorus).....................................67
15: TSS vs. Metals Correlation (Total Copper)............................................70
16: TSS vs. Metals Correlation (Total Zinc)..............................................71
17: Developed Land Cover Map for 25-km Radius around Denver, CO (2001).................74
18: Developed Land Cover Map for 25-km Radius around Denver, CO (2006).................75
19: Developed Land Cover Map for 25-km Radius around Denver, CO (2011).................76
20: Comparison of Sub-Basin TSS Loading...................................................79
21: Lakewood Gulch Watershed Map.........................................................80
22: Lakewood Gulch Watershed Map with Delineated Sub-Basins..............................81
23: Validation of Matrix with Mixed Use Sampling Location................................83
IX


LIST OF ABBREVIATIONS
AGNPS Agricultural Non-Point Source Pollution Model
BMP Best Management Practice
CWA Clean Water Act
CWP Center for Watershed Protection
DP Dissolved Phosphorus
EMC event mean concentration
EMxC event maximum concentration
LID Low Impact Development
MS4 Municipal Separate Storm Sewer System
NO2+NO3 Nitrite plus Nitrate
NPDES National Pollution Discharge Elimination System
NSQD National Stormwater Quality Database
NURP National Urban Runoff Program
TKN Total Kjeldahl Nitrogen
TMDL Total Maximum Daily Load
TNPL Total Non-Point Loading
TP Total Phosphorus
TSS Total Suspended Solids
U S EPA U.S. Environmental Protection Agency
UDFCD Urban Drainage and Flood Control District
WEF Water Environment Federation
WWE Wright Water Engineers, Inc.


1. Introduction
Water is the driving force of all nature.
Leonardo da Vinci
1.1 Purpose of the Study
Optimal watershed management practices and strategies to evaluate impairments
in rivers and streams have become a critical objective for water resource engineers and
environmentalists to improve qualities of streams and lakes (Kaplowitz andLupi 2012;
Carey etal. 2013). Research and environmental assessments continue to provide strong
evidence urbanization from increased human development creates excess nutrients,
metals, and sediments that directly impact ecological properties and stability of surface
waters (Mitchell 2005; Maestre and Pitt 2006; Walsh et al. 2012; Son et al. 2015; Park
and Park 2015). These excess pollutants create biogeochemical instabilities in
ecosystems as urban deposition is washed into receiving streams and lakes during storm
events (Lee and Bang 2000).
In response, many cities have begun to adopt green infrastructure programs using
sustainable urban planning techniques by implementing and retrofitting low impact
development (LID) to address non-point stormwater runoff and reduce the water footprint
left from urban regimes (U.S. EPA 1986; Benedict and McMahon 2006; Dietz 2007; U.S.
EPA 2007b; Chau 2009). These LID designs, such as water quality detention basins, rain
gardens, constructed wetlands, and grass swales, have prominent potential to treat non-
point runoff by permitting multi-functional uses that rely on natural hydrologic
principles. (Guo 2009; U.S. EPA 2007a; UDFCD 2013). Utilization of these natural
hydrologic processes is key as LID designs can act as buffers for reducing pollutants that
1


reach streams (i.e. utilization of sediment collection pads, metal reductions with select
soil media, and optimizing phytoremediation processes for nutrients with select native
vegetation) (Wulliman and Thomas 2005; UDFCD 2013). With proper design, planning
and maintenance, these LID systems can effectively treat urban runoff and reduce that
magnitude of pollutants that enter urban streams and lakes.
1.2 Scope of the Study
As built environments continue to grow, increased human activity and large-scale
development drastically stress receiving urban streams resulting in the current impaired
and degraded state of urban surface waters. (U.S. EPA 2010; Stevens and Slaughter
2012). In response, integrated management and sustainable urban planning have become
necessary to mitigate impacts from new developments. The following study provides a
detailed statistical and geospatial analysis for two developed land uses that impact urban
stream systems. These land uses include developed residential and commercial areas that
have imperviousness values greater than 20%. Using metropolitan Denver, Colorado as a
case study, this thesis presents a general evaluation method to identify critical source
watersheds and associated sub-basins contributing high non-point urban pollutant loads
during the first flush of storm events. This process can be used to aid stormwater
professionals by developing a general methodology for locating retrofits of water quality
features within urban areas that can be used reduce, capture and treat stormwater runoff
prior to entering the surrounding natural streams and river systems. Additionally, this
process can be to aid environmental regulatory agencies and other interested parties in
finding locations that can be key sources of degradation to nearby streams.
2


2. Literature Review
2.1 Overview
Although stormwater has long been regarded as a major factor in flooding of
urban areas, only since the 1980s have policymakers, engineers, and environmentalists
recognized the additional roles that stormwater plays in the impairment of urban
watersheds and natural ecosystems (National Research Council 2009aj. Since the
creation of the Clean Water Act (CWA) in the 1970s, the U.S. Environmental Protection
Agency (U.S. EPA) has continued to develop and establish new regulations to address
discharges of pollutants into streams and rivers (Regas 2005; U.S. EPA 2007a).
Motivation and incentives to address surface water impairments have led designers,
planners, and engineers to pollutants located in urban stormwater runoff.
2.2 Regulations and Standards
Urban populations have grown significantly since of creation of the Clean Water
Act, so consequently stormwater pollutants have become the primary cause of
impairment for urban surface waters (U.S. EPA 2015). The CWA, which originated to
address point discharges into streams and lakes, continues to mature rapidly as
organizational research provides evidence to support integrated and sustainable water
resource management practices. Organizations and programs including the U.S. EPAs
National Urban Runoff Program (NURP), American Society of Civil Engineers
Environmental and Water Resources Institute (EWRI), the nonprofit organization Center
for Watershed Protection (CWP), and the sanitary engineering organization Water
Environment Federation (WEF) continue providing research and water quality studies
that focus on lakes and rivers prone to stormwater pollutants left unregulated by the
3


CWA. These pollutants have become the main concerns as built environments such as
residential developments and large commercial areas are a major contributors to
impairments in urban surface waters (U.S. EPA 2007).
In response to this organizational research, many regulations and standards have
been adopted under the CWA to address and evaluate these urban stormwater pollutants.
Under Section 402 of the CWA, the National Pollutant Discharge Elimination System
(NPDES) was adopted in order to set effluent-based standards and ensure compliance
from dischargers. NPDES Municipal Separate Stormwater Sewer System (MS4) permits
are issued from state governments and approved by the US EPA as a way to implement
regulations for dischargers. These MS4 permits have worked to address additional gaps
that leave rivers and lakes prone to once overlooked unregulated source pollutants.
Preliminary data summaries have been reported to characterize Phase 1 NPDES
stormwater data for more than 200 municipalities throughout the country (Pitt et al.
2003).
Additionally, under Sections 305(b) and 303(d) of the CWA, the U.S. EPA
requests states to report on water quality conditions through National Water Quality
Assessment Reports (Regas 2005). These bi-annual integrated reports are used to assess
streams and surface waters, identify impaired waters, and their causes, and track the
status of actions being taken to restore impaired waters to their ambient state. Using the
most recent Colorado Water Quality Assessment Report from 2010, approximately 19%
of assessed streams and 49% of assessed lakes are impaired, which triggers the
requirement for a Total Maximum Daily Load (TMDL) plan (National Research Council
2009b; U.S. EPA 2010). Despite an investment of over hundreds of millions of dollars
4


through the 1990s to improve the quality of the nations waters, the impact from
unregulated non-point sources continues to limit attaining the ultimate goal of
swimmable and fishable waters (National Research Council 1999).
2.3 Stormwater Quality Databases
Multiple stormwater quality databases have been created over past decades to
assess stormwater runoff quality based on a number of different characteristics. Of these,
the National Stormwater Quality Database (NSQD) and the International Stormwater
BMP Database are two resources used to aid water quality assessments and research
efforts to improve urban stormwater conditions prior to entering urban surface waters.
Water quality assessments for streams and rivers is a continuous process that
requires a thorough understanding of stream water management and good monitoring
practices (Maestre and Pitt 2006). With support from the U.S. EPA and the CWP, the
NSQD was recently updated to the NSQD Version 4.02 and is the major resource for all
stormwater data that has been collected over recent years. This NSQD is a national urban
stormwater runoff database that serves as an important resource for urban runoff data,
categorized by location, land use, years of record, along with several other
characterizations (Pitt 2015). Additionally, the International BMP database provides
BMP stormwater monitoring data that has been collected for BMP monitored sites. These
BMP site locations, more often than not, have a reference site at which the BMP system
is compared with. This reference location, at which stormwater quality samples are also
taken, is used as a resource within for the NSQD as it aligns with urban stormwater data
that has already been included in the database. Both databases will be prominent
resources used in this study as they address urban runoff pollutants for different land use
5


categories and can be used for understanding sampling site locations and descriptions that
are found within both databases.
2.4 Urbanization Impacts
As mentioned in this chapter, non-point urban runoff from residential and
commercial areas greatly contribute to the disruption of natural stream processes by
discharging excess sediments, nutrients and metals into the drainageways. These urban
pollutants, which have caused impairments in urban surface waters, are commonly
analyzed by their different stormwater pollutant categories. Although there is an
extensive list of constituents that could have been analyzed such as organic elements,
additional fuel by-products, pathogens, or other contaminants of evolving concern, three
common pollutant categories associated with stormwater runoff were selected (Maestre
and Pitt 2007; Pitt 2015). First flush effects have become a primary area of research as
certain pollutants have evaluated strong relations to rainfall events, however all
monitoring practices for each of the data sites would be required (Hathaway et al. 2012).
Three pollutant categories, which will be used for this analysis, include sediments,
nutrients, and metals. These three categories are the most common urban stormwater
pollutants that are studied in relation to stormwater quality runoff (UDFCD 2013; Pitt
2015; Son et al. 2015). The following section provides a background analysis of three
common stormwater pollutant categories and how urbanization impacts from these
pollutants affect natural ecosystems.
The first category is sediments. Total Suspended Solids (TSS) are often used to
evaluate stormwater runoff in the form of the sediment pollutant category. TSS consists
of loose particles that are carried during storm events into streams and lakes. These loose
6


particulates, which commonly are made up of sands, silts, clays and other small particles,
are naturally occurring as waters erode natural landscapes and carry these particles into
waters. This sediment transport cycle has been rapidly increasing as urban development
induces additional sediment loading into streams and rivers. These sediments, which are
carried off of parking lots and large impervious areas through diverted stormwater
systems, bring heavy stormwater sediments loads to specific points within streams and
rivers. This excessive sediment loading affects geomorphological properties within rivers
resulting in adverse effects downstream of this location. With proper planning and
management, the impact felts from urban sediments can be reduced as implementation of
stormwater quality detention basins allow loose particulate found in stormwater to settle
to lower depths and removed at times when there is no surface water (Guo 2009).
The second pollutant category is nutrients. Nutrients in the form of nitrogen and
phosphorus compounds are often used to evaluate stormwater runoff in the form of the
nutrient pollutant category. Nitrogen and phosphorus are both essential in providing
healthy, natural ecosystems the ability to function at optimal capacities (Smith et al.
2003). Excessive nutrients can create unwanted conditions for natural ecosystems as
accelerated plant and algae growth can deplete oxygen levels in streams in rivers. In
response to this degradation of surface water quality, some or a combination of adverse
effects can follow. Accompanied with the degradation of water quality, disruption of
natural process such as removing large forested areas along river banks and large
developments near lakes can cause significant impacts on natural processes unless
properly designed. These impacts include reduced fish populations, non-swimmable and
boatable waters, and even destruction of entire ecosystems. Proper utilization of
7


constructed wetlands and engineered gardens with native tolerant vegetation can help
reduce these sediments prior to finding their way into streams and rivers (UDFCD 2013).
The last pollutant category is metals. Metals, which are often associated with
sediments, are found to cause direct problems in ecosystems as living organisms are
unable to naturally uptake these heavy elements. Copper and Zinc are two heavy metals
associated with human and industrial deposition and are commonly analyzed in relation
to the metal pollutant category (Seattle Public Utilities 2009; WWE et al. 2013).
Although metals have become a primary concern, similar implementation of water
quality detention basins that reduce sediment loads also help reduce heavy metal loading
(Walker and Hurl 2002; UDFCD 2013). Additionally, chemical, biological and other
processes carried out by wetlands are able to reduce heavy metals in the form of copper,
zinc and lead (Walker and Hurl 2002).
2.5 Objectives of Case Study for Denver, Colorado
Identification of watersheds contributing high pollutant loads can be a complex
task and methods for determining land use source contribution are uncertain. Much
research and data has been collected in relation to stormwater runoff and the ability to
plan for this minor, yet impactful detail in design and planning can be crucial for
prosperous, healthy engineered ecosystems. The motivation for this thesis is that most of
the impaired rivers and lakes in lakes in Colorado, which are primarily located along the
eastern slope of the Rocky Mountains (i.e., the Front Range), have unknown sources of
contamination. One of the major categories within these unknown sources is urban-
related stormwater runoff. By 2011, over 60% of land use within a 25-km radius around
Denver, Colorado had been developed with imperviousness ranging from 20 to 100
8


percent (Homer etal. 2015). These developed areas, which consist of both single and
multi-family homes, and highly developed commercial and industrial areas, create excess
urban deposition that is transported during storm events through engineered stormwater
systems or directly into receiving streams. Understanding there are significant differences
between stormwater pollutants for various land use types, this study was developed to
evaluate these uncertainties of unknown non-point stormwater sources using water
quality assessment and geospatial methods.
The following study uses metropolitan Denver, Colorado to evaluate urbanization
impacts with respect to collected non-point stormwater runoff data for residential and
commercial land uses. This thesis study uses a two-phase analysis: (1) collection of non-
point stormwater quality data for metropolitan Denver, Colorado and (2) geospatial
analysis linking land cover datasets to urbanization impacts for different land uses. This
analysis led to the development of a non-point stormwater watershed assessment model
that can be used to evaluate and identify watersheds acting as major sources for
sediments, nutrients, and metals, along other additional pollutant loading. Lakewood
Gulch and its surrounding watershed, which falls within the study boundary, was used to
demonstrate the non-point watershed assessment matrix that was developed in response
to a water quality and geospatial analysis methods to determine regional relationships.
9


3. Methods
3.1 Overview
According to the Colorado Water Quality Control Commission (CWQCC), the
availability of event mean concentration (EMC)-based urban runoff data is sufficient to
characterize the nutrient loads within the state of Colorado (Wright Water Engineers, Inc.
et al. 2013). Evaluation of non-point source loading from this urban runoff data that has
been collected enables standards to be developed for the state. However, understanding
the variability of non-point urban water quality runoff with respect to different land uses,
geographical locations, imperviousness, and type of sampling procedure can be a
challenging feat when selecting appropriate site locations for non-point stormwater
quality features (Maestre and Pitt 2006).
Variation of different non-point stormwater land use samples, with respect to
regionalization, was a key aspect in this study when using such large datasets. Analysis of
relevant stormwater site locations was required prior to analyzing sample data. In order to
minimize these variations, strict data acceptance criteria were created to determine
relevant water quality sampling locations. This four part sampling criteria includes: (1)
site location and descriptions, (2) land use type distinguished by database records, (3)
greater than two years of sampling data, and (4) available constituent data for the three
pollutant categories, sediments, nutrients, and metals.
To correspond with land cover datasets and the scope of this study, all
commercial, industrial and institutional land classes designated by the stormwater
databases, were combined into the one category, commercial due that common high
impervious areas (greater than 80%). Additionally, mixed areas were not included in the
10


analysis as they do not align with the geospatial analysis process that separates the land
cover datasets into one of the three classifications, Developed Low Intensity (NLCD
Class 22), Developed Medium Intensity (NLCD Class 23), and Developed High
Intensity (NLCD Class 24). In particular, only non-mixed NSQD classifications were
included, meaning sample sites were required to be 100% residential or 100%
commercial, industrial or institutional.
The following analysis is based on non-point stormwater runoff data accessed
through the National Stormwater Quality Database (NSQD) and land cover classification
rasters accessed through the National Land Cover Dataset (NLCD). Data were
augmented using additional sources including the International BMP database and raw
data collected from personal communication with Holly Piza at UDFCD (2015). This
study uses a two-phase process to develop a geospatial analysis method for evaluating
watersheds as non-point stormwater sources for stormwater pollutants into streams and
lakes. The first phase is a non-point water quality assessment to query, locate and
evaluate stormwater samples for residential and commercial land uses within
metropolitan Denver, Colorado. The second phases uses a geospatial analysis for land
cover datasets that were collected and analyzed through geographic information system
process to determine land cover classifications for metropolitan Denver, Colorado. This
regionalized process was then used to develop a simple and discrete matrix to evaluate
watershed loading in the form of n number of constituents. Methods used for this analysis
will be further explained within this chapter.
11


3.2 Site Selection
The following thesis uses metropolitan Denver, Colorado as a case study area to
analyze urbanization trends with respect to water quality from non-point stormwater
runoff data. Denver, which is located along the Front Range of the Rocky Mountains, is
the first large, urban region through which its streams and rivers flow as they descend
from the adjacent mountainous terrain. Due to population growth, Denver has grown
significantly since 2000 as large residential and commercial developments have been
built across the region. Although many of these new developments are equipped with
stormwater quality detention basins or some form of stormwater pollution prevention, the
impact from non-point stormwater runoff has not been eliminated, which is why the
South Platte River and its major tributaries are listed by the US EPA as impaired waters
with TMDL controls needed (U.S. EPA 2010; Stevens and Slaughter 2012).
3.3 Non-Point Stormwater Quality Analysis
3.3.1 Data Selection
In order to distinguish non-point stormwater sampling locations from other
sampling locations, the four site selection criteria, listed in Chapter 3.1, enabled a query-
based selection process. Using the NSQD v4.02 as the main stormwater database, data
was augmented in the following process.
The first data selection uses the first criteria, location, that was specific to only the
state of Colorado. Of the 690 sampling site locations in the database, 49 of these
sampling sites were in Colorado. The second data selection was again location specific.
Only counties that were located in or around metropolitan Denver, Colorado were
included. These counties include Arapahoe, Adams, Jefferson, and Denver, respectively.
12


Of these 49 sampling locations within Colorado, 38 of these sampling sites were located
inside one of the four counties listed. The third data selection uses the second criteria,
land use type, to select only residential and commercial areas for the analysis. As
mentioned in this chapter, these commercial areas include all commercial, industrial, and
institution database classification for the sampling sites. All mixed sampling site
locations were omitted during this selection. Of the 38 sampling locations within the
selected counties around Denver, 26 of these sampling sites were selected based on the
two land use types, residential and commercial. The fourth selection uses the third
criteria, year of record, to select only location specific, residential and commercial
sampling locations that had more than two years of sample records. This criteria, which is
mainly used to prevent any unique loading for a given year such a construction projects
or after a flood, enables yearly assessment to be recorded, rather than only during one
period in time.
Of the 26 sampling locations determined through the first three criteria, 12 final
site locations were selected for the constituent analysis after year of record had been
determined. It is important to note, that these 12 sites had some form of overlap. Two
site locations, UDFCD Modular Porous Pavement and Shop Creek Wetland Pond, each
had two periods of record that were listed as separate datasets within the database. The
reason for this is unknown, but could correspond with a change in monitoring practice or
other uncertainty not listed. For this study, these two location datasets were combined for
the each site location as each represents either residential or commercial land use.
The selection process results in a total of 10 sampling locations that were selected
from the original 690 sites in the database. These ten sampling locations, identified using
13


the first three established criteria, were made up of five residential and five commercial
locations. These locations can be seen in Figure 1. Additionally, Table 1 provides
locations and descriptions including Site ID, coordinates and sampling events over the
year of record for each of the ten sites augmented.
Table 1: Final Site Locations and Descriptions from Selection Criteria
Site ID Sampling Site Name Year of Record Coordinate Location No. of Event
Residential
COLAIRIS 21st and Iris Rain Garden 2011-2014 39.7488 N 105.1066 W 54
CODEGRHE Grant Heron 2000-2009 39.6197 N 105.0582W 29
CODEGRRE Grant Reflect 2000-2009 39.6184 N 105.0594W 25
COAUSHCR Shop Creek Wetland Pond 1990-1997 39.6291 N 104.7415 W 55
CODEORPO UDFCD Orchard Pond 2000-2011 39.6211 N 105.0598W 106
Commercial
COACWWL3 Arapahoe Country Water & Wastewater Authority (L3) 2008-2009 39.6005 N 104.8379 W 19
COACW6W7 Arapahoe Country Water & Wastewater Authority (W6W7) 2008-2009 39.5919 N 104.8206 W 17
CODEWAWA Denver Wastewater Building 2008-2014 39.7209 N 105.0106 W 67
COLASHOP Lakewood Shops 2005-2015 39.8748 N 105.1630 W 121
COLAMOPA UDFCD Modular Porous Pavement 1994-2006 39.8833 N 105.2000W 48
Note: Shop Creek Wetland Pond and UDFCD Modular Porous Pavement had two sets of sampling years listed in the NSQD. This may reflect a change in monitoring
practices or another uncertainty not listed in the NSQD v4.02.
Source: Pitt (2015), WWE (2014), WWE (2012), and personal communication (UDFCD 2015)
14



Legend Counties Roads Sampling Locations | |Lakes
Adams Freeway Other Maj or Road Cd Commercial Industrial Major Streams
.Arapahoe Minor Roads Residential
Denver Connecting R o ads
Douglas Jefferson
NAD 1953
Figure 1: Residential and Commercial Sampling Locations for Denver, Colorado
15


3.3.2 Selection of Constituent Data
As mentioned in Chapter 2, excessive sediments, nutrients, and metals in response
to urbanization cause impairments in streams and lakes. Non-point runoff from these
urban areas create chemical unbalances within ecosystems ranging from decreased
beneficial uses to loss of life itself. In order to restore these impairments listed by the
U.S. EPA, water quality assessments on streams and lakes need to be carefully analyzed
in order to plan future remediation strategies. Having augmented the dataset into the
selected site locations, the selection of constituents based on available data could begin.
Analysis of the three pollutant categories, sediments, nutrients, and metals, were
determined in response to the two land use types that were under analysis. A total of
seven constituents were analyzed within the three categories. These seven stormwater
pollutants can be seen in Table 2 below along with their potential urban sources.
Table 2: Common Urban Stormwater Pollutants and Potential Sources
Stormwater Pollutant Potential Source
Sediments
Total Suspended Solids [TSS] Construction sites, erosion, poorly vegetated lands, large-commercial vehicles
Common Nutrients
Total Kjeldahl Nitrogen [TKN] Lawn fertilizers, domestic animal waste, vegetative matter, detergents
Nitrite + Nitrate [NO2+NO3]
Total Phosphorus [TP]
Dissolved Phosphorus [DP]
Metals
Total Copper [Cu] Atmospheric deposition from fuel combustion and industrial processes, vehicles, soil erosion
Total Zinc [Zn]
Sources: U.S. EPA (2007b), Seattle Public Utilities (2009), and UDFCD (2013)
16


After established criteria were used to select the sampling site locations, the final
criterion was used on the selected seven constituent data recorded from the database. This
final criterion, which analyzed available constituent data of the final ten sites, sourced an
in-depth and detailed quality assurance check on each of the datasets to address any
unique, repeated or missing values. Identification of unique and repeated sample events,
which can skew results, included addressing variations in monitoring practices, locating
human data entry errors, and backfilling missing and additional data not included in the
dataset.
Additionally, the NSQD v4.02 had not been updated with sampling events from
2014, which prompted personal communication with Holly Piza at Urban Drainage and
Flood Control District and Jane Clary at Wright Water Engineers to collect raw data for
the 2014 sampling season. This raw data collected through personal communication
enabled additional sample data for the ten sites provided in the NSQD. Additional raw
data was also provided that enabled manual data entry for missing parts of the sample
events recorded in the NSQD from past years. With the final criteria addressed, final
datasets on all sampling events for each of the ten sites was complete. Statistical analysis
results on the final dataset will be further explained in Chapter 4.
3.4 Geospatial Land Cover Analysis
3.4.1 Data Collection
Land cover for the continental United States was retrieved from the Multi-
Resolution Land Characteristics Consortium (MRLC) to determine land use types in
relation to the water quality sampling locations. Supported by the U.S. Department of the
Interior (DOI) and the U.S. Geological Survey (USGS), the MRLC specifies land covers
17


for uses ranging from agricultural to forest to developed areas. The 2001, 2006, and 2011
NLCD rasters, which provide 30 meter by 30 meter cells for each of the land covers in
the form of a TIFF file, were retrieved and analyzed using geographic information system
processes listed in this section. Rasters are images in the form of rectangular gridded
pixels or group of cells that represent some form matrix data structure such as elevation
or land cover.
There are two important items to note about the datasets that were retrieved. First,
the 1992 NLCD was unable to be used in the study because the classifications for land
cover were inconsistent with the 2001, 2006, and 2011 datasets (Vogelmann el al. 2001).
Second, to limit the scope of the study, only developed land cover areas were analyzed as
they are the key locations for urbanization. These developed areas, which include Low
Intensity, Medium Intensity, and High Intensity land classifications, were used in
conjunction with the collected non-point land use runoff data from the water quality
analysis. The three developed land use classes can be seen in Table 3 along with their
descriptions (Homer et al. 2015). Areas identified as Developed Open Space, which
include golf courses, parks, and developed recreational and aesthetic areas and account
for less than 20% imperviousness, were not included in the analysis as non-point land use
stormwater quality runoff can greatly vary from these locations. In relation to the water
quality analysis, Developed Low and Medium Intensity classifications correspond to
residential locations, while Developed High Intensity classifications were set to
correspond to commercial locations.
18


Table 3: NLCD Classifications for Developed Land Use
NLCD Class ID NLCD Classification Description for Developed Areas
22 Developed, Low Intensitv Areas with a mixture of constructed materials and vegetation. Impervious surfaces account for 20% to 49% percent of total cover. These areas most commonly include single-family housing units.
23 Developed, Medium Intensitv Areas with a mixture of constructed materials and vegetation. Impervious surfaces account for 50% to 79% of the total cover. These areas most commonly include single-family housing units.
24 Developed High Intensitv Highly developed areas where people reside or work in high numbers. Examples include apartment complexes, row houses and commercial/industrial. Impervious surfaces account for 80% to 100% of the total cover.
Source: Homer et al. (2015)
3.4.2 Geospatial Processing Methods
Having identified the sample locations, an arbitrary circular boundary was created
around Denver for the purpose of limiting the geographic scope under analysis. Centered
at the county courthouse in downtown Denver, this 25 km radius encompasses all
stormwater sampling locations. Once the study boundary was determined, relevant
shapefiles and file databases were added to the maps for the study. Shapefiles, which
include geographic features such as rivers, counties, and site locations, store vector data
relating to locations, shapes, and attributes for each feature dataset that was presented. In
order to maintain proper file management, a master geodatabase was created to store,
query and manage all GIS data for processed rasters and features in one central file
location. All features and rasters were converted to the same coordinate system, North
American Datum 1983 (NAD 1983) prior to any GIS processes or analyses.
19


Each NLCD raster was processed using the Clip Raster tool in ArcMap 10.3 for
the radial 25-km study area boundary that was created. Using this process for each of the
2001 NLCD, 2006 NLCD, and 2011 NLCD, three maps, shown in Figure 17-19, were
developed to show urbanization trends for the metropolitan Denver, Colorado. Results
and discussion of this geospatial analysis for Denver, Colorado will be further discussed
in Chapter 4.
Original attempts to convert this clipped raster to unique polygons were
unsuccessful as the GIS process returned over 180,000+ polygons for developed land use
classifications within the 25-km radius boundary for Denver, Colorado. In efforts to
address this problem and not enlarge cell raster sizes, a watershed was selected for which
this analysis would be applied to. A similar Clip Raster process was used for the
Lakewood Gulch watershed to reduce the raster to a manageable file size. After removing
unwanted classifications from the 2011 NLCD raster dataset, the Raster to Polygon tool
was used in ArcMap 10.3 to convert all cells classified as developed into 4,000+
polygons, which was much more manageable. An attribute table was then created for
these polygons that could be exported into an adaptable Excel file format to identify
developed land use areas and percentages for the watershed.
The last process in the geospatial analysis used the Union tool to join overlapping
features into a new output feature class. This tool used two inputs to address overlapping
layers. These two inputs included the polygon features created in the previous step for the
Lakewood Gulch watershed and a polygon shapefile of eight delineated sub-basins within
the watershed. Using the Union tool in ArcMap 10.3, final land use areas and
percentages could be calculated for the different sub-basins. These sub-basins land use
20


areas and percentages will be used for the non-point assessment matrix found in the next
part of this chapter.
3.5 Geospatial Non-Point Stormwater Assessment Matrix
3.5.1 Development of Geospatial Assessment Matrix
Using the following method, as described in the section below, determination of
watersheds and associated sub-basins potential on non-point urban impacts can be
evaluated. This method uses regional land cover data in relation with a geospatial
analysis. Watershed size, collected land use data relevant to a given watershed and event
rainfall are required for the matrix to be evaluated. Similar models have been developed
by the National Resources Conservation Service (NRCS) including the Agricultural Non-
Point Source Pollution Model (AGNPS), which focuses on non-point runoff from
agricultural areas (Bingner and Theurer 2009). This AGNPS model is not applicable to
large urban areas as it was developed to address different agricultural uses and associated
runoff properties. The model developed in this thesis through the two-phase analysis
described previously in the chapter introduces a simple area weighting and a discrete
matrix evaluation method for analyzing any number of constituents for different land uses
with a given study area.
The developed geospatial assessment matrix uses land use percentages as
weighting coefficients to identify urban watersheds that may be significant sources of
non-point stormwater runoff data. This model does not account for any engineered
stormwater quality treatment such as water quality detention basins, rain gardens or other
water quality treatment features. Additionally, this model can also be applied to smaller
sub-basins once an urban watershed is located, to evaluate sub-assessment analyses.
21


The following non-point geospatial assessment matrix requires four inputs. Two
variables, volume of runoff and watershed area, and two sets of data that are able to be
converted to the [m x n] and [nxl] matrix form. The first dataset converted to the
[m x n] matrix form includes water quality data for various constituents in the form of
mass per volume. The m variable represents rows which correspond to similar
constituents for different land use classifications. The n variable represents columns that
corresponds to different constituents with similar land use type. The second dataset
converted to the [nxl] matrix form includes geospatial land use data as a percentage
over the total study area. The n variable represents number of land use types within the
study area. Using this developed method, which is shown below in both single and
multiple constituent form, total non-point loading (TNPL) can be evaluated for a given
urban watershed and associated sub-basins.
Simple Area Weighting for Single Constituent
TNPLa = DA{C1LU1 + C2LU2 + C3LU3 + + CanLUn)
Discrete Area Weighting for Multiple Constituents
\TNPLal cai Ca2 Can -LU1
TNPLb = DA Cb2 lu2
TNPLn_ - h h ln - LUn_
Where,
D represents depth of runoff (M)
A represent area of the watershed or sub-basin (M2)
C represent different non-point stormwater runoff statistical data for different
constituents in analysis. (M/L3)
o Letters represent different constituents and numbers represent associated
land use type with that constituent
LU represents different land use percentages for study area [%]
TNPL represents Total Non-Point Loading for each of the constituents Ca, Cb, to
Cn over a watershed or sub-basin area [M]
22


3.5.2 Applications of Matrix using Lakewood Gulch Example
Application of the matrix that was developed to analyze non-point watershed
loading could be applied after statistical and geospatial analysis methods had been
determined. As mentioned previously, the South Platte River and its tributaries are listed
as impaired by the U.S. EPA in the most recent 2010 assessment report (U.S. EPA 2015).
Using the Lakewood Gulch tributary of the South Platte River, which is listed by the U.S.
EPA as impaired in Aquatic Life Warm Water Class 2 and Recreational Primary
Contact designated categorical uses, application of the non-point stormwater matrix could
be applied with data that was collected within the regional non-point stormwater analysis
summary (U.S. EPA 2015).
This straightforward model revolves around a geospatial analysis principle using
land use percentages as weighting coefficients to evaluate urban watersheds based on
non-point stormwater runoff data. This method can be used to identify watersheds that
are major contributors of pollutants from unregulated non-point locations for urban areas
and multiple constituents and multiple land use within the same watershed and/or smaller
sub-basins comparisons within a given watershed.
Using selected basins that drain to Lakewood Gulch tributary of the South Platte
River, land use polygons were created using similar geospatial process as for Denver,
Colorado. Thee developed land use polygon sets that relate to land use types, which can
be seen in Figure 12 and Table 4, were analyzed using the collected land use data for the
seven pollutants in the study. Using mean concentrations for the seven pollutants,
evaluation of the watershed loading could be completed with the Non-Point Stormwater
Geospatial Watershed Loading Matrix.
23


This matrix, which assumes no BMP implementation, uses 2011 NLCD for three
land use polygons within the Lakewood Gulch watershed and non-point stormwater mean
values from the water quality analysis. The example provided only evaluates a 2-year
rainfall depth event over the watershed area, which corresponds with regulated design
practices and treatment of minor storm event runoff for the area (UDFCD 2013).
Calculations for the Lakewood Gulch assessment matrix example can be found in the
Appendix A. 2.
24


4. Results and Discussion
4.1 Non-Point Stormwater Quality Results
Using the data collected for the stormwater sampling sites, final statistical and
correlative analyses can be determined for the two different land use types. The following
section compares median concentrations for commercial and residential areas, and
summarizes a correlation analysis of the constituents listed on Table 2. Figures and tables
provided throughout each section and are discussed in the text. Raw data that was
collected and used for non-point stormwater quality statistical analysis and correlations is
provided in the Appendix.
4.1.1 Statistical Analysis
The first analysis on the final data set of stormwater quality runoff constituents
uses box-and-whisker plots to show event sample mean, range, and normality, along with
another statistical evaluation for the seven pollutants within the two land use categories.
A t-Test was run on each of the land use datasets to determine corresponding p-values for
both one-tailed and two-tailed tests. P-values were used to determine significance of the
relationships between mean concentrations for residential and commercial land uses. A
null hypothesis was created that mean values are equal for each land use type and p-
values were used to either accept or reject this hypothesis. Plots can be seen below along
with corresponding tables that are provided for each of the seven constituents in the
analysis and are discussed within this section. Final comparisons to the U.S. EPA et al.
1983 for median and coefficient of variation values. Comparison of event median
concentrations with the 1983 NURP study are shown in Table 22 and Table 23.
25


A. Sediments: TSS
As mentioned in the Literature Review, Total Suspended Solids (TSS) are often
used to evaluate stormwater runoff in the form of the sediment pollutant category. There
were a total of 507 recorded TSS samples for the two land use sampling locations, 246
samples for residential and 261 samples for commercial. Multiple statistical categories
were analyzed to determine a final TSS classification summary for residential and
commercial land uses within Denver, Colorado. Results from the statistical analysis can
be seen in Table 4 and Table 5, along with a box-and-whisker plot for the TSS
constituent in Figure 2. Additionally, Table 6 provides individual site statistics calculated
for mean, standard deviation, max, and median TSS values.
First, event mean concentrations (EMCs) for residential and commercial sampling
locations were compared. TSS in residential areas had an EMC of 204 mg/L with a
standard deviation of 239 mg/L for the samples that were analyzed. TSS in commercial
areas had an EMC of 193 mg/L with a standard deviation of 333 mg/L for samples that
were analyzed. EMC values for residential areas were higher than commercial areas by
11 mg/L. To better understand significance of these event mean concentrations, t-tests
were run on each land use dataset to compare event mean values assuming unequal
variances. Using a null hypothesis of an event mean concentration difference of 0, 472
degrees of freedom were calculated with a 0.42 t-statistic for TSS residential and
commercial land use mean values. For the two-tailed t-test, p-values for the two mean
were calculated at 0.67. This TSS p-value is considerably greater than the level of
significance for the TSS analysis of 0.05, thus the null hypothesis was accepted for
residential and commercial TSS event mean concentrations.
26


Second, event maximum concentrations (EMxCs) for residential and commercial
sampling locations were compared. TSS in residential areas had an EMxC of 1310 mg/L.
TSS in commercial areas had an EMxC of 2260 mg/L of samples that were analyzed.
Commercial sampling location EMxC values were considerably greater than residential
sampling sites. Seven commercial samples recorded values greater than the residential
maximum of 1310 mg/L. Further analysis using the TSS plot suggest there were many
lower event values that were recorded to offset large events in commercial areas.
Last, event median concentrations for residential and commercial sampling
locations were compared. TSS in residential areas had an event median concentration
value of 121 mg/L. TSS in commercial areas had an event median concentration value of
66 mg/L. Event median concentration differed by 55 mg/L as residential sites had a
higher event median concentration when compared to commercial locations.
Overall, the final determination of TSS classification summary is as follows.
Total Suspended Solids is not significantly different between residential and commercial
areas andfurther analysis is needed. Although commercial loading recorded several
maximum values greater than that recorded at a residential location, event mean and
event median concentrations were higher for residential areas. This suggests that
variability within commercial sites is larger, and uncertainty for the commercial loading
needs to be addressed for individual site characteristics and events. In relation to the null
hypothesis that event mean concentrations are similar for residential and commercial
areas, the one tailed p-value is significant as it does in fact help show that the observed
mean values do not differ between the two land uses.
27


Table 4: Summary of Non-Point Runoff Statistics (TSS)
Summary of Non- TSS (mg/L)
Residential Commercial
Average 204 193
Standard Deviation 239 333
Sample Size 246 261
Median 120.5 66
Max 1310 2260
Standard Error 25.1 33.9
Table 5: Summary of t-Test (TSS)
TSS (units of mg/L) Residential Commercial
Mean 204 193
Variance 57100.4 110773.6
Observations 246 261
Hypothesized Mean Difference 0
Degrees of Freedom 472
t-statistic 0.4227
P(T<=t) one-tail 0.3364
t Critical one-tail 1.6481
P(T<=t) two-tail 0.6727
t Critical two-tail 1.9650
Note: t-Test for Two-Sample Assuming Unequal Variances
28


Total Suspended Solids [TSS]
All Sites
Figure 2: Box-and-Whisker Plots for Residential and Commercial Locations (TSS)
29


Table 6: Individual Site Statistics for TSS Analysis
Site ID Sample Size Average Standard Deviation Max Median
Residential Sampling Locations (measured in mg/L)
COLAIRIS 54 296.6 300.6 1310 168
CODEGRHE 29 202.4 192.1 814 157
CODEGRRE 23 272.9 296.9 1210 169.5
COAUSHCR 38 198.9 209.6 999 134
CODEORPO 102 142.4 189.9 1280 71.5
All Residential Sites 246 204 239 1310 121
Commercial Sampling Locations (measured in mg/L)
COACWWL3 19 74.0 14.1 302 43.7
COACW6W7 17 39.0 42.2 186 20.8
CODEWAWA 66 383.7 472.4 2260 211.5
COLASHOP 113 182.7 297.2 1940 64
COLAMOPA 46 53.0 80.6 450 21.5
All Commercial Sites 261 193 333 2260 66
B. Nutrients: TKN and NO2 NO3
Total Kjeldahl Nitrogen (TKN) and Nitrite+Nitrate (NO2+NO3) datasets were
used to evaluate stormwater runoff in nutrient pollutant category. There were a total of
416 recorded TKN samples for the two land use sampling locations, 196 samples for
residential and 220 samples for commercial. Additionally, there were a total of 435
recorded NO2+NO3 samples for the two land uses types, 226 samples for residential and
209 samples for commercial. Multiple statistical categories were analyzed to determine a
final nitrogen classification summary, which is based on TKN and NO2+NO3 sample
data, for residential and commercial land uses within Denver, Colorado. Results from the
statistical analysis can be seen in Tables 7-9 along with a box-and-whisker plot for the
TKN and NO2+NO3 constituents in Figure 3 and Figure 4. Additionally, individual site
30


statistics were also calculated for the mean, standard deviation, max and median TKN
and NO2+NO3 values. These results can be seen in Table 10 and Table 11.
First, event mean concentrations (EMCs) for residential and commercial sampling
locations were compared. TKN in residential areas had an EMC of 3.41 mg/L with a
standard deviation of 2.37 mg/L for the samples that were analyzed. TKN in commercial
areas had an EMC of 2.53 mg/L with a standard deviation of 2.44 mg/L for samples that
were analyzed. NO2+NO3 in residential areas had an EMC of 1.07 mg/L with a standard
deviation of 0.85 mg/L for the samples that were analyzed. NO2+NO3 in commercial
areas had an EMC of 0.70 mg/L with a standard deviation of 0.56 mg/L for samples that
were analyzed. To better understand significance of event mean concentrations for TKN
and NO2+NO3, t-tests were run on each land use dataset to compare event mean values
assuming unequal variances. Using a null hypothesis of an event mean concentration
difference of 0, 411 degrees of freedom were calculated with a 3.69 t-statistic for TKN
residential and commercial land use mean values. For the two-tailed t-test, TKN p-values
were in the order of lxlO"4. The p-value is much less than the level of significance for the
TKN analysis of 0.05, thus the null hypothesis was rejected for residential and
commercial EMC values. Again, using a null hypothesis of an event mean concentration
difference of 0, 393 degrees of freedom were calculated with a 5.32 t-statistic for
NO2+NO3 residential and commercial areas. For the two-tailed t-test, NO2+NO3 p-values
were in the order of lxlO"7. Since p-values is much less than the level of significance for
the NO2+NO3 analysis of 0.05, thus we can reject the null hypothesis that residential and
commercial NO2+NO3EMC values are similar.
31


Second, event maximum concentrations (EMxCs) for residential and commercial
sampling locations were compared for both nitrogen constituents. TKN in residential
areas had an EMxC of 13.4 mg/L. TKN in commercial areas had an EMxC of 23.4 mg/L.
NO2+NO3 in residential areas had an EMxC of 8.32 mg/L. NO2+NO3 in commercial
areas had an EMxC of 3.61 mg/L of samples that were analyzed. Further analysis of both
the TKN and NO2+NO3 suggest that the maximum value for commercial might have been
some type of extreme event, however, no comments are provided for this event.
Last, event median concentrations for residential and commercial sampling
locations were compared for the two nitrogen constituents. TKN in residential areas had
an event median concentration value of 2.85 mg/L. TKN in commercial areas had an
event median concentration value of 2.00 mg/L. NO2+NO3 in residential areas had an
event median concentration value of 0.91 mg/L. NO2+NO3 in commercial areas had an
event median concentration value of 0.58 mg/L. Event median concentration for
residential sites had a higher event median concentration for both TKN and NO2+NO3
when compared to commercial locations. TKN and NO2+NO3 event median
concentration values were 1.4 times and 1.6 times higher in residential areas than in
commercial areas, respectively.
Overall, the final determination of nitrogen classification summary is as follows.
Higher loading in the form of TKN and NO2+NO3 can be seen in residential sampling
locations. All phases of the nitrogen statistical analysis provide supporting evidence
TKN and NO2+NO3 concentrations are higher in residential areas. Both plots show
residential upper and lower bounds are higher compared to commercial sampling
locations. This further supports the notion that the maximum data point within the
32


commercial location might have been an extreme or misreported event. However, this
data is still included in the analysis, thus mean values might be slightly higher than
expected for commercial land use in this study. In relation to both constituent null
hypotheses checked, p-values were less than level of significance of 0.05 and both null
hypotheses were rejected.
Table 7: Summary of Non-Point Runoff Statistics (TKN and NO2+NO3)
Summary of Non- TKN (mg/L) N02+NO3 (mg/L as N)
Point Runoff Data Residential Commercial Residential Commercial
Average 3.41 2.53 1.07 0.70
Standard Deviation 2.37 2.44 0.85 0.56
Sample Size 196 220 226 209
Median 2.85 2.00 0.91 0.58
Max 13.4 23.9 8.32 3.61
Standard Error 0.3 0.3 0.1 0.1
Table 8: Summary of One-Tailed and Two-Tailed t-Test (TKN)
TKN (units of mg/L) Residential Commercial
Mean 3.41 1 2.53
Variance 5.63 5.95
Observations 196 220
Hypothesized Mean Difference 0
Degrees of Freedom 411
t-statistic 3.69
P(T<=t) one-tail lxl O'4
t Critical one-tail 1.6486
P(T<=t) two-tail lxl O'4
t Critical two-tail 1.9658
Note: t-Test Two-Sample Assuming Unequal Variances
33


Table 9: Summary of One-Tailed and Two-Tailed t-Test (NO2+NO3)
NO2+NO3 (units of mg/L) Residential Commercial
Mean 1.07 0.70
Variance 0.72 0.31
Observations 226 209
Hypothesized Mean Difference 0
Degrees of Freedom 393
t-statistic 5.32
P(T<=t) one-tail lxl O'7
t Critical one-tail 1.6487
P(T<=t) two-tail lxl O'7
t Critical two-tail 1.9660
Note: t-Test Two-Sample Assuming Unequal Variances
34


14
12
10
8
3
'Sfc
J,
6
4
2
0
igure
Total Kjedhal Nitrogen [TKN]
All Sites

o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
8
o
H b

H b

Residential Commercial
: Comparison of TKNfor Residential and Commercial Sampling Locations
35


Nitrite + Nitrate [N02+NCK]
All Sites
Figure 4: Comparison ofN02+N03 for Residential and Commercial Sampling Locations
36


Table 10: Individual Site Statistics for TKN Analysis
Site ID Sample Size Average Standard Deviation Max Median
Residential Sampling Locations (measured in mg/L)
COLAIRIS 54 3.24 2.19 13.40 2.95
CODEGRHE 12 3.79 1.70 7.50 3.73
CODEGRRE 13 3.41 1.94 7.20 3.45
COAUSHCR 37 3.83 3.10 12.20 2.90
CODEORPO 80 3.26 2.28 13.10 2.76
All Residential Sites 196 3.41 2.37 13.4 2.85
Commercial Sampling Locations (measured in mg/L)
COACWWL3 0 N/A
COACW6W7 0 N/A
CODEWAWA 66 3.69 3.30 23.90 2.85
COLASHOP 108 1.93 1.30 8.90 1.60
COLAMOPA 46 2.30 2.52 13.00 1.45
All Commercial Sites 220 2.53 2.44 23.90 2.00
Table 11: Individual Site Statistics for NO 2+NO 3 Analysis
Site ID Sample Size Average Standard Deviation Max Median
Residential Sampling Locations (measured in mg/L)
COLAIRIS 54 0.74 0.45 2.57 0.66
CODEGRHE 25 1.24 0.70 3.46 1.04
CODEGRRE 21 1.02 0.56 2.09 1.02
COAUSHCR 25 2.11 1.72 8.32 1.71
CODEORPO 101 0.95 0.52 2.70 0.85
All Residential Sites 226 1.07 0.85 8.32 0.91
Commercial Sampling Locations (measured in 1 mg/L)
COACWWL3 0 N/A
COACW6W7 0 N/A
CODEWAWA 60 0.64 0.65 3.61 0.55
COLASHOP 102 0.63 0.47 2.98 0.49
COLAMOPA 47 0.96 0.56 2.88 0.85
All Commercial Sites 209 0.70 0.56 3.61 0.58
37


C. Nutrients: Total Phosphorus and Dissolved Phosphorus
Similar to nitrogen and mentioned in the Literature Review, elemental phosphorus
is required by natural ecosystems to provide beneficial uses in a healthy state. Since
phosphorus reacts with microbes found in soil and consumed through phytoremediative
processes, Total Phosphorus (TP) and Dissolved Phosphorus (DP) datasets were used to
evaluate stormwater runoff in the nutrient pollutant category. There were a total of 416
recorded TP samples for the two land use sampling locations, 235 samples for residential
and 267 samples for commercial. Additionally, there were a total of 364 recorded DP
samples for the two land uses types, 192 samples for residential and 172 samples for
commercial. Multiple statistical categories were analyzed to determine a final phosphorus
category classification summary, which is based on TP and DP sample data, for
residential and commercial land uses within Denver, Colorado. Orthophosphate was not
considered in this analysis. Results from the statistical analysis can be seen in Tables 12-
14 along with a box-and-whisker plot for the TP and DP constituents in Figure 5 and
Figure 6. Additionally, individual site statistics were calculated and can be seen in Table
15 and Table 16.
First, event mean concentrations (EMCs) for residential and commercial sampling
locations were compared. TP in residential areas had an EMC of 0.51 mg/L with a
standard deviation of 0.33 mg/L for the samples that were analyzed. TP in commercial
areas had an EMC of 0.28 mg/L with a standard deviation of 0.38 mg/L for samples that
were analyzed. DP in residential areas had an EMC of 0.25 mg/L with a standard
deviation of 0.22 mg/L for the samples that were analyzed. DP in commercial areas had
an EMC of 0.09 mg/L with a standard deviation of 0.16 mg/L for samples that were
analyzed. EMC values for both phosphorus pollutant forms analyzed, TP and DP, were
38


higher in residential areas when compared to commercial areas. To further understand
significance of event mean concentrations for TP and DP, t-tests were run on each land
use dataset to compare event mean values assuming unequal variances. Using a null
hypothesis of an event mean concentration difference of 0, 500 degrees of freedom were
calculated with a 7.38 t-statistic for TP residential and commercial land use mean values.
For the two-tailed t-test, TP p-values were in the order of lxlO"13. The p-value is much
less than the level of significance for the TP analysis of 0.05, thus the null hypothesis was
rejected for residential and commercial EMC values. Again, using a null hypothesis of an
event mean concentration difference of 0, 344 degrees of freedom were calculated with a
8.04 t-statistic for DP residential and commercial areas. For the two-tailed t-test, DP p-
values were in the order of lxlO"13. Since p-values were less than the level of significance
for the DP analysis of 0.05, we can reject the null hypothesis that residential and
commercial DP EMC values are similar.
Second, event maximum concentrations (EMxCs) for residential and commercial
sampling locations were compared for both pollutant constituents. TP in residential areas
had an EMxC of 1.91 mg/L. TP in commercial areas had an EMxC of 4.44 mg/L. DP in
residential areas had an EMxC of 1.62 mg/L. DP in commercial areas had an EMxC of
1.59 mg/L of samples that were analyzed. Further analysis of the TP plot suggest that the
maximum value for commercial might have been some sort of extreme event, however it
corresponds with the same sample event data recorded in the TKN EMxC that had
occurred. Both TP and DP values were plotted on the same y-axis as some values are not
shown in the box-and-whisker plots.
39


Last, event median concentrations for residential and commercial sampling
locations were compared for the two pollutant constituents. TP in residential areas had an
event median concentration value of 0.43 mg/L. TP in commercial areas had an event
median concentration value of 0.17 mg/L. DP in residential areas had an event median
concentration value of 0.18 mg/L. DP in commercial areas had an event median
concentration value of 0.05 mg/L. Event median concentration for residential sites had a
higher event median concentration for both TP and DP when compared to commercial
locations. TP and DP event median concentration values were 2.5 times and 3.6 times
higher in residential areas than in commercial areas, respectively.
Overall, the final determination of phosphorus category classification summary is
as follows. Higher loading in the form of TP and DP can be seen in residential sampling
locations. All phases of the phosphorus statistical analysis provide supporting evidence
nutrients in the form of TP and DP concentrations are higher in residential areas. Both
plots, TP and DP, show residential upper and lower bounds greater than commercial
sampling locations. Similar to nitrogen, the same axis for both box-and-whisker plots
were used for TP and DP. In relation to both constituent null hypotheses checked, p-
values were much less than the significance level set, thus no similar relationship
between means can be seen between residential and commercial areas.
40


Table 12: Summary of Non-Point Stormwater Statistics (TP and DP)
Summary of Non- Total Phosphorus Dissolved Phosphorus
Point Runoff Data (mg/L) (mg/L)
Residential Commercial Residential Commercial
Average 0.51 0.28 0.25 0.09
Standard Deviation 0.33 0.38 0.22 0.16
Sample Size 235 267 192 172
Median 0.43 0.17 0.18 0.05
Max 1.91 4.44 1.62 1.59
Table 13: Summary of One-Tailed and Two-Tailed t-Test (TP)
TP (units of mg/L) Residential Commercial
Mean 0.51 0.28
Variance 0.11 0.14
Observations 235 267
Hypothesized Mean Difference 0
Degrees of Freedom 500
t-Statistic 7.38
P(T<=t) one-tail lxlO"13
t Critical one-tail 1.6479
P(T<=t) two-tail lxlO"12
t Critical two-tail 1.9647
Note: t-Test Two-Sample Assuming Unequal Variances
Table 14: Summary of One-Tailed and Two-Tailed t-Test (DP)
DP (units of mg/L) Residential Commercial
Mean 0.25 0.09
Variance 0.05 0.02
Observations 192 172
Hypothesized Mean Difference 0
Degrees of Freedom 344
t-Statistic 8.04
P(T<=t) one-tail lxlO'14
t Critical one-tail 1.6493
P(T<=t) two-tail lxlO'14
t Critical two-tail 1.9669
Note: t-Test Two-Sample Assuming Unequal Variances
41


Total Phosphorus (mg/L)
Total Phosphorus [TP]
All Sites
Figure 5: Comparison of TP for Residential and Commercial Sampling Locations
42


Dissolved Phosphorus (mg/L)
Dissolved Phosphorus [DP]
All Sites
Figure 6: Comparison of DP for Residential and Commercial Sampling Locations
43


Table 15: Individual Site Statistics for TP Analysis
Site ID Sample Size Average Standard Deviation Max Median
Residential Sampling Locations (measured in mg/L)
COLAIRIS 46 0.56 0.41 1.91 0.42
CODEGRHE 25 0.52 0.24 1.03 0.53
CODEGRRE 22 0.62 0.42 1.71 0.44
COAUSHCR 46 0.51 0.35 1.83 0.44
CODEORPO 96 0.46 0.28 1.73 0.38
All Residential Sites 235 0.51 0.33 1.91 0.43
Commercial Sampling Locations (measured in mg/L)
COACWWL3 19 0.17 0.15 0.54 0.12
COACW6W7 17 0.26 0.10 0.40 0.30
CODEWAWA 66 0.57 0.64 4.44 0.39
COLASHOP 119 0.19 0.16 0.93 0.15
COLAMOPA 46 0.13 0.11 0.50 0.10
All Commercial Sites 267 0.28 0.38 4.44 0.17
Table 16: Individual Site Statistics for DP Analysis
Site ID Sample Size Average Standard Deviation Max Median
Residential Sampling Locations
COLAIRIS 48 0.18 0.19 1.23 0.13
CODEGRHE 23 0.27 0.13 0.71 0.24
CODEGRRE 19 0.32 0.27 1.33 0.25
COAUSHCR - - - - -
CODEORPO 102 0.26 0.23 1.62 0.17
All Residential Sites 192 0.25 0.22 1.62 0.18
Commercial Sampling Locations
COACWWL3 17 0.04 0.03 0.12 0.04
COACW6W7 15 0.18 0.07 0.30 0.16
CODEWAWA 63 0.12 0.21 1.59 0.06
COLASHOP 77 0.06 0.11 0.97 0.04
COLAMOPA 0 - - - -
All Commercial Sites 172 0.09 0.16 1.59 0.05
44


I). Metals: Total Copper and Total Zinc
As mentioned in the Literature Review, metals and other chemical pollutants are
common remnants from human activity and industrial processes. Total Copper and Total
Zinc sampling datasets for the ten sampling locations were used to evaluate stormwater
runoff in the form of the metals pollutant category. Although additional metals could
have been selected, these two metals had prominent data records and are two common
metal pollutants found in urban areas. There were a total of 271 recorded Total Copper
samples for the two land use sampling locations, 186 samples for residential and 85
samples for commercial. Additionally, there were a total of 238 recorded Total Zinc
samples for the two land uses types, 155 samples for residential and 83 samples for
commercial. Multiple statistical categories were analyzed to determine a final metal
category classification summary, which is based on Total Copper and Total Zinc sample
data, for residential and commercial land uses within Denver, Colorado. Results from the
statistical analysis can be seen in Tables 17-19 along with a box-and-whisker plot for the
Total Copper and Total Zinc constituents in Figure 7 and Figure 8. Additionally,
individual site statistics were calculated and can be seen in Table 20 and Table 21.
First, event mean concentrations (EMCs) for residential and commercial sampling
locations were compared. Total Copper in residential areas had an EMC of 20.1 pg/L
with a standard deviation of 18.2 pg/L for the samples that were analyzed. Total Copper
in commercial areas had an EMC of 27.6 pg/L with a standard deviation of 34.8 pg/L for
samples that were analyzed. Total Zinc in residential areas had an EMC of 104.0 pg/L
with a standard deviation of 87.2 pg/L for the samples that were analyzed. Total Zinc in
commercial areas had an EMC of 143.1 pg/L with a standard deviation of 216.9 pg/L for
samples that were analyzed. EMC values for both metal pollutant forms analyzed, Total
45


Copper and Total Zinc, were higher in commercial areas when compared to residential
areas. To further understand significance of event mean concentrations for Total Copper
and Total Zinc, t-tests were run on each land use dataset to compare event mean values
assuming unequal variances. Using a null hypothesis of an event mean concentration
difference of 0, 106 degrees of freedom were calculated with a -1.87 t-statistic for Total
Copper residential and commercial event mean values. For the two-tailed t-test, Total
Copper p-values were 0.06. The p-value is slightly greater than the level of significance
for the Total Copper analysis of 0.05, thus the null hypothesis was accepted for
residential and commercial EMC values. Again, using a null hypothesis of an event mean
concentration difference of 0, 96 degrees of freedom were calculated with an -1.58 t-
statistic for Total Zinc residential and commercial area event mean values. For the two-
tailed t-test, Total Zinc p-values were 0.12. Since p-values were greater than the
significance level set of 0.05, we can accept the null hypothesis that residential and
commercial Total Zinc EMC values are similar.
Second, event maximum concentrations (EMxCs) for residential and commercial
sampling locations were compared for both metal constituents. Total Copper in
residential areas had an EMxC of 130 pg/L. Total Copper in commercial areas had an
EMxC of 224 pg/L. Total Zinc in residential areas had an EMxC of 590 pg/L. Total Zinc
in commercial areas had an EMxC of 1440 pg/L of samples that were analyzed. In
correlation with mean values, further analysis using the data plot suggest both metal
constituents, Total Copper and Total Zinc, are likely to have higher extreme events occur
at commercial locations.
46


Last, event median concentrations for residential and commercial sampling
locations were compared for the two metal constituents. Total Copper in residential areas
had an event median concentration value of 14.3 pg/L. Total Copper in commercial areas
had an event median concentration value of 16.3 pg/L. Total Zinc in residential areas had
an event median concentration value of 80 pg/L. Total Zinc in commercial areas had an
event median concentration value of 72 mg/L. Event median concentration differed for
each of the metal constituents. Total Copper for commercial sites had higher event
median concentrations for residential sites when compared to commercial locations. On
the other hand, Total Zinc event median concentrations had higher values recorded at
commercial sampling sites. This provides additional insight as to why p-values were
both greater than the significance level set of 0.05 when analyzing the EMC values.
Overall, the final determination of metals category classification summary is as
follows. Both Total Copper and Total Zinc have similar loading in residential and
commercial areas andfurther analysis is needed. All phases of the statistical analysis,
event mean, median, and maximum values, in addition to the box-and-whisker plot,
suggest concentrations at commercial areas will be higher when compared to residential
areas. However, in relation to both constituent null hypotheses checked, two-tailed t-test
p-values were greater than the 0.05 significance level set. Only minor relationships each
metal constituent EMC mean p-values for the two land uses are shown as these null
hypotheses were accepted. Additionally, further analysis would look to collect more data
from commercial locations as residential sampling areas had almost two times as many
sampling events.
47


Table 17: Summary of Non-Point Stormwater Statistical Analysis (Cu and Zn)
Summary of Non- Total Copper (pg/L) Total Zinc (pg/L)
Point Runoff Residential Commercial Residential Commercial
Average 20.1 27.6 104.0 143.1
Standard Deviation 18.2 34.8 87.2 216.9
Sample Size 186 85 155 83
Median 14.3 15.95 80 72.1
Max 130 224 590 1440
Standard Error 2.2 6.2 11.5 39.2
Table 18: Summary of One-Tailed and Two-Tailed t-Test (Total Copper)
Total Copper (units of pg/L) Residential Commercial
Mean 20.1 27.6
Variance 330.9 1209.1
Observations 186 85
Hypothesized Mean Difference 0
Degrees of Freedom 106
t-statistic -1.87
P(T<=t) one-tail 0.0319
t Critical one-tail 1.6594
P(T<=t) two-tail 0.0637
t Critical two-tail 1.9826
Note: t-Test: Two-Sample Assuming Unequal Variances
48


Table 19: Summary of One-Tailed and Two-Tailed 1-Test (Total Zinc)
Total Zinc (units of pg/L) Residential Commercial
Mean 104.0 143.1
Variance 7602.8 47025.5
Observations 155 83
Hypothesized Mean Difference 0
Degrees of Freedom 96
t-statistic -1.58
P(T<=t) one-tail 0.06
t Critical one-tail 1.660
P(T<=t) two-tail 0.12
t Critical two-tail 1.985
Note: t-Test: Two-Sample Assuming Unequal Variances
49


Total Copper [Cu]
All Sites
hJ
3
-
Q.
O.
O
U
o
H
Residential
Commercial
Figure 7: Comparison of Total Copper for Residential and Commercial Sampling
50


Total Zinc (fig/L)
Total Zinc [Zn]
All Sites
Residential
Commercial
Figure 8: Comparison of Total Zinc for Residential and Commercial Sampling Locations
51


Table 20: Individual Site Statistics for Total Copper Analysis
Site ID Sample Size Average Standard Deviation Max Median
Residential Sampling Locations (measured in PgIL)
COLAIRIS 54 18.7 11.8 53.6 14.7
CODEGRHE 0 - - - -
CODEGRRE 0 - - - -
COAUSHCR 54 30.1 26.3 130.0 24.5
CODEORPO 78 14.1 10.8 73.6 10.9
All Residential Sites 186 20.1 18.2 130.0 14.3
Commercial Sampling Locations (measured in fig/L)
COACWWL3 0 - - - -
COACW6W7 0 - - - -
CODEWAWA 42 45.4 42.1 224.0 33.1
COLASHOP 43 10.2 7.9 39.5 8.4
COLAMOPA 0 - - - -
All Commercial Sites 85 27.6 34.8 224.0 16.3
Table 21: Individual Site Statistics for Total Zinc Analysis
Site ID Sample Size Average Standard Deviation Max Median
Residential Sampling Locations (measured in PgIL)
COLAIRIS 46 126.3 91.3 468.0 97.4
CODEGRHE 0 - - - -
CODEGRRE 0 - - - -
COAUSHCR 53 115.2 107.6 590.0 100.0
CODEORPO 56 75.1 46.1 240.0 65.6
All Residential Sites 155 104.0 87.2 590.0 80.0
Commercial Sampling Locations (measured in fig/L)
COACWWL3 0 - - - -
COACW6W7 0 - - - -
CODEWAWA 40 247.1 273.4 1440.0 167.5
COLASHOP 43 46.4 50.1 215.0 34.4
COLAMOPA 0 - - - -
All Commercial Sites 83 143.1 216.9 1440.0 72.1
52


Table 22: Comparison to 1983 NURP Study (Residential)
Pollutant Units US EPA 1983 This Study 2015
Median cov Median COV
TSS mg/L 101 0.96 121 1.17
TKN mg/L 1.9 0.73 2.85 0.70
N02+N03 mg/L 0.736 0.83 0.91 0.79
Total Phosphorus mg/L 0.38 0.69 0.43 0.65
Diss. Phosphorus mg/L 0.14 0.46 0.1775 0.89
Total Copper gg/L 33 0.99 14 0.91
Total Zinc gg/L 135 0.84 80 0.84
Table 23: Comparison to 1983 NURP Study (Commercial)
Pollutant Units US EPA 1983 This Study 2015
Median COV Median COV
TSS mg/L 69 0.85 66 1.72
TKN mg/L 1.18 0.43 2 0.96
N02+N03 mg/L 0.572 0.48 0.58 0.79
Total Phosphorus mg/L 0.201 0.67 0.17 1.37
Diss. Phosphorus mg/L 0.8 0.71 0.05 1.75
Total Copper gg/L 29 0.81 16 1.72
Total Zinc gg/L 226 1.07 72 0.96
53


4.1.2 Correlation Analysis
The second statistical analysis provides correlations among different constituents
for individual sampling events. These plots, which use a linear regression analysis for
correlations, were developed for each of the three stormwater pollutant categories,
sediments, nutrients, and metals. Regression was used to evaluate measured vs. predicted
relationships between constituents and to aid in understanding trends and associated
errors of these linear relationships. Only relevant linear trend lines for measured and
predicted values were provided and can be seen on Figures 9-11 within this section.
Additionally, the linear estimate (LINEST) function in Excel was used to calculate a
statistical array that was used to analyze relationships of the linear regression for each
correlation. The developed plots, along with the statistical analyses evaluation, were used
to show positive, negative or null linear correlations among different constituents within
these pollutant categories. Raw data that was used to develop these correlations is
provided in the Appendix.
A. Constituent vs. Constituent
Correlations between similar constituents were analyzed using linear regression.
Three categories, which include nitrogen, phosphorus, and metals, were used to create
constituent vs. constituent correlations as seen in Figures 9-11. The LINEST function was
used to evaluate the relationships of these regressions.
First, the nitrogen category, which uses TKN vs. NO2+NO3, had R-squared values
of 0.004 for residential and 0.22 for commercial. Both land use categories have positive
linear slopes, however, from the LINEST evaluation, the slope of the TKN vs. NO2+NO3
regression was not significantly different from zero, indicating a lack of correlation..
54


When compared to the residential LINEST results, commercial results show less of a
standard error on the slope, however, they are still not significant. The final plot of the
nitrogen regression analysis is shown in Figure 9. Two samples, one commercial and one
residential, were not shown on the chart for plotting reasons. These values had a TKN
greater than 14 mg/L or N02+N03 EMC values greater than 4.0 mg/L. Overall, no
significant correlations can be seen between the two nitrogen constituents.
Second, the phosphorus category, which uses Total vs. Dissolved Phosphorus, had
R-squared values of 0.30 for residential and 0.36 for commercial. Both land use
categories have positive linear slopes and from the LINEST evaluation, these slopes are
significant. Both residential and commercial have standard errors within ten percent of
the slope. With these LINEST results and analysis of the plot, we can see there are in fact
positive correlations that are significant with respect to the standard error of the mean for
both residential and commercial areas.
Last, the metal category, which uses Total Zinc vs. Total Copper, had R-squared
values of 0.33 for residential and 0.93 for commercial. Both land uses, residential and
commercial, had positive slopes. From the LINEST evaluation, both slopes are
significant. The standard error of the residential slope is about 11%, while the standard
error of the commercial slope is about 3%. Results for each land use type using the
calculated LINEST array can be seen in Tables 24-26. With the LINEST results and
analysis of Figures 9-11, positive correlations between metals can be seen for both land
uses, especially commercial areas.
55


Table 24: LINEST Results for TKNvs. N02+N03
TKN vs. N02+N03 LINEST Statistics Variable Residential Commercial
Slope m 0.1703 2.1103
Standard Error (Slope) Seslope 0.1978 0.2743
Intercept b 3.2914 1.0812
Standard Error (Intercept) Seintercept 0.279 0.2452
Coefficient of Determination R2 0.0041 0.2232
Standard Error (y estimate) sey 2.43 2.20
F Statistic F 0.742 59.177
Degrees of Freedom df 182 206
Regression Sum of Squares SSreg 4.36 286.19
Residual Sum of Squares SSresid 1070.3 996.2
Table 25: LINEST Results for TP vs. DP
TP vs. DP
LINEST Statistics Variable Residential Commercial
Slope m 0.8149 1.7019
Standard Error (Slope) SCslope 0.0935 0.1727
Intercept b 0.3074 0.1781
Standard Error (Intercept) Seintercept 0.0305 0.0310
Coefficient of Determination R2 0.3029 0.3636
Standard Error (y estimate) sey 0.27 0.35
F Statistic F 76.0 97.1
Degrees of Freedom df 175 170
Regression Sum of Squares SSreg 5.65 12.07
Residual Sum of Squares SSresid 13.0 21.1
56


Table 26: LINEST Results for Total Zinc us. Total Copper
Total Zinc vs. Total Copper
LINEST Statistics Variable Residential Commercial
Slope m 2.6146 5.9862
Standard Error (Slope) Seslope 0.3038 0.1906
Intercept b 47.2118 -24.3649
Standard Error (Intercept) Seintercept 8.8602 9.1446
Coefficient of Determination R2 0.3291 0.9337
Standard Error (y estimate) sey 71.87 58.34
F Statistic F 74.1 986.6
Degrees of Freedom df 151 70
Regression Sum of Squares SSreg 382579 3358023
Residual Sum of Squares SSresid 779948 238259
57


TKN (mg/L)
TKN vs. N02+N03 Correlations using Regression Analysis
All Sites
A Residential o Commercial ----------------Linear (Commercial)
Figure 9: Constituent Correlations (TKN and NO2+NO3)
58


Total Phosphorus (mg/L)
TP vs. DP- Correlations using Regression Analysis
All Sites
A Residential o Commercial
----- Linear (Residential) ------Linear (Commercial)
Figure 10: Constituent Correlations (TP vs. DP)
59


Total Zinc (pg/L)
Total Zinc vs. Total Copper Correlations using Regression Analysis
All Sites
A Residential o Commercial
Linear (Residential) ----------Linear (Commercial)
Figure 11: Constituent Correlations (Total Zinc us. Total Copper)
60


B. TSS us. Nutrients
Evaluation of the different nutrients with respect to TSS was selected to determine
if relationships could be determined between the two pollutant categories, sediments and
nutrients. In order to account for high magnitudes of TSS loading, a log scale was used
for the y-axis for all plots when comparing the TSS values to the different pollutant
categories.
The first TSS vs. Nutrients correlation compares TSS and TKN for the two
sampling land uses. R2 values were 0.21 for residential and 0.28 for commercial. Both
land use categories have positive linear slopes and from the LINEST evaluation, both
slopes are significant. Both residential and commercial have standard errors on the slopes
equaling 16% and 11% respectively. With these LINEST results and analysis of the plot,
we can see there are in fact positive correlations that are significant with respect to the
standard error of the mean for both residential and commercial areas. Results for each
land use type using the calculated LINEST array can be seen in Table 27 along with the
Regression plot in Figure 12.
The second correlation TSS vs. Nutrients compares TSS vs. NO2+NO3 for the two
sampling land uses. R2 values were 0.02 for residential and 0.001 for commercial.
Correlation for both residential and commercial sampling locations were similar and
classified as no correlation, which corresponds with R2 values near 0. Slopes for the two
land use categories vary greatly. Residential shows a negative slope, while commercial
shows a positive slope for the TSS vs. N02+N03 evaluation. From the LINEST
evaluation, the standard error of the slope in relation to the residential sampling location
proves there is no correlation between the two constituents. Both residential and
commercial standard error on the slopes are greater than the slope itself, which shows
61


these results are not significant. Results for each land use type using the calculated
LINEST array can be seen in Table 28 along with the Regression plot in Figure 13.
The last TSS vs. Nutrients correlation compares TSS vs. Total Phosphorus for the
two sampling land uses. R2 values were 0.28 for residential and 0.51 for commercial.
Both land use categories have positive linear slopes and from the LINEST evaluation, the
standard errors on these slopes are significant. Both residential and commercial have
standard errors on the slopes equaling 9% and 6%, respectively. With these LINEST
results and analysis of the plot, we can see there are in fact positive correlations that are
significant with respect to the standard error of the mean for both residential and
commercial areas. To further this, stronger positive correlations are seen between the
Total Zinc and Total Copper for commercial areas, when compared to residential areas.
Results for each land use type using the calculated LINEST array can be seen in Table 29
along with the Regression plot in Figure 14.
62


Table 27: LINEST Results for TSS vs. TKN
TSS vs. TKN
LINEST Statistics Variable Residential Commercial
Slope m 39.6 79.8
Standard Error (Slope) SCslope 6.38 8.54
Intercept b 63.2 14.0
Standard Error (Intercept) Seintercept 26.66 30.45
Coefficient of Determination R2 0.17 0.30
Standard Error (y estimate) sey 209.8 305.5
F Statistic F 38.6 87.3
Degrees of Freedom df 187 207
Regression Sum of Squares SSreg 1.70E+06 8.15E+06
Residual Sum of Squares SSresid 8.23E+06 1.93E+07
Table 28: LINEST Results for TSS vs. N02+N03
TSS vs. N02+N03
LINEST Statistics Variable Residential Commercial
Slope m -16.03 25.30
Standard Error (Slope) SCslope 19.36 46.28
Intercept b 227.5 208.1
Standard Error (Intercept) Seintercept 26.3 42.2
Coefficient of Determination R2 0.003 0.002
Standard Error (y estimate) sey 245.3 371.2
F Statistic F 0.69 0.30
Degrees of Freedom df 222 197
Regression Sum of Squares SSreg 4.13E+04 4.12E+04
Residual Sum of Squares SSresid 1.34E+07 2.71E+07
63


Table 29: LINEST Results for TSS vs. TP
TSS vs. TP
LINEST Statistics Variable Residential Commercial
Slope m 432.2 668.3
Standard Error (Slope) SCslope 38.7 36.9
Intercept b -10.9 10.8
Standard Error (Intercept) Seintercept 23.9 17.2
Coefficient of Determination R2 0.4 0.6
Standard Error (y estimate) sey 195.9 221.9
F Statistic F 124.4 328.6
Degrees of Freedom df 222 255
Regression Sum of Squares SSreg 4.78E+06 1.62E+07
Residual Sum of Squares SSresid 8.52E+06 1.26E+07
64


TSS (mg/L)
TSS vs. TKN- Correlations using Regression Analysis
All Sites
A Residential O Commercial
Linear (Residential)------Linear (Commercial)
Figure 12: TSS vs. Nutrients Correlations (TKN)
65


TSS (mg/L)
TSS vs. N02+N03- Correlations using Regression Analysis
All Sites
10000.00 -|
1000.00
100.00
10.00
oA
A O O O
A
A
1.00 H-----1------1------1------1------1-----1------1------1
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
N02+N03 (mg/L)
A Residential o Commercial
Figure 13: TSS vs. Nutrients Correlations (NO2+NO3)
9.00
66


TSS (mg/L)
TSS vs. TP- Correlations using Regression Analysis
All Sites
10000.00
y = 668.28X + 10.841
R2 = 0.5631
1000.00 -O
o
0.0
o o' 0 > o
6aa
(2k
.Ar'
100.00
10.00

O
432.2x- 10.939
R2 = 0.3592
1.00
0.00
1.00 2.00 3.00
Total Phosphorus (mg/L))
4.00
5.00
A Residential O Commercial
Linear (Residential)------Linear (Commercial)
Figure 14: TSS uv. Nutrients Correlation (Total Phosphorus)
67


C. TSS us. Metals
Both metals, copper and zinc, were evaluated with respect to TSS to determine
relations between the different constituents. Similar to before, the TSS uses a logarithmic
y-axis to account for orders of magnitude.
The first comparison uses TSS vs. Total Copper. R2 values for this correlation
were 0.29 for residential and 0.59 for commercial. Both land use categories have positive
linear slopes and from the LINEST evaluation, both slopes are significant. Both
residential and commercial have standard errors on the slopes equaling 19% and 7%,
respectively. With these LINEST results and analysis of the plot, we can see there are in
fact positive correlations that are significant with respect to the standard error of the mean
for both residential and commercial areas. To further this, stronger positive correlations
are seen between the TSS and Total Copper for commercial areas, when compared to
residential areas. Results for each land use type using the calculated LINEST array can be
seen in Table 30 along with the Regression plot in Figure 15.
The second comparison uses TSS vs. Total Zinc. R2 values for this correlation
were 0.29 for residential and 0.77 for commercial. Both land use categories have positive
linear slopes and from the LINEST evaluation, the standard errors on these slopes are
significant. Both residential and commercial have standard errors on the slopes equaling
11% and 6%, respectively. With these LINEST results and analysis of the plot, we can
see there are in fact positive correlations that are significant with respect to the standard
error of the mean for both residential and commercial areas. To further this, stronger
positive correlations are seen between the TSS and Total Zinc for commercial areas,
when compared to residential areas. Results for each land use type using the calculated
LINEST array can be seen in Table 31 along with the Regression plot in Figure 16.
68


All equations for each of the two metals are displayed on the plots for each of the
land use types. Significant standard errors on the slopes and strong positive correlations
were seen in commercial areas as R2 values were 0.59 for TSS vs. Total Copper and 0.77
for TSS vs. Total Zinc, thus positive correlations can be seen between both metals.
Table 30: LINESTResults for TSS vs. Total Copper
TSS vs. Total Copper
LINEST Statistics Variable Residential Commercial
Slope m 5.2 10.9
Standard Error (Slope) SCslope 1.0 0.8
Intercept b 109.1 10.0
Standard Error (Intercept) Seintercept 25.8 33.6
Coefficient of Determination R2 0.13 0.71
Standard Error (y estimate) sey 229.9 241.7
F Statistic F 26.2 201.6
Degrees of Freedom df 168 82
Regression Sum of Squares SSreg 1.38E+06 1.18E+07
Residual Sum of Squares SSresid 8.88E+06 4.79E+06
Table 31: LINEST Results for TSS vs. Total Zinc
TSS vs. Total Zinc
LINEST Statistics Variable Residential Commercial
Slope m 1.64 1.88
Standard Error (Slope) SCslope 0.18 0.12
Intercept b 50.5 49.5
Standard Error (Intercept) Seintercept 24.8 31.7
Coefficient of Determination R2 0.37 0.79
Standard Error (y estimate) sey 191.2 216.2
F Statistic F 81.6 263.0
Degrees of Freedom df 138 69
Regression Sum of Squares SSreg 2.98E+06 1.23E+07
Residual Sum of Squares SSresid 5.04E+06 3.23E+06
69


TSS (mg/L)
TSS vs. Total Copper- Correlations using Regression Analysis
All Sites
A Residential O Commercial
Linear (Residential)------Linear (Commercial)
Figure 15: TSS us. Metals Correlation (Total Copper)
70


TSS (mg/L)
TSS vs. Total Zinc Correlations using Regression Analysis
All Sites
10000.00
y = 1.8781X +49.508
R2 = 0.7921
1000.00
O
.-er'

aa
Q A Oa 1.6365X + 50.52
* ' R2 = 0.3716
100.00
10.00
1.00 -i-------1---------1--------1---------1---------1--------1---------1--------1
0.00 200.00 400.00 600.00 800.00 1000.00 1200.00 1400.00 1600.00
Total Zinc (mg/L))
A Residential O Commercial
Linear (Residential)------Linear (Commercial)
Figure 16: TSS us. Metals Correlation (Total Zinc)
71


4.2 Geospatial Land Cover Results
The geospatial analysis using the process listed in the methods section proved to
be a great tool for analyzing land use over the large urban area of Denver, Colorado. In
2011, developed land use accounted for 49.6% of land cover in the form of Low
Intensity, Medium Intensity, and High Intensity NLCD classifications. Developed Low
Intensity, which includes land use ranging from 20% to 49% imperviousness and is used
as a residential in this study, makes up 522 km2 of the total 1964 km2 study area.
Developed Medium Intensity, which includes land use ranging from 50% to 79%
imperviousness and is used as a residential in this study, makes up 327 km2 of the total
1964 km2 study area. Developed High Intensity, which ranges from 80% to 100%
imperviousness and is used as a commercial land use, makes up 125 km2 of the total 1964
km2 study area.
When compared to the 2001 land cover dataset, land cover has slightly increased
for all three developed land cover classes, Low Intensity, Medium Intensity, and High
Intensity. Developed Low Intensity has increased by 0.4%, Developed Medium
Intensity has increased by 2.9%, and Developed High Intensity has increased by 1.3%;
all with respect to the radial study boundary. Results for the 2001, 2006, and 2011 land
cover can be seen in the Table 32 and Figures 17-19. This geospatial analysis, which was
applied to the 1964 km2 study area for Denver, Colorado, was used to show developed
land use increases from urbanization for 2001, 2006, and 2011. Further application of this
geospatial analysis method that was used will be provided in the next section in relation
to the geospatial assessment matrix that was developed. This application uses the
developed non-point watershed loading matrix to analyze the sub-basins within the
72


Lakewood Gulch watershed and identify residential and commercial areas contributing to
high urban pollutant runoff.
Table 32: NLCD Analysis of Developed Areas within 25-km Radial Boundary
NLCD
Class ID
Classification
Description for
Developed Areas
(From NLCD
Legend)
Area
(km2)
%of
Total
Area
. % of . % of
Area ... . Area ... .
.. lotal ,, ,, lotal
(km) Area (lflrf) Area
Developed Land
LTse Change
(2001 to 2011)
Area
22 Develoned. Low Intensitv 515.3 26.2% 524.4 26.7% 522.4 26.6% 7.1 0.4%
23 Develoned. Medium Intensitv 269.5 13.7% 310.4 15.8% 327.1 16.7% 57.6 2.9%
24 Develoned Ilisih Intensitv 100.2 5.1% 115.4 5.9% 125.3 6.4% 25.1 1.3%
- Other 1,079 54.9% 1,014 51.6% 989 50.4%
TOTALS 1,964 100% 1,964 100% 1,964 100%
Note: Based on site classifications from the NLCD Legend, NLCD Class 22 and 23 were used as residential areas and NLCD Class 24 was used as the commercial/industrial area in relation non-point stormwater data.
73


Legend
2001 Developed Land Cover
Value
| | Developed (LowIntensity)
Developed (Medium Intensity)
Developed (High Intensity)
Surface Water Features
South Platte River
\fcjor Drainage way
| | Lakes
| | Study Boundary
NAD 1983
241
Kilometers
S
Figure 17: Developed Land Cover Map for 25-km Radius around Denver, CO (2001)
74


Developed (Medium Intensity)
Developed (High Intensity)
Surface Water Features
South Platte River
\£j or Drainage way
| | Lakes
| | Study Boundary
NAD 1983
241
Kilometers
Figure 18: Developed Land Cover Map for 25-km Radius around Denver, CO (2006)
75


NAD 1983
Figure 19: Developed Land Cover Map for 25-km Radius around Denver, CO (2011)
76


4.3 Geospatial Non-Point Stormwater Assessment Matrix
4.3.1 Application of Matrix to Lakewood Gulch Study Area
As mentioned previously, the Lakewood Gulch tributary of the South Platte River
is classified by the U.S. EPA as impaired in two categorical uses including Aquatic Life
Warm Water and Recreational Primary Contact. Lakewood Gulch is located just west of
central Denver. This watershed and associated sub-basins was selected in response to the
discussion and determination that the study area is very near fully developed and no new
developments are planned.
Without implementation of any BMP, the Lakewood Gulch would experience
excessive loading over the watershed from the three non-point land uses as listed in the
matrix. With further analysis using this matrix, eight sub-basins that were delineated by
UDFCD using geoprocessed digital elevations maps were compared to one another to
identify locations within this watershed that could potential be key contributors to the
urban stormwater impairments in the Lakewood Gulch drainageway. These results can
be seen in Tables 28-29 and in Figures 20-21.
From this analysis and application of the matrix on a sub-basin level, Sub-Basin
3, Sub-Basin 5, and Sub-Basin 7 are the key locations within the watershed where action
should be implemented. Sub-Basin 5, which has the largest area for all three land uses
(Residential A, Residential B, and Commercial), would provide the highest pollutant
loadings in the form of non-point runoff. If enough research is collected for the areas
shown in grey in Figure 33 and Figure 34, a complete non-point watershed loading
profile can be developed for the study watershed draining to Lakewood Gulch.
77


With a complete analysis, Sub-Basin 5 would serve a top priority for
implementation of BMP sites. Use of LID such as rain gardens for small residential
areas, implementation of large vegetated water quality detention basins in large
residential neighbors, and selection of soil media below permeable pavements to remove
high loads of metals; all result in efforts that can be made to reduce non-point runoff
pollutants that are directly carried into streams.
Table 33: Lakewood Gulch Geospatial Analysis Example using Polygons
NLCD Class ID NLCD Classification Description for Developed Areas Land Use Type Variable Area (km2) %of Total Area
22 Developed, Low Intensity RES. 1 Land Use 1 LU1 20.5 40.3%
23 Developed, Medium Intensity RES. 2 Land Use 2 -LU2 10.3 20.2%
24 Developed High Intensity COM. Land Use 3 -LU3 4.7 9.1%
- Other N/A N/A 15.6 31.4%
78


Table 34: Geospatial Summary for Sub-Basins for Lakewood Gulch Matrix Application
Cull Sub- %of Land Use 1 Residential A Land Use 2 Residential B Land Use 3 Commercial Developed Land Cover Totals
Basins Area (km2) Total Area Area (km2) %of Sub- Basin Area (km2) %of Sub- Basin Area (km2) %of Sub- Basin Total Area (km2) % Developed
1 3.5 6.8% 1.7 48.6% 0.8 22.5% 0.2 6.1% 2.7 77.2%
2 5.1 9.9% 2.7 52.5% 0.8 14.9% 0.2 4.2% 3.6 71.5%
3 7.4 14.5% 3.2 43.0% 1.6 22.1% 0.7 9.3% 5.5 74.4%
4 4.2 8.3% 1.8 42.3% 0.9 22.0% 0.5 12.3% 3.2 76.6%
5 9.7 19.0% 3.9 39.9% 2.3 23.3% 1.5 15.6% 7.6 78.8%
6 5.6 10.9% 2.7 48.6% 0.8 15.0% 0.1 2.5% 3.7 66.2%
7 9.4 18.5% 2.7 28.7% 1.6 17.4% 0.9 9.1% 5.2 55.2%
8 6.3 12.3% 2.0 31.4% 1.5 23.3% 0.5 8.5% 4.0 63.2%
Average 6.4 13% 2.6 41.9% 1.3 20.1% 0.6 8.4% 4.4 70.4%
Entire Watershed 51.1 100% 20.5 40.2% 10.3 20.1% 4.7 9.1% 35.5 69.6%
(/)
P
40.000
35.000
30.000
25.000
20.000
15.000
10.000
5,000
0
Sub-Basin Loading Comparison for TSS
Using 2-yr rainfall event
5
TSS
Figure 20: Comparison of Sub-Basin TSS Loading
79


Figure 21: Lakewood Gulch Watershed Map
80
Kilometers


Figure 22: Lakewood Gulch Watershed Map with Delineated Sub-Basins
NAD 1983


4.3.2 Validation of Matrix with Mixed Land Use Location
Validation of the matrix required returning to the NSQD to locate a mixed use
sampling location from which predicted and measured values could be compared. This
sampling location followed the similar site selection process, however, 100% residential
and 100% commercial, industrial and institutional were not used as a mixed land use
sampling location criteria. Using this process, the North Avenue at Denver Federal
Center was selected for validation as it provided a complete sampling record for two
years. This 0.28 km2 sampling area is composed of 33% residential, 30% commercial and
37% open space land use. There were twenty sampling events collected at the North
Avenue location from 1980 to 1981. The database for these locations provides the storm
event runoff that was used in the matrix for the depth of runoff variable for each event.
The relationship between predicted and measured values is shown in Figure 23.
82


Predicted Values (mg)
Matrix Validation using Data Collected from North
Avenue at Denver Federal Center
Predicted vs. Measured Values
Figure 23: Validation of Matrix with Mixed Use Sampling Location
83


5 Conclusion
The following study provides a straightforward methodology and analysis process
for applying regional non-point stormwater quality data to associate land cover datasets
to analyze impacts from urbanization in developing areas. Research on unknown sources
of contaminations continue to drive the U.S. EPA to set stricter regulations as
amendments to the Clean Water Act look to restore natural streams to their once
glorified, beneficial, and fishable state. Although efforts on reviving damaged or lifeless
ecosystems remain costly and time extensive, growing surface water impairments
continue to drive urban planners, environmentalists, and water resource engineers to look
for new methods to treat urban impacts from decades of urban pollution.
The following study provides residential and commercial urban stormwater
quality analysis that support to final determination summary for each urban land use
constituent. The regional stormwater quality analysis results provide general statistics,
regression analysis of correlations, and t-tests to determine p-values for associated non-
equal variances. These results are applied to land cover classifications to develop
geospatial analysis process to determine land use percentages for a given study area.
Once all data is collected, the regional matrix can then be applied for the non-
point stormwater assessment analysis. Results from this matrix application are used to
evaluate and identify urban areas in need of BMP implementation strategies to reduce
impacts felt from developed urban.
Overall, the following methods used in this study help understand impacts of
different land uses with respect to regional water quality data. Using a geospatial
approach to evaluate collected runoff data, identification and prioritization of watershed
84


and sub-basins can be determined to locate areas contributing to high pollutant loads to
nearby impaired surface waters. Residential and commercial developments act as key
factors limiting these impairments to be restored and repaired. With proper planning,
evaluation and selection of optimal BMP designs and strategies, installments and retrofits
of efficient stormwater treatment systems can play key factor moving forward as
populations and developments continue to grow. Learning to understand impacts prior to
installments can be a key factor for evaluating stormwater runoff. Not only to reduce
extreme flows during heavy storm events with the use of detention basins, but also to
produce treatment during minor events that work to protect and restore principles of the
urban regime. Implementation of water quality treatment features and integrated
prevention plans will continue to expand as protection of natural lands and restoration of
the urban regime will continue to remain a key objective for engineers, planners, and
environmentalists who develop design and standards for innovative sustainable
engineering process moving forward.
85


REFERENCES
Benedict, M. and McMahon, E. (2006) Green Infrastructure: Smart Conservation for the
21st Century. Island Press. Washington, DC.
Bingner, R. and Theurer, F. (2009) Agricultural Non-Point Source Pollution Model.
(AGNPS). U.S. Department of Agriculture: Natural Resources Conservation
Service. . (1 June 2015).
Carey, R., Hochmuth, G., Martinez, C., Boyer, T., Dukes, M., Toor, G., and Cisar, J.
(2013). Evaluating nutrient impacts in urban watersheds: Challenges and research
opportunities. Environmental Pollution, 173, 138-149.
Chau, H. (2009) Green Infrastructure for Los Angeles: Addressing Urban Runoff and
Water Supply through Low Impact Development. (Doctoral Dissertation). Los
Angeles, CA.
Dietz, M. (2007) Low Impact Development Practices: A Review of Current Research
and Recommendations for Future Directions. Water, Air and Soil Pollution, 186(1-
4), 351-363.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and
Wickham, J., (2011). Completion of the 2006 National Land Cover Database for the
Conterminous United States. Photogrammetric Engineering and Remote Sensing,
77(9), 858-864.
Guo, J. (2009). Retrofitting Detention Basin with Water Quality Control Pool. Journal
of Irrigation and Drainage Engineering. 135(5), 671-675.
Hathaway, J., Tucker, R., Spooner, J., and Hunt, W. (2012). A traditional analysis of the
first flush effect for nutrients in stormwater runoff from two small urban
catchments. Water, Air, & Soil Pollution, 223(9), 5903-5915.
Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow,
A., VanDriel, J.N., and Wickham, J. (2007). Completion of the 2001 National Land
Cover Database for the Conterminous United States. Photogrammetric Engineering
and Remote Sensing, 73(4), 337-341.
Homer, C. G., Dewitz, J. A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J.,
Herold, N. D., Wickham, J. D., and Megown, K., (2015) Completion of the 2011
National Land Cover Database for the conterminous United States-Representing a
decade of land cover change information. Photogrammetric Engineering and
Remote Sensing, 81(5), 345-354.
Kaplowitz, M. and Lupi, F. (2012). Stakeholder preferences for best management
practices for non-point source pollution and stormwater control. Journal of
Landscape and Urban Planning, 104(3-4), 364-372.
Lee, J. and Bang, K. (2000). Characterization of urban stormwater runoff. Water
Research, 34(6), 1773-1780.
86


Maestre, A. and Pitt, R. (2006). Identification of Significant Factors Affecting
Stormwater Quality Using the NSQD. Stormwater and Urban Systems Modeling
Proceedings, Monograph 14. 287-326.
Mitchell, G. (2005) Mapping hazard from urban non-point pollution: a screening model
to support sustainable urban drainage planning. Journal of Environmental
Management. 74(1), 1-9.
National Research Council (1999) AW Strategies for Americas Watersheds. National
Academies Press. Washington, D.C.
National Research Council. (2009a) Urban Stormwater Management in the United States.
The National Academies Press. Washington, D.C.
National Research Council. (2009b) Assessing the TMDL Approach to Water Quality
Management. National Academies Press. Washington, D.C.
Park, M. and Park, M. (2015) Evaluation of Watershed Susceptibility to Contaminants
of Emerging Concern. Journal of American Water Works Association. 107(4), 96-
97.
Pitt, R., Maestre, A., Morquecho, R., Brown, T., Swann, C., Cappiella, K., and Schueler,
T. (2003). Evaluation of NPDES Phase I municipal stormwater monitoring data.
National conference on urban stormwater: enhancing the programs at the local
level. (February 2003)
Pitt, R. (2015) The National Stormwater Quality Database (NSQD) Version 4.04.
Center for Watershed Protection. Ellicott, City, MA.
Regas, D. (2005). Guidance for 2006 Assessment, Listing and Reporting Requirements
Pursuant to Sections 303(d), 305(b) and 314 of the Clean Water Act. U.S. EPA
Office of Wetlands, Oceans and Watersheds. (15 May, 2015)
Seattle Public Utilities (2009). "Quality Assurance Project Plan NPDES Phase I
Municipal Stormwater Permit Stormwater Treatment Best Management Practices
Evaluation. Seattle, WA.
Smith, A., Alexander, B., and Schwarz, G. (2003). Natural background concentrations
of nutrients in streams and rivers of the conterminous United States. Environmental
Science & Technology, 37(14), 3039-3047.
Son, J., Goodwin, S., and Carlson, K. (2015) Total Phosphorus Input to the Cache la
Poudre River in Northern Colorado. Water Environment Research. 87(2), 169-178.
Stevens, M.R., and Slaughter, C.B. (2012) Summary and evaluation of the quality of
stormwater in Denver, Colorado, 2006-2010: U.S. Geological Survey Open-File
Report 2012-1052. U.S. Geological Survey. Reston, VA.
U.S. EPA (1986) Methodology for Analysis of Detention Basins for Control of Urban
Runoff Quality. U.S. Environmental Protection Agency, EPA 440/5-87-001.
87


U.S. EPA (2007a) Developing Your Stormwater Pollution Prevention Plan: A Guide for
Construction Sites. U.S. Environmental Protection Agency. Washington, DC.
U.S. EPA (2007b). Reducing Stormwater Costs through Low Impact Development
(LID) Strategies and Practices. Nonpoint Source Control Branch. Washington, DC.
U.S. EPA (2013) My WATERS Mapper. U.S. EPA Office of Water.
U.S. EPA (2015) Colorado Water Quality Assessment Report. Watershed Assessment,
Tracking & Environmental Results.
. (15 May 2015)
Urban Drainage and Flood Control District. (2013) Urban Storm Drainage Criteria
Manual, Volume 3 Best Management Practices. Denver, CO.
Vogelmann, J., Howard, S., Yang, L., Larson, C., Wylie, B., and Van Driel, J. (2001)
Completion of the 1990's National Land Cover Data Set for the conterminous
United States Photogrammetric Engineering and Remote Sensing. (67), 650-662.
Walsh, C., Fletcher, T., and Burns, M. (2012) Urban Stormwater Runoff: A New Class
of Environmental Flow Problem. PLoS ONE. 7(9): e45814.
doi: 10.1371/joumal.pone.0045814
Walker, D., and Hurl, S. (2002). The reduction of heavy metals in a stormwater
wetland. Ecological Engineering, 18(4), 407-414.
Wright Water Engineers, Inc. and Geosyntec Consultants. (2012) International Best
Management Practice (IBMP) Database. Denver, CO. (5
March 2015).
Wright Water Engineers, Inc., Geosyntec Consultants, Pitt, R. and Roesner, L. (2013)
Regulation 85 Data Gap Analysis Report. Urban Drainage and Flood Control
District. Denver, Colorado.
Wulliman, J. and Thomas, P. (2005) Learning from Nature: Reducing Urban
Stormwater Impacts. Lake Line Magazine. Urban Drainage and Flood Control
District, Denver, CO.
88


APPENDIX
A. 1 Raw Data
The dataset provided below is the final data set that was used for the statistical and
correlative analyses. This dataset includes the sampling site locations, land use types,
locations and all constituent data from the stormwater databases and personal
communication. The datasets are separated by residential and commercial.
Residential
ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L) TKN (mgO NOz+NOs (mg/L) TP (mg/L) DP (mg/L) Total Copper (P-g/L) Total Zinc (M-g/L)
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 7/9/1996 4.00 2.00 6.00 0.19 4.00 40.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 4/23/1995 9.00 0.50 1.63 0.08 5.00 2.50
COAUSHC R Shop Creek Wedand-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 5/9/1996 13.00 1.00 1.12 0.12 5.00 2.50
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 6/1/1991 14.00 0.25 5.00 30.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/29/1990 18.00 1.80 0.32 10.00 50.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 6/22/1991 20.00 0.37 30.00 60.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/21/1991 30.00 0.34 100.00 30.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 9/20/1995 40.00 1.10 0.76 0.22 9.00 40.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/12/1992 44.00 10.00 190.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 10/4/1995 59.00 1.60 1.78 0.25 5.00 80.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 9/19/1995 61.00 3.20 1.62 0.51 10.00 2.50
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 5/29/1990 72.00 1.50 0.39 10.00 50.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 8/3/1991 72.00 0.16 50.00 150.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 5/16/1995 76.00 2.00 3.23 0.18 11.00 60.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/2/1992 88.00 30.00 200.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 6/28/1995 103.00 0.30 8.32 0.20 5.00 2.50
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 6/19/1990 122.00 3.00 0.54 60.00 10.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 10/22/1995 124.00 10.10 1.77 1.04 20.00 130.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 9/6/1996 128.00 3.90 1.85 0.45 23.00 110.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 6/22/1994 140.00 10.50 3.16 1.03 12.00 130.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 5/25/1996 150.00 3.20 1.71 0.46 39.00 140.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 4/18/1995 153.00 3.70 0.93 0.53 5.00 2.50
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/5/1990 164.00 2.30 0.37 50.00 110.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 7/13/1995 175.00 11.60 1.68 1.00 130.00 590.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 8/18/1995 182.00 4.40 2.85 0.58 17.00 110.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 9/29/1995 203.00 3.00 2.16 0.10 23.00 150.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 6/22/1996 221.00 12.20 0.08 0.61 20.00 220.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/8/1990 292.00 2.00 0.47 30.00 130.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/15/1992 294.00 80.00 240.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 4/10/1995 306.00 1.80 0.55 0.43 5.00 2.50
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 6/13/1995 324.00 6.20 1.82 0.75 26.00 200.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 8/17/1990 352.00 2.50 1.08 30.00 60.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 4/23/1995 389.00 4.60 1.68 1.83 5.00 2.50
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 8/23/1996 438.00 6.90 1.81 0.81 31.00 250.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 6/15/1996 501.00 5.70 1.32 0.71 5.00 160.00
COAUSHC R Shop Creek Wetland-Pond 1995- 97 SI 95-97 39.629 1 104.7415 RE 9/11/1996 521.00 1.40 1.04 0.11 16.00 170.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/14/1990 656.00 4.10 0.87 50.00 200.00
COAUSHC R Shop Creek Wetland-Pond 1990- 94 SI 39.629 1 104.7415 RE 7/19/1994 999.00 7.30 1.45 1.16 68.00 420.00
89


Full Text

PAGE 1

GEOSPATIAL EVALUATION OF NON POINT STORMWATER RUNOFF FOR DEVELOPED RESIDENTIAL AND COMMERCIAL LAND USES: CASE STUDY OF DENVER, COLORADO by BRIK R. ZIVKOVICH B.S., University of Pittsburgh, 2013 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 Master of Science Civil Engineering 2015

PAGE 2

2015 BRIK R. ZIVKOVICH ALL RIGHTS RESERVE D

PAGE 3

ii This thesis for the Master of Science degree by Brik R. Zivkovich has been approved for the Department of Civil Engineering by James C.Y. Guo, Chair David C. Mays Wesley E. Marshall August 1 st 2015

PAGE 4

iii Zivkovich, Brik R. (M.S., Civil Engineering) Geospatial Evaluation of Non Point Stormwater Runoff for Developed Residential and Commercial Land Uses: Case Study of Denver, Colorado Thesis directed by Associa te Pro f essor David C. Mays. ABSTRACT Understanding and evaluating urban impacts on natural ecosystem processes has become an increasingly complex task for engineers, planners and environmental scientists. As built environments continue to grow, increased human activity and large scale development drastically stress receiving urban streams and lakes resulting in the current impai red and degraded state of surface waters. In response, integrated water quality management programs have been adopted t o address these unregulated non point sources by utilizing best management practices to treat this runoff as close to the source as possible. The following study provides a detailed statistical and geospatial analysis process to analyze how different land uses affect urban stream systems. Using metropolitan Denver, Colorado as a case study, this paper presents a general evaluation method to identify critical non point pollutant source watersheds and associated sub basins The two phase analysis led to the d evelopment of a non point stormwater assessment matrix that can be used to aid stormwater professionals to evaluate and specify retrofits of water quality features within urban areas The selected water quality features can be used reduce, capture and tre at stormwater runoff specific to pollutant loading prior to entering nearby urban surface waters as location is vital for optimizing pollutant reduction found in non point stormwater runoff. The form and content of this abstract are approved. I recommend i ts publication Approved: David C. Mays

PAGE 5

iv ACKNOWLEDGEMENTS Over the past two years, I have been truly overwhelmed and grateful for all the support and g uidance I have received since I have begun my graduate studies at the University of Colorado Denver. Not only has this university helped enhance a nd improve my engineering skill set, but it has become a second home away from family. There are many people I would like to thank that have helped me along the way as I wo uld not be in the seat I am today without their support and guidance. First, I would like to thank is my advisor, Dr. David Mays, for all of his support and guidance throughout my graduate studies at here at the University of Colorado Denver He has been a great mentor and professor that has provided endless support and advice in r elation to my coursework as well as my pushed me to reach my full ability as the past two years of classes and thesis meetings have enabled gr eat education experience for me. Next, I would like to thank my two committee members, Dr. James Guo and Dr. Wesley Marshall. Both have been influential professors who through their classes enabled this thesis topic to be performed. Additionally, both pro vided great advice and pressed me to look at different ways to address these obstacles in order to find an alternative way to analyze or think about a problem. Next, I would like to send out an extremely gracious thank you to all employees at Urban Drainage and Flood Control District. After acquiring a graduate internship in the master planning program at the start of my graduate studies, my skillset has exponentially grown a s I have been exposed to real world engineering problems that coincide perfectly with my studies. A spe cial thanks goes out to my bosses Ken MacKenzie, She a

PAGE 6

v Thomas and Holly Piza, as well as Julia Bailey, as they have helped provide advice and guidance whenever I had a question relating to my thesis work. Additionally, a special thanks also goes out to contacts met through this internship program including Jane Clary at Wright Water Engineers, Inc. and David Delagarza at RESPEC Engineering Last and most important, an unmeasurable thank you goes out to my family and friends who have continue to support me as I get one step closer to my career dreams. They have always been there to care and provide support for me in every decision that I make and I am truly grateful them in my life. Thank you again for everything be here without you.

PAGE 7

vi TABLE OF CONTENTS Chapter 1. Introduction ................................ ................................ ................................ ................. 1 1.1 Purpose of the Study ................................ ................................ ............................ 1 1.2 Scope of the Study ................................ ................................ ................................ 2 2. Literature Review ................................ ................................ ................................ ........ 3 2.1 Overview ................................ ................................ ................................ .............. 3 2.2 Regulations and Standards ................................ ................................ ................... 3 2.3 Stormwater Quality Databases ................................ ................................ ............. 5 2.4 Urbanization Impacts ................................ ................................ ........................... 6 2.5 Objectives of Case Study for Denver, Colorado ................................ .................. 8 3. Methods ................................ ................................ ................................ ..................... 10 3.1 Overview ................................ ................................ ................................ ............ 10 3.2 Site Selection ................................ ................................ ................................ ...... 12 3.3 Non Point Stormwater Quality Analysis ................................ ........................... 12 3.3.1 Data Selection ................................ ................................ ............................. 12 3.3.2 Selection of Constituent Data ................................ ................................ ..... 16 3.4 Geospatial Land Cover Analysis ................................ ................................ ........ 17 3.4.1 Data Collection ................................ ................................ ........................... 17 3.4.2 Geospatial Processing Methods ................................ ................................ .. 19 3.5 Geospatial Non Point Stormwater Assessment Matrix ................................ ...... 21 3.5.1 Development of Geospatial Assessment Matrix ................................ ......... 21 3.5.2 Applications of Matrix using Lakewood Gulch Example .......................... 23 4. Results and Discussion ................................ ................................ .............................. 25 4.1 Non Point Stormwater Quality Results ................................ .............................. 25 4.1.1 Statistical Analysis ................................ ................................ ...................... 25

PAGE 8

vii 5.1.1 Correlations ................................ ................................ ................................ 54 4.2 Geospatial Land Cover Results ................................ ................................ .......... 72 4.3 Geospatial Non Point Stormwater Assessment Matrix Results ......................... 77 5 Conclusion ................................ ................................ ................................ ................. 84 References ................................ ................................ ................................ ......................... 86 Appendix ................................ ................................ ................................ ........................... 89 A.1 Raw Data ................................ ................................ ................................ ................ 89 A.2 Calculations ................................ ................................ ................................ .......... 101

PAGE 9

viii L IST OF TABLES Table 1 : Final Site Locations and Descriptions from Selection Criteria ................................ .................. 14 2: Common Urban Stormwater Pollutants and Potential Sources ................................ ................. 16 3: NLCD Classifications for Developed Land Use ................................ ................................ ........ 19 4: Summary of Non Point Runoff Statistics (TSS) ................................ ................................ ........ 28 5: Summary of t Test (TSS) ................................ ................................ ................................ ........... 28 6: Individual Site Statistics for TSS Analysis ................................ ................................ ................ 30 7: Summary of Non Point Runoff Statistics (TKN and NO 2 +NO 3 ) ................................ .............. 33 8: Summary of One Tailed and Two Tailed t Test (TKN) ................................ ............................ 33 9: Summary of One Tailed and Two Tailed t Test (NO 2 +NO 3 ) ................................ .................... 34 10: Individual Site Statistics for TKN Analysis ................................ ................................ ............. 37 11: Individual Site Statistics for NO 2 +NO 3 Analysis ................................ ................................ .... 37 12: Summary of Non Point Stormwater Statistics (TP and DP) ................................ .................... 41 13: Summary of One Tailed and Two Tailed t Test (TP) ................................ ............................. 41 14: Summary of One Tailed and Two Tailed t Test (DP) ................................ ............................. 41 15: Individual Site Statistics for TP Analysis ................................ ................................ ................ 44 16: Individual Site Statistics for DP Analysis ................................ ................................ ................ 44 17: Summary of Non Point Stormwater Statistical Analysis (Cu and Zn) ................................ .... 48 18: Summary of One Tailed and Two Tailed t Test (Total Copper) ................................ ............. 48 19: Summary of One Tailed and Two Tailed t Test (Total Zinc) ................................ ................. 49 20: Individual Site Statistics for Total Copper Analysis ................................ ................................ 52 21: Individual Site Statistics for Total Zinc Analysis ................................ ................................ .... 52 22: Comparison to 1983 NURP Study (Residential) ................................ ................................ ..... 53 23: Comparison to 1983 NURP Study (Commercial) ................................ ................................ ... 53 24: LINEST Results for TKN vs. NO2+NO3 ................................ ................................ ................ 56 25: LINEST Results for TP vs. DP ................................ ................................ ................................ 56 26: LINEST Results for Total Zinc vs. Total Copper ................................ ................................ .... 57 27: LINEST Results for TSS vs. TKN ................................ ................................ ........................... 63 28: LINEST Results for TSS vs. NO2+NO3 ................................ ................................ ................. 63 29: LINEST Results for TSS vs. TP ................................ ................................ .............................. 64 30: LINEST Results for TSS vs. Total Copper ................................ ................................ .............. 69 31: LINEST Results for TSS vs. Total Zinc ................................ ................................ .................. 69 32: NLCD Analysis of Developed Areas within 25 km Radial Boundary ................................ .... 73 33: Lakewood Gulch Geospatial Analysis Example using Polygons ................................ ............ 78 34: Geospatial Summary for Sub Basins for Lakewood Gulch Matrix Application ..................... 79

PAGE 10

ix L IST OF FIGURES Figure 1: Residential and Commercial Sampling Locations for Denver, Colorado ................................ .. 15 2: Box and Whisker Plots for Residential and Commercial Locations (TSS) .............................. 29 3: Comparison of TKN for Residential and Commercial Sampling Locations ............................. 35 4: Comparison of NO 2 +NO 3 for Residential and Commercial Sampling Locations ..................... 36 5: Comparison of TP for Residential and Commercial Sampling Locations ................................ 42 6: Compari son of DP for Residential and Commercial Sampling Locations ................................ 43 7: Comparison of Total Copper for Residential and Commercial Sampling ................................ 50 8: Comparison of Total Zinc for Residential and Commercial Sampling Locations ..................... 51 9: Constituent Correlations (TKN and NO 2 +NO 3 ) ................................ ................................ ........ 58 10: Constituent Correlations (TP vs. DP) ................................ ................................ ...................... 59 11: Constituent Correlations (Total Zinc vs. Total Copper) ................................ .......................... 60 12: TSS vs. Nutrients Correlations (TKN) ................................ ................................ ..................... 65 13: TSS vs. Nutrients Correlations (NO 2 +NO 3 ) ................................ ................................ ............ 66 14: TSS vs. Nutrients Correlation (Tot al Phosphorus) ................................ ................................ .. 67 15: TSS vs. Metals Correlation (Total Copper) ................................ ................................ ............. 70 16: TSS vs. Metals Correlation (Total Zinc) ................................ ................................ .................. 71 17: Developed Land Cover Map for 25 km Radius around Denver, CO (2001) ........................... 74 18: Developed Land Cover Map for 25 km Radius around Denver, CO (2006) ........................... 75 19: Developed Land Cover Map for 25 km Radius around Denver, CO (2011) ........................... 76 20: Comparison of Sub Basin TSS Loading ................................ ................................ .................. 79 21: Lakewood Gulch Watershed Map ................................ ................................ ........................... 80 22: Lakewood Gulch Watershed Map with Delineated S ub Basins ................................ .............. 81 23: Validation of Matrix with Mixed Use Sampling Location ................................ ...................... 83

PAGE 11

x L IST OF ABBREVIATIONS AGNPS Agricultural Non Point Source Pollution Model BMP Best Management Practice CWA Clean Water Act CWP Center for Watershed Protection DP Dissolved Phosphorus EMC event mean concentration EMxC event maximum concentration LID Low Impact Development MS4 Municipal Separate Storm Sewer System NO 2 +NO 3 Nitrite plus Nitrate NPDES National Pollution Discharge Elimination System NSQD National Stormwater Quality Database NURP National Urban Runoff Program TKN Total Kjeldahl Nitrogen TMDL Total Maximum Daily Load TNPL Total Non Point Loading TP Total Phosphorus TSS Total Suspended Solids U.S. EPA U.S. Environmental Protection Agency UDFCD Urban Drainage and Flood Control District WEF Water Environment Federation WWE Wright Water Engineers, Inc.

PAGE 12

1 1. Introduction Leonardo da Vinci 1.1 Purpose of the Study Optimal watershed management practices and strategies to evaluate impai rments in rivers and streams have become a critical objective for water resource engineers and environmentalists to improve qualities of streams and lakes (Kaplowitz and Lupi 2012; Care y et al. 2013) Research and environmental assessments continue to provide strong evidence urbanization from increased human development creates excess nutrients, metals, and sediments that directly impact ecological properties and stability of surface wat ers (Mitchell 2005; Maestre and Pitt 2006; Walsh et al. 2012; Son et al. 2015; Park and Park 2015) Th ese excess pollutants create biogeochemical instabilities in ecosystems as urban deposition is washed into receiving stream s and lakes during storm events (Lee and Bang 20 0 0) In response, many cities have begun to adopt green infrastructure programs using sustainable urban planning techniques by implementing and retrofitting low impact development (LID) to address non point stormwater runoff and reduce the water footprint left from urban regimes ( U.S. EPA 1986; Benedict and McMahon 2006; Dietz 2007; U.S. EPA 2007 b ; Chau 2009) These LID designs, such as water quality detention basins, rain gardens, constructed wetlands, and grass swales, have prominent potential to treat non point runoff by permitting multi functional uses that rely on natural hydrologic principles. ( Guo 200 9 ; U.S. EPA 2007 a ; UDFCD 2013) Utilization of these natural hydrologic process es is key as LID designs can act as buffers for reducing pollutants that

PAGE 13

2 reach streams ( i.e utilization of sediment collection pads, metal reductions with select soil media, and o ptimizing phytoremediation processes for nutrients with select native vegetation) (Wulliman and Thomas 2005; UDFCD 2013) With proper design, planning and maintenance, these LID systems can effective ly treat urban runoff and reduce that magnitude of pollutants that enter urban streams and lakes. 1.2 Scope of the Study As built environments continue to grow, increased human activity and large scale development drastically stress receiving urban streams resulting in the current impaired and degraded state of urban surface waters. (U.S. EPA 2010; Stevens and Slaughter 2012) In response, integrated management and sustainable urban planning have become necessary to mitigate impacts from new developments. The following study provides a detailed statistical and geospatial analysis for two developed land uses that impact urban stream systems. Th ese land uses include developed residential and commercial areas that have imperviousness values greater than 20%. Using metropolitan Denver, Colorado as a case study, thi s thesis presents a general evaluation method to identify critical source watersheds and associated sub basins contributing high non point urban pollutant loads during the first flush of storm events This process can be used to aid stormwater professional s by developing a general methodology for locating retrofits of water quality features within urban areas that can be used reduce, capture and treat stormwater runoff prior to entering the surrounding natural streams and river systems. Additionally, t his process can be to aid environmental regulatory agencies and other interested parties in finding locations that can be key sources of degradation to nearby streams.

PAGE 14

3 2. Literature Review 2.1 Overview Although stormwater has long been rega rded as a major factor in flooding of urban areas, only since the 1980s have policymakers, engineers, and environmentalists recognized the additional roles that stormwater plays in the impairment of urban watersheds and natural ecosystems (National Researc h Council 2009a) Since the creation of the C lean Water Act (CWA) in the 1970s the U.S. Environmental Protection Agency (U.S. EPA) has continued to develop and establish new regulations to address discharges of pollutants into streams and rivers (Regas 2005; U.S. EPA 2007 a ) Motivation and incentive s to address surface water impairments have led designers, planners, and engineers to pollutants located in urban stormwater runoff 2.2 Regulations and Standards Urban p opulations have grown significantly since of creation of the Clean Water Act, so consequently stormwater pollutants have become the primary cause of impairment for urban surface waters (U.S. EPA 2015) The CWA, which originated to address point discharges into streams and lakes, continues to matu re rapidly as organizational research provides evidence to support integrated and sustainable water National Urban Runoff Program (NURP), s Environmental and Water Resources Institute (EWRI), the nonprofit organization Center for Watershed Protection (CWP), and the sanitary engineering organization Water Environment Federation (WEF) continue provid ing research and water quality studies that focus on lakes and rivers prone to stormwater pollutants left unregulated by the

PAGE 15

4 CWA These pollutants have become the main concerns as built environments such as residential developments and large commercial areas are a major contributors to impairments in urban surface waters (U.S. EPA 2007) In response to this organizational research many regulations and standards have been adopted under the CWA to address and evaluate these urban stormwater pollutants. Under Section 402 of the CWA, the National Pollutant Discharge Elimination System (NPDES) was adopted in order to set effluent based standards and ensure compliance from dischargers. NPDES Municipal Separate Stormwater Sewer System (MS4) permits are issued from state governments and approved by the US EPA as a way to implement regulations for dischargers. These MS4 permits have worked to address additional gaps that leave rivers and lakes prone to once overlooked unregulated source pollutants. Preliminary data summaries have been reported to characterize Phase 1 NPDES stormwater data for more than 200 municipalities throughout the coun try (Pitt et al. 2003). Additionally, u nder Sections 305(b) and 303(d) of the CWA, the U.S. EPA requests states to report on water quality conditions through National Water Quality Assessment Reports (Regas 2005) These bi annual integrated reports are used to assess streams and surface waters, identify impaired waters, and their causes, and track the status of actions being taken to restore impaired waters to their ambient state. Using the most recent Colorado Wa ter Quality Assessment Report from 2010, approximately 19% of assessed streams and 49% of assessed lakes are impaired, which triggers the requirement for a Total Maximum Daily Load (TMDL) plan (National Research Council 2009b; U S EPA 2010) Despite an investment of over hundreds of millions of dollars

PAGE 16

5 through the 1990s unregulated non point sources continue s to limit attaining the ultimate goal of swimmable and fishable waters (National Rese arch Council 1999). 2.3 Stormwater Quality Databases M ultiple stormwater quality datab ases have been created over past decades to assess stormwater runoff quality based on a number of different characteristics. Of these, the National Stormwater Quality Databa se (NSQD) and the International Stormwater BMP Database are two resources used to aid water quality assessments and research efforts to improve urban storm water conditions prior to entering urban surface waters Water quality assessments for streams and rivers is a continuous process that requires a thorough understanding of stream water managemen t and good monitoring practices (Maestre and Pitt 2006) With support from the U S EPA and the CWP, the NSQD was recently updated to the NSQD Version 4.02 and is the major resource for all stormwater data that has been collected over recent years. This NSQD is a national urban stormwater runoff database that serves as an important reso urce for urban runoff data, categorized by location, land use, years of record, along with several other characterizations (Pitt 2015) Additionally, the I nternational BMP database provides BMP stormwater monitoring data that has been collected for BMP mo nitored sites. These BMP site locations, more often than not, have a reference site at which the BMP system is compared with. This reference location, at which stormwater quality samples are also taken, is used as a resource within for the NSQD as it align s with urban stormwater data that has already been included in the database. Both databases will be prominent resources used in this study as they address urban runoff pollutants for different land use

PAGE 17

6 categories and can be used for understanding sampling site locations and descriptions that are found within both databases 2.4 Urbanization Impacts As mentioned in this chapter non point urban runoff from residential and commercial areas greatly contribute to the disruption of natural stream processes by discharging excess sediments, nutrients and metals into the drainageways. These urban pollutants, which have caused impairment s in urban surface waters, are commonly analyzed by their different stormwater pollutant categories. Although there is an extensive list of constituents that c ould have been analyzed such as organic elements, additional fuel by products, pathogens, or oth er contaminants of evolving concern, three common pollutant categories associated with stormwater runoff were selected (Maestre and Pitt 2007; Pitt 2015). Fi r st flush effects have become a primary area of research as certain pollutants have evaluated stron g relations to rainfall events however all monitoring practices for each of the data sites would be required (Hathaway et al. 2012). Three p ollutant categories, which will be used for this analysis, include sediments, nutrients, and metals These three categories are the most common urban stormwater pollutants that are studied in relation to stormwater quality runoff (UDFCD 2013; Pitt 2015; Son et al. 2015) The following section provides a background analysis of three common stormwater pol lutant categories and how urban ization impacts from these pollutants affect natural ecosystems. The first category is sediments. Total Suspended Solids (TSS) are often used to evaluate stormwater runoff in the form of the sediment pollutant category. TS S consists of loose particles that are carried during storm events into streams and lakes. These loose

PAGE 18

7 particulate s, which commonly are made up of sands, silts, clays and other small particles, are naturally occurring as waters erode natural landscapes and carry these particles into waters. This sediment transport cycle has been rapidly increasing as urban develo pment induce s additional sediment load ing into streams and rivers. These sediments, which are carried off of parking lots and large impervious areas through diverted stormwater systems, bring heavy stormwater sediments loads to specific points within streams and rivers. This excessive sed iment loading affects geomorphological properties within rivers resulting in adverse effects downstream of this location. With proper planning and management, the impact felts from urban sediments can be reduced as implementation of stormwater quality dete ntion basins allow loose particulate found in stormwater to settle to lower depths and removed at times when there is no surface water (Guo 2009) The second pollutant category is nutrients. Nutrients in the form of nitrogen and phosphorus compounds are of ten used to evaluate stormwater runoff in the form of the nutrient pollutant category. Nitrogen and phosphorus are both essential in providing healthy, natural ecosystems the ability to function at optimal capacities (Smith et al. 2003). Excessive nutrient s can create unwanted conditions for natural ecosystems as accelerated plant and algae growth can deplete oxygen levels in streams in rivers. In response to this degradation of surface water quality some or a combination of adverse effects can follow. Accompanied with the degradation of water quality, disruption of natural process such as removing large forested areas along river banks and large developments near lakes can cause significant impacts on natural processes unless properly designed. These impacts include reduced fish populations, non swimmable and boatable waters, and even destruction of entire ecosystems. Proper utilization of

PAGE 19

8 constructed wetlands and engineered gardens with native toleran t vegetation can help reduce these sediments prior to finding their way into streams and rivers (UDFCD 2013) The last pollutant category is metals. Metals, which are often associated with sediments, are found to cause direct problems in ecosystems as livi ng organisms are unable to na turally uptake these heavy elements. Copper and Zinc are two heavy metals associated with human and industrial deposition and are commonly analyzed in relation to the metal pollutant category (Seattle Public Utilities 2009 ; WWE et al. 2013 ) Although metals have become a primary concern, similar implementation of water quality detention basins that reduce sediment loads also help reduce heavy metal loading (Walker and Hurl 2002; UDFCD 2013) Add itionally, chemical, biological an d other processes carried out by wetlands are able to reduce heavy metals in the form of copper, zinc and lead (Walker and Hurl 2002). 2.5 Objectives of Case Study for Denver, Colorado Identification of watersheds contributing high pollutant loads can be a complex task and methods for determining land use sou rce contribution are uncertain Much research and data has been collected in relation to stormwater runoff and the ability to plan for this minor yet impactful detail in design and planning ca n be crucial for prosperous, healthy engineered ecosystems. The motivation for this thesis is that most of the impaired rivers and lakes in lakes in Colorado, which are primarily located along the eastern slope of the Rocky Mountains ( i.e. the Front Range ), have unknown sources of contamination. One of the major categories within these unknown sources is urban related stormwater runoff. By 2011, over 60% of land use within a 25 km radius around Denver, Colorado had been developed with imperviousness rangin g from 20 to 100

PAGE 20

9 percent (Homer et al. 2015) These developed areas, which consist of both single and multi family homes, and highly developed commercial and industrial areas, create excess urban deposition that is transported during storm events through e ngineered stormwater systems or directly into receiving streams. Understanding there are significant differences between stormwater pollutants for various land use types, this study was developed to evaluate these uncertainties of unknown non point stormwa ter sources using water quality assessment and geospatial methods. The following study uses metropolitan Denver, Colorado to evaluate urbanization impacts with respect to collected non point stormwater runoff data for residential and commercial land uses. This thesis study uses a two phase analysis: (1) collection of non point sto rmwater quality data for metropolitan Denver, Colorado and (2) geospatial analysis linking land cover datasets to urbanization impacts for different land uses. This analysis led t o the development of a non point stormwater watershed assessment model that ca n be used to evaluate and identify watersheds acting as major sources for sediment s nutrient s a nd metals, along other additional pollutant loading Lakewood Gulch and its surrounding watershed, which falls within the study boundary, was used to demonstrate the non point watershed assessment matrix that was developed in response to a water quality and geospatial analysis methods to determine regional relationships.

PAGE 21

10 3. Methods 3.1 Overview According to the Colorado Water Quality Control Commission (CWQCC), the availability of event mean concentration (EMC) based urban runoff data is sufficient to characterize the nutrient loads within the state of Colorado (Wright Water Engineers, Inc. et al 2013) Evaluation of non point source loading from this urban runoff data that has been collected enables standards to be developed for the state However, understanding the variability of non point urban water quality runoff with respect to different la nd uses, geographical locations, imperviousness, and type of sampling procedure can be a challenging feat when selecting appropriate site locations for non point stormwater quality features (Maestre and Pitt 2006). Variation of different non point stormwater land use samples, with respect to regionalization was a key aspect in this study when using such large datasets. Analysis of relevant stormwater site locations was required prior to analyzing sample data. In ord er to minimize these variations, strict data acceptance criteria were created to determine relevant water quality sampling locations. This four part sampling criteria includes: (1) site location and descriptions, (2) land use type distinguished by databas e records, (3) greater than two years of sampling data, and (4) available constituent data for the three pollutant categories, sediments, nutrients, and metals. To correspond with land cover datasets and the scope of this study, all commercial, industrial and institutional land classes designated by the stormwater databases, were combined into the one category, commercial due that common high impervious areas (greater than 80%). Additionally, mixed areas were not included in the

PAGE 22

11 analysis as they do not ali gn with the geospatial analysis process that separates the land cover datasets into one of the three classifications, Developed Low Intensity (NLCD Class 22), Developed Medium Intensity (NLCD Class 23), and Developed High Intensity (NLCD Class 24). I n particular, only non mixed NSQD classifications were included, meaning sample sites were required to be 100% residential or 100% commercial, industrial or institutional. The following analysis is based on non point stormwater runoff data accessed throu gh the National Stormwater Quality Database (NSQD) and land cover classification rasters accessed through the National Land Cover Dataset (NLCD). Data were augmented using additional sources including the I nternational BMP database and raw data collected from personal communication with Holly Piza at UDFCD (2015). This study uses a two phase process to develop a geospatial analysis method for evaluating watersheds as non point stormwater sources for stormwater pollutants into streams and lakes. The first phase is a non point water quality assessment to query locate and evaluate stormwater samples for residential and commercial land uses within met ropolitan Denver, Colorado. The second phases uses a geospatial analysis for land cover datasets that were col lected and analyzed through geographic information system process to determine land cover classifications for metropolitan Denver, Colorado. This regionalized process was then used to develop a simple and discrete matrix to evaluate watershed loading in th e form of n number of constituents. Methods used for this analysis will be further explained within this chapter.

PAGE 23

12 3.2 Site Selection The following thesis uses metropolitan Denver, Colorado as a case study area to analyze urbanization trends with respect to wat er quality from non point stormwater runoff data. Denver, which is located along the Front Range of the Rocky Mountains, is the first large, urban region through which its streams and rivers flow as they descend from the adjacent mountainous terrain. Due to population growth, Denver has grown significantly since 2000 as large residential and commercial developments have been built across the region. Although many of these new developments are equipped with stormwater quality detention basins or some form o f stormwater pollution prevention, the impact from non point stormwater runoff has not been eliminated, which is why the South Platte River and its major tributaries are listed by the US EPA as impaired waters with TMDL controls needed (U.S. EPA 2010; Stev ens and Slaughter 2012) 3.3 Non Point Stormwater Quality Analysis 3.3.1 Data Selection In order to distinguish non point stormwater sampling locations from other sampling locations, the four site selection criteria, listed in Chapter 3.1 enabled a query based selection process. Using the NSQD v4.02 as the main stormwater database, data was augmented in the following process. The first data selection uses the first criteria, location, that was specifi c to only the state of Colorado. Of the 690 sampling site location s in the database, 49 of these sampling sites were in Colorado. The second data selection was again location specific. Only counties that were located in or around metropolitan Denver, Colorado were included. These counties include Arapahoe, Adams, Jefferson, and Denver, respectively.

PAGE 24

13 Of these 49 sampling locations within Colorado, 38 of thes e sampling sites were located inside one of the four counties listed. The third data selection uses the second criteria, land use type, to select only residential and commercial areas for the analysis. As mentioned in this chapter these commercial areas include all commercial, industrial, and institution database classification for the sampling sites. All mixed sampling site locations were omitted during this selection Of the 38 sampling locations within the selected co unties around Denver, 26 of these sampling sites were selected based on the two land use types, residential and commercial. The four th selection uses the third criteria, year of record, to select only location specific, residential and commerc ial sampling locations that had more than two years of sample records. This criteria, which is mainly used to prevent any unique loading for a given year such a construction projects or after a flood, enables yearly assessment to be recorded, rather than o nly during one period in time. Of the 26 sampling locations determined through the first three criteria 12 final site locations were selected for the constituent analysis after year of record had been determined. It is important to note, tha t these 12 sites had some form of overlap. Two site locations, UDFCD Modular Porous Pavement and Shop Creek Wetland Pond, each had two periods of record that were listed as separate datasets within the database. The reason for this is unknown, but could c orrespond with a change in monitoring practice or other uncertainty not listed. For this study, these two location datasets were combined for the each site location as each represents either residential or commercial land use. The selection p rocess results in a total of 10 sampling locations that were selected from the original 690 sites in the database. These ten sampling location s identified using

PAGE 25

14 the first three established criteria, were made up of five residential and five commercial locations These locations can be seen in Figure 1 Additionally, Table 1 provides locations and descriptions including Site ID, coordinate s and sampling events over the year of record for each of the ten sites augmented. Table 1 : Final Site Location s and Description s from Selection Criteria Site ID Sampling Site Name Year of Record Coordinate Location No. of Event Residential COLAIRIS 21 st and Iris Rain Garden 2011 2014 39.7488 N 105.1066 W 54 CODEGRHE Grant Heron 2000 2009 39.6197 N 105.0582 W 29 CODEGRRE Grant Reflect 2000 2009 39.6184 N 105.0594 W 25 COAUSHCR Shop Creek Wetland Pond 1990 1997 39.6291 N 104.7415 W 55 CODEORPO UDFCD Orchard Pond 2000 2011 39.6211 N 105.0598 W 106 Commercial COACWWL3 Arapahoe Country Water & Wastewater Authority (L3) 2008 2009 39.6005 N 104.8379 W 19 COACW6W7 Arapahoe Country Water & Wastewater Authority (W6W7) 2008 2009 39.5919 N 104.8206 W 17 CODEWAWA Denver Wastewater Building 2008 2014 39.7209 N 105.0106 W 67 COLASHOP Lakewood Shops 2005 2015 39.8748 N 105.1630 W 121 COLAMOPA UDFCD Modular Porous Pavement 1994 2006 39.8833 N 105.2000 W 48 Note: Shop Creek Wetland Pond and UDFCD Modular Porous Pavement had two sets of sampling years listed in the NSQD. This may reflect a change in monitoring practices or another uncertainty not listed in the NSQD v4.02. Source: Pitt (2015), WWE (2014), WWE (2012), and personal communication (UDFCD 2015)

PAGE 26

15 Figure 1 : Residential and Commercial Sampling Locations for Denver, Colorado

PAGE 27

16 3.3.2 Selection of Constituent Data As mentioned in Chapter 2 excessive sediments, nutrients, and metals in response to urbanization cause impairments in streams and lakes. Non point runof f from these urban areas create chemical unbalances within ecosystems ranging from decreased beneficial uses to loss of life itself. In order to restore these impairments listed by the U.S. EPA, water quality assessments on streams and lakes need to be car efully analyzed in order to plan future remediation strategies. Having augmented the dataset into the selected site locations, the selection of constituents based on available data could begin. Analysis of the three pollutant categories, sediments, nutrie nts, and metals, were determined in response to the two land use types that were under analysis. A total of seven constituents were analyzed within the three categories. These seven stormwater pollutants can be seen in Table 2 below along with their potent ial urban sources Table 2 : Common Urban Stormwater Pollutants and Potential Sources Stormwater Pollutant Potential Source Sediments Total Suspended Solids [TSS] Construction sites, erosion, poorly vegetated lands, large commercial vehicles Common Nutrients Total Kjeldahl Nitrogen [TKN] Lawn fertilizers, domestic animal waste, vegetative matter, detergents Nitrite + Nitrate [NO 2 +NO 3 ] Total Phosphorus [TP] Dissolved Phosphorus [DP] Metals Total Copper [Cu] Atmospheric deposition from fuel combustion and industrial processes, vehicles, soil erosion Total Zinc [Zn] Sources: U.S. EPA (2007 b ), Seattle Public Utilities (2009), and UDFCD (2013)

PAGE 28

17 After established criteria were used to select the sampling site locations, the final criteri on was used on the selected seven constituent data recorded from the database. This final criterion which analyzed available constituent data of t he final ten sites sourced an in depth and detailed quality assurance check on each of the datasets to address any unique, repeated or missing values. Identification of unique and repe ated sample events, which can skew results, included addressing variations in monitoring practices, locating human data entry errors, and backfilling missing and additional data not included in the dataset. Additionally, the NSQD v4.02 had not been updated with sampling events from 2014, which prompted personal communication with Holly Piza at Urban Drainage and Flood Control District and Jane Clary at Wright Water Engineers to collect raw data for the 2014 sampling season. This raw data collected through personal communication enabled additional sample data for the ten sites provided in the NSQD Additional raw data wa s also provided that enabled manual data entry for missing parts of the sample events recorded in the NSQD from past years With the final criteria address ed final datasets on all sampling events for eac h of the ten sites was complete Statistical analys is results on the final dataset will be further explained in Chapter 4. 3.4 G eospatial Land Cover Analysis 3.4.1 Data Collection Land cover for the continental United States was retrieved from the Multi Resolution Land Characteristics Consortium (MRLC) to determine land use types in relation to the water quality sampling locations. Supported by the U.S. Department of the Interior (D OI) and the U.S. Geological Survey (USGS), the MRLC specifies land covers

PAGE 29

18 for uses ranging from agricultural to forest to developed areas. The 2001, 2006, and 2011 NLCD rasters, which provide 30 meter by 30 meter cells for each of the land covers in the fo rm of a TIFF file, were retrieved and analyzed using geographic information system processes listed in this section. Rasters are images in the form of rectangular gridded pixels or group of cells that represent some form matrix data structure such as elevation or land cover. There are two important items to note about the datasets that were retrieved. First, the 1 992 NLCD was un able to be used in the study because the classifications for land cover were inconsistent with the 2001, 2006, and 2011 datase ts (Vogelmann et al. 2001) Second, to limit the scope of the study, only developed land cover areas were analyzed as they are the key locations for urbanization. These developed areas, which include Low Intensity, Medium Intensity, and High Intensity lan d classifications, were used in conjunction with the collected non point land use runoff data from the water quality analysis. The three developed land use classes can be seen in Table 3 along with their descriptions (Homer et al. 2015) A reas identified a s Developed Open Space, which include golf courses, parks, and developed recreational and aesthetic areas and account for less than 20% imperviousness, were not included in the analysis as non point land use stormwater quality runoff can greatly vary fro m these locations. In relation to the water quality analysis, Developed Low and Medium Intensity classifications correspond to residential locations, while Developed High Intensity classifications were set to correspond to commercial locations.

PAGE 30

19 Table 3 : NLCD Classifications for Developed Land Use NLCD Class ID NLCD Classification Description for Developed Areas 22 Developed, Low Intensity Areas with a mixture of constructed materials and vegetation. Impervious surfaces account for 20% to 49% percent of total cover. These areas most commonly include single family housing units. 23 Developed, Medium Intensity Areas with a mixture of constructed materials and vegetation. Impervious surfaces account for 50% to 79% of the total cover. These areas most commonly include single family housing units. 24 Developed High Intensity Highly developed areas where people reside or work in high numbers. Examples include apartment complexes, row houses and commercial/industrial. Impervious surfaces account for 80% to 100% of the total cover. Source: Homer et al. (2015) 3.4.2 Geospatial Processing Methods Having identified the sample locations, a n arbitrary circular boundary was created around Denver for the purpose of limiting the geographic scope under analysis Centered at the county courthouse in downtown Denver, this 25 km radius encompasses all stormwater sampling locations. Once the study boundary wa s determined, relevant shapefiles and file databases were added to the maps for the study. Shapefiles, which include geographic features such as rivers, counties, and site locations, store vector data relating to locations, shapes, and attributes for each feature dataset that was presented In order to maintain proper file management, a master geodatabase was created to store, query and manage all GIS data for processed rasters and features in one central file location. All features and rasters were convert ed to the same coordinate system, North American Datum 1983 (NAD 1983) prior to any GIS processes or analyses.

PAGE 31

20 Each NLCD raster was processed using the Clip Raster tool in ArcMap 10.3 for the radial 25 km study area boundary that was created. Using this pr ocess for each of the 2001 NLCD, 2006 NLCD, and 2011 NLCD, three maps, shown in Figure 17 19, were developed to sh ow urbanization trends for the metropolitan Denver, Colorado. Result s and discussion of this geospatial analysis for Denver, Colorado will be further discussed in Chapter 4. Original attempts to convert this clipped raster to unique polygons were unsuccessful as the GIS process returned over 180,000 + polygons for developed land use classifications within the 25 km radius boundary for De nver, Colorado In efforts to address this problem and not enlarge cell raster sizes, a watershed was selected for which this analysis would be applied to. A similar Clip Raster process was used for the Lakewood Gulch watershed to reduce the raster to a manageable file size. After removing unwanted classifications from the 2011 NLCD raster dataset, the Raster to Polygon tool was used in ArcMap 10.3 to convert all cells classified as developed into 4,000 + polygons which was much more manageable An attrib ute table was then created for these polygons that could be exported into an adaptable Excel file format to identify developed land use areas and percentages for the watershed The last process in the geospatial analysis used the Union tool to join overla pping features into a new output feature class This tool used two inputs to address overlapping layers. These two inputs included the polygon features created in the previous step for the Lakewood Gulch watershed and a polygon shapefile of eight delineate d sub basin s within the watershed. Using the Union tool in ArcMap 10.3, final land use areas and percentages could be calculated for the different sub basins. These sub basins land use

PAGE 32

21 areas and percentages will be used for the non point assessment matrix found in the next part of th is chapter. 3.5 G eospatial Non Point Stormwater Assessment Matrix 3.5.1 Development of Geospatial Assessment Matrix Using the following method as described in the section below determination of watersheds and associated sub b potential on non point urban impacts can be evaluated This method uses regional land cover data in relation with a geospatial analysis. W atershed size, collected land use data relevant to a given watershed and event rainfall are required for the matrix to be evaluated. Similar models have been developed by the National R esources Conservation Service (NRCS) including the Agricultural Non Point Source Pollution Model (AGNPS), which focuses on non point runoff from agricultural areas (Bingner and Theurer 2009). This AGNPS model is not applicable to large urban areas as it was developed to address different agricultural uses and associated runoff properties The model developed in this thesis through the two phase analysis descri bed previously in the chapter introduce s a simple area weighting and a discrete matrix evaluation method for analyzing any number of constituents for different land uses with a given study area The developed geospatial assessment matrix uses land use pe rcentages as weighting coefficients to identify urban watersheds that may be significant sources of non poi nt stormwater runoff data. T his model does not account for any engineere d stormwater quality treatment such as water quality detention basins, rain gardens or other water quality treatment features. Additionally, this model can also be applied to smaller sub basins once an urban watershed is located, to evaluate sub asse ssment analyses.

PAGE 33

22 The following non point geospatial assessment matrix requires four inputs. T wo variables, volume of runoff and water shed area, and two sets of data that are able to be converted to the [m x n] and [n x 1] matrix form. The first dataset converted to the [m x n ] matrix form includes water quality data for various constituents in the form of mass per volume. The m variable represents rows which correspond to similar constituents for different land use classifications. The n variable represents columns that corresponds to different constituents with similar land use type The second dataset co nverted to the [n x 1] matrix form includes g eospatial land use data as a percentage over the total study area. The n variable represents number of land use types within the study area. Using this developed method which is shown below in both single and multiple constituent form t otal non point loading (TNP L) can be evaluated for a given urban watershed and associated sub basins Simple Area Weighting for Single Constituent Discrete Area Weighting for Multiple Constituents Where, D represents depth of runoff ( M ) A represent area of the watershed or sub basin (M 2 ) C represent different non point stormwater runoff statistical data for dif ferent constituents in analysis. ( M/L 3 ) o Letters represent different constituents and numbers represent associated land use type with that constituent LU represents different land use percentages for study area [%] TNP L represents Total Non Point Loading for each of the constituents C a C b to C n over a watershed or sub basin area [M]

PAGE 34

23 3.5.2 Applications of Matrix using Lakewood Gulch Example Application of the matrix that was developed to analyze non point watershed loading could be applied after statistical and geospati al analysis methods had been de termined As mentioned previously the South Platte River and its tributaries are listed as impaired by the U.S. EPA in the most recent 2010 assessment report (U.S. EPA 2015). Using the Lakewood Gulch tributary of t he South Platte River, which is listed by the U.S. EPA as impaired in Aquatic Life Warm Water Class 2 and Recreational Primary Contact designated categorical uses, application of the non point stormwater matrix could be applied with d ata that was collected within the regional non point stormwater analysis summary (U.S. EPA 2015). This straightforward model revolves around a geospatial analysis principle using land use percentages as weighting coefficients to evaluate urban watersheds based on non point stormwater runoff data. This method can be used to identify watersheds that are major contributors of pollutants from unregulated non point locations for urban areas and multiple constituents and multiple land use within the same waters hed and/ or smaller sub basins comparisons within a given watershed. Using selected basins that drain to Lakewood Gulch tributary of the South Platte River land use polygons were created using similar geospatial process as for Denver Colorado Thee devel oped land use polygon sets that relate to land use types, which can be seen in Figure 12 and Table 4, were analyzed using the collected land use data for the seven pollutants in the study. Using mean concentrations for the seven pollutants, evaluation of t he watershed loading could be completed with the Non Point Stormwater Geospatial Watershed Loading Matrix.

PAGE 35

24 This matrix, whic h assumes no BMP implementation, uses 2011 NLCD for three land use polygons within the Lakewood Gulch watershed and non point stormwater mean values from the water quality analysis. The example provided o nly evaluates a 2 year rainfall depth event over the watershed area, which corresponds with regulated design practices and treatment of minor storm event r unoff for the area (UDFCD 2013). Calculations for the Lakewood Gulch assessment matrix example can be found in the Appendix A.2.

PAGE 36

25 4. Results and Discussion 4.1 Non Point Stormwater Quality Results Using the data collected for the stormwater sampling sites, final statistical and correlative analyses can be determined for the two different land use types. The following section compares median concentrations for commercial and residential areas, and summarizes a correlation an alysis of the constituents listed on Table 2. Figures and tables provided throughout each section and are discussed in the text. Raw data that was collected and used for non point stormwater quality statistical analysis and correlations is provided in the Appendix. 4.1.1 Statistical Analysis The first analysis on the final data set of stormwater quality runoff constituents uses box and whisker plo ts to show event sample mean, range, and normality, along with another statistical evaluation for the seven pollutants within the two land use categories. A t Test was run on each of the land use data sets to determine corresponding p values for both one tailed and two tailed tests. P values were used to determine significance of the relationships between mean concentrations for residential and commercial land use s. A null hypothesis was created that mean values are equal for each land use type and p values were used to either accept or reject th is hypothesis Plots can be seen below along with corresponding tables that are provided for each of the seven constituents in the analysis and are discussed within this section. Final comparisons to the U.S. EPA et al. 1983 for median and coefficient of variation values Comparison of event median concentrations with the 1983 NURP study are shown in Tab le 22 and Table 23

PAGE 37

26 A. Sediments: TSS As mentioned in the Literature Review, Total Suspended Solids (TSS) are often used to evaluate stormwater runoff in the form of the sediment pollutant category. There were a total of 507 recorded TSS samples for the two land use sampling locations, 246 samples for residential and 261 samples for commercial. Multiple statistical categorie s were analyzed to determine a final TSS classification summary for residential and commercial land uses within Denver, Colorado. Results from the statistical analysis can be seen in Table 4 and Table 5, along with a box and whisker plot for the TSS consti tuent in Figure 2. Additionally, Table 6 provides individual site statistics calculated for mean, standard deviation, max, and median TSS values. First, event mean concentrations (EMCs) for residential and commercial sampling locations were compared. TSS i n residential areas had an EMC of 204 mg/L with a standard deviation of 239 mg/L for the samples that were analyzed. TSS in commercial areas had an EMC of 193 mg/L with a standard deviation of 333 mg/L for samples that were analyzed EMC values for residential areas were higher than commercial areas by 11 mg/L To better understand significance of these e vent mean concentrations, t test s were run on each land use dataset to compare event mean values assuming unequal variances. Using a null hypothesis of an event mean concentration difference of 0, 472 degrees of freedom were calculated with a 0.42 t statistic for TSS residential and commercial land use mean values. For the two tailed t test, p values for the two mean were calculated a t 0. 67 This TSS p value is considerably greater than the level of significance for the TSS analysis of 0.05, thus the null hypothesis was accepted for residential and commercial TSS event mean concentrations.

PAGE 38

27 Second, event maximum concentrations (EMxCs) for residential and commercial sampling locations were compared. TSS in residential areas had an EMxC of 1310 mg/L. TSS in commercial areas had an EMxC of 2260 mg/L of samples that were analyzed. Commercial samp ling location EMxC values were considerably greater than residential sampling sites. Seven commercial samples recorded values greater than the residential maximum of 1310 mg/L. Further analysis using the TSS plot suggest t here were many lower event v alues that were recorded to offset large events in commercial areas. Last event median concentrations for residential and commercial sampling locations were compared. TSS in residential areas had an event median concentration value of 121 mg/L. TSS in commercial areas had an event median concentration value of 66 mg/L. Event m edian concentration differed by 55 mg/L as residential sites had a higher event median concentration when compared to commercial locations. Overall, the final determinat ion of TSS classification summary is as follows. Total Suspended Solids is not significantly different between residential and commercial areas and further analysis is needed. Although commercial loading recorded several maximum values greater than that recorded at a residential location, event mean and event median concentrations were higher for residential areas. This suggests that variability within commercial sites is larg er, and uncertainty for the commercial loading needs to be addressed for individual site characteristics and events. In relation to the null hypothesis that event mean concentrations are similar for residential and commercial areas the one tailed p value is significant as it does in fact help show that the observed mean values do not differ between the two land uses.

PAGE 39

28 Table 4 : Summary of Non Point Runoff Statisti cs (TSS) Summary of Non Point Runoff Data TSS (mg/L) Residential Commercial Average 204 193 Standard Deviation 239 333 Sample Size 246 261 Median 120.5 66 Max 1310 2260 Standard Error 25.1 33.9 Table 5 : Summary of t Test (TSS) TSS (units of mg/L) Residential Commercial Mean 204 193 Variance 57100.4 110773.6 Observations 246 261 Hypothesized Mean Difference 0 Degrees of Freedom 472 t statistic 0.4227 P(T<=t) one tail 0.3364 t Critical one tail 1.6481 P(T<=t) two tail 0.6727 t Critical two tail 1.9650 Note: t Test for Two Sample Assuming Unequal Variances

PAGE 40

29 Figure 2 : Box and W hisker Plots for Residential and Commercial Locations (TSS) 0 500 1000 1500 2000 2500 Residential Commercial TSS (mg/L) Total Suspended Solids [TSS] All Sites

PAGE 41

30 Table 6 : Individual Site Statistics for TSS Analysis Site ID Sample Size Average Standard Deviation Max Median Residential Sampling Locations (measured in mg/L) COLAIRIS 54 296.6 300.6 1310 168 CODEGRHE 29 202.4 192.1 814 157 CODEGRRE 23 272.9 296.9 1210 169.5 COAUSHCR 38 198.9 209.6 999 134 CODEORPO 102 142.4 189.9 1280 71.5 All Residential Sites 246 204 239 1310 12 1 Commercial Sampling Locations (measured in mg/L) COACWWL3 19 74.0 74.7 302 43.7 COACW6W7 17 39.0 42.2 186 20.8 CODEWAWA 66 383.7 472.4 2260 211.5 COLASHOP 113 182.7 297.2 1940 64 COLAMOPA 46 53.0 80.6 450 21.5 All Commercial Sites 261 193 33 3 2260 66 B. Nutrients: T KN and NO 2 +NO 3 Total Kjeldahl Nitrogen (TKN) and Nitrite + Nitrate (NO 2 +NO 3 ) datasets were used to evaluate stormwater runoff in nutrient pollutant category. There were a total of 416 recorded TKN samples for the two land use sampling locations, 196 samples for residential and 220 samples for commercial. Additionally, there were a total of 435 recorded NO 2 +NO 3 sampl es for the two land uses types, 226 samples for residential and 209 samples for commercial. Multiple statistical categories were analyzed to determine a final nitrogen classification summary, which is based on TKN and NO 2 +NO 3 sample data, for residential a nd commercial land uses within Denver, Colorado. Results from the statistical analysis can be seen in Tables 7 9 along with a box and whisker plot for the TKN and NO 2 +NO 3 constituents in Figure 3 and Figure 4. Additionally, individual site

PAGE 42

31 statistics were also calculated for the mean, standard deviation, max and median TKN and NO 2 +NO 3 values. These results can be seen in Table 10 and Table 11. First, event mean concentrations (EMCs) for residential and commercial sampling locations were compared. TKN in residential areas had an EMC of 3.41 mg/L with a standard deviation of 2.37 mg/L for the samples that were analyzed. TKN in commercial areas had an EMC of 2.53 mg/L with a standard deviation of 2.44 mg/L for samples that were analyzed. NO 2 +NO 3 in residenti al areas had an EMC of 1.07 mg/L with a standard deviation of 0.85 mg/L for the samples that were analyzed. NO 2 +NO 3 in commercial areas had an EMC of 0.70 mg/L with a standard deviation of 0.56 mg/L for samples that were analyzed. To better understand sign ificance of event mean concentrations for TKN and NO 2 +NO 3 t tests were run on each land use dataset to compare event mean values assuming unequal variances. Using a null hypothesis of an event mean concentration difference of 0 411 degrees of freedom were calculated with a 3.69 t statistic for TKN residential and commercial land use mean values. For the two tailed t test, TKN p values were in the order of 1x10 4 The p value is much less than the level of significance for the T KN analysis of 0.05, thus the null hypothesis was rejected for residential and commercial EMC values. Again, using a null hypothesis of an event mean concentration difference of 0, 393 degrees of freedom were calculated with a 5.32 t statistic for NO 2 +NO 3 residential and commercial areas. For the two tailed t test, NO 2 +NO 3 p values were in the order of 1x10 7 Since p values is much less than the level of significance for the NO 2 +NO 3 analysis of 0.05, thus we can reject the null hypothesis that residential and commercial NO 2 +NO 3 EMC values are similar.

PAGE 43

32 Second, event maximum concentrations (EMxCs) for residential and commercial sampling locations were compared for both nitrogen constituents. TKN in residential areas had an EMxC of 13.4 mg/L. TKN i n commercial areas had an EMxC of 23.4 mg/L. NO 2 +NO 3 in residential areas had an EMxC of 8.32 mg/L. NO 2 +NO 3 in commercial areas had an EMxC of 3.61 mg/L of samples that were analyzed. Further analysis of both the TKN and NO 2 +NO 3 suggest that the maximum value for commercial might have been some type of extreme event, however, no comments are provided for this event Last event median concentrations for residential and commercial sampling locations were compared for the two nitrogen constituents. TKN in residential areas had an event median concentration value of 2.85 mg/L. TKN in commercial areas had an event median concentration value of 2.00 mg/L. NO 2 +NO 3 in residential areas had an event median concen tration value of 0.91 mg/L. NO 2 +NO 3 in commercial areas had an event median concentration value of 0.58 mg/L. Event median concentration for residential sites had a higher event median concentration for both TKN and NO 2 +NO 3 when compared to c ommercial locations. TKN and NO 2 +NO 3 event median concentration values were 1.4 times and 1.6 times higher in residential areas than in commercial areas, respectively. Overall, the final determination of nitrogen classification summary is as follows. Higher loading in the form of TKN and NO 2 +NO 3 can be seen in residential sampling locations. All phases of the nitrogen statistical analysis provide supporting evidence TKN and NO 2 +NO 3 concentrations are higher in residential areas. Both plots show residential upper and l ower bounds are higher compared to commercial sampling locations. This further supports the notion that the maximum data point within the

PAGE 44

33 commercial location might have been an extreme or misreported event. However, this data is sti ll included in the analysis, thus mean values might be slightly higher than expected for commercial land use in this study. In relation to both constituent null hypothes e s checked, p values were less than level of significance of 0.05 and both null hypothese s were rejected. Table 7 : Summary of Non Point Runoff Statistics (TKN and NO 2 +NO 3 ) Summary of Non Point Runoff Data TKN (mg/L) N02+NO3 (mg/L as N) Residential Commercial Residential Commercial Average 3.41 2.53 1.07 0.70 Standard Deviation 2.37 2.44 0.85 0.56 Sample Size 196 220 226 209 Median 2.85 2.00 0.91 0.58 Max 13.4 23.9 8.32 3.61 Standard Error 0.3 0.3 0.1 0.1 Table 8 : Summary of One Tailed and Two Tailed t Test (TKN) TKN (units of mg/L) Residential Commercial Mean 3.41 2.53 Variance 5.63 5.95 Observations 196 220 Hypothesized Mean Difference 0 Degrees of Freedom 411 t statistic 3.69 P(T<=t) one tail 1x10 4 t Critical one tail 1.6486 P(T<=t) two tail 1x10 4 t Critical two tail 1.9658 Note: t Test Two Sample Assuming Unequal Variances

PAGE 45

34 Table 9 : Summary of One Tailed and Two Tailed t Test (NO 2 +NO 3 ) NO 2 +NO 3 (units of mg/L) Residential Commercial Mean 1.07 0.70 Variance 0.72 0.31 Observations 226 209 Hypothesized Mean Difference 0 Degrees of Freedom 393 t statistic 5.32 P(T< =t) one tail 1x10 7 t Critical one tail 1.6487 P(T<=t) two tail 1x10 7 t Critical two tail 1.9660 Note: t Test Two Sample Assuming Unequal Variances

PAGE 46

35 Figure 3 : Comparison of TKN for Residential and Commercial Sampling Locations 0 2 4 6 8 10 12 14 Residential Commercial TKN (mg/L) Total Kjedhal Nitrogen [TKN] All Sites

PAGE 47

36 Figure 4 : Comparison of NO 2 +NO 3 for Residential and Commercial Sampling Locations 0 2 4 6 8 10 12 14 Residential Commercial NO 2 +NO 3 (mg/L) Nitrite + Nitrate [ NO 2 +NO 3 ] All Sites

PAGE 48

37 Table 10 : Individual Site Statistics for TKN Analysis Site ID Sample Size Average Standard Deviation Max Median Residential Sampling Locations (measured in mg/L) COLAIRIS 54 3.24 2.19 13.40 2.95 CODEGRHE 12 3.79 1.70 7.50 3.73 CODEGRRE 13 3.41 1.94 7.20 3.45 COAUSHCR 37 3.83 3.10 12.20 2.90 CODEORPO 80 3.26 2.28 13.10 2.76 All Residential Sites 196 3.41 2.37 13.4 2.85 Commercial Sampling Locations (measured in mg/L) COACWWL3 0 N/A COACW6W7 0 N/A CODEWAWA 66 3.69 3.30 23.90 2.85 COLASHOP 108 1.93 1.30 8.90 1.60 COLAMOPA 46 2.30 2.52 13.00 1.45 All Commercial Sites 220 2.53 2.44 23.90 2.00 Table 11 : Individual Site Statistics for NO 2 +NO 3 Analysis Site ID Sample Size Average Standard Deviation Max Median Residential Sampling Locations (measured in mg/L) COLAIRIS 54 0.74 0.45 2.57 0.66 CODEGRHE 25 1.24 0.70 3.46 1.04 CODEGRRE 21 1.02 0.56 2.09 1.02 COAUSHCR 25 2.11 1.72 8.32 1.71 CODEORPO 101 0.95 0.52 2.70 0.85 All Residential Sites 226 1.07 0.85 8.32 0.91 Commercial Sampling Locations (measured in mg/L) COACWWL3 0 N/A COACW6W7 0 N/A CODEWAWA 60 0.64 0.65 3.61 0.55 COLASHOP 102 0.63 0.47 2.98 0.49 COLAMOPA 47 0.96 0.56 2.88 0.85 All Commercial Sites 209 0.70 0.56 3.61 0.58

PAGE 49

38 C. Nutrients: Total Phosphorus and Dissolved Phosphorus Similar to nitrogen and mention ed in the Literature Review, elemental phosphorus is required by natural ecosystem s to provide beneficial uses in a healthy state. Since phosphorus reacts with microbes found in soil and consumed through phytoremediative processes, Total Phosphorus (TP) and Dissolved Phosphorus (DP) datasets were used to evaluate stormwater runoff in the nutrient pollutant category. There were a total of 416 recorded TP samples for the two land use sampling locations, 235 samples for residential and 267 samples f or commercial. Additionally, there were a total of 364 recorded DP samples for the two land uses types, 192 samples for residential and 172 samples for commercial. Multiple statistical categories were analyzed to determine a final phosphorus category class ification summary, which is based on TP and DP sample data, for residential and commercial land uses within Denver, Colorado. Orthophosphate was not considered in this analysis. Results from the statistical analysis can be seen in Tables 12 14 along with a box and whisker plot for the TP and DP constituents in Figure 5 and Figure 6. Additionally, individual site statistics were calculated and can be seen in Table 15 and Table 16. First, event mean concentrations (EMCs) for residential and commercial samplin g locations were compared. TP in residential areas had an EMC of 0.51 mg/L with a standard deviation of 0.33 mg/L for the samples that were analyzed. TP in commercial areas had an EMC of 0.28 mg/L with a standard deviation of 0.38 mg/L for samples that wer e analyzed. DP in residential areas had an EMC of 0.25 mg/L with a standard deviation of 0.22 mg/L for the samples that were analyzed. DP in commercial areas had an EMC of 0.09 mg/L with a standard deviation of 0.16 mg/L for samples that were analyzed. EMC values for both phosphorus pollutant forms analyzed, TP and DP, were

PAGE 50

39 higher in residential areas when compared to commercial areas. To further understand significance of event mean concentrations for TP and DP t tests were run on each land use dataset to compare event mean values assuming unequal variances. Using a null hypothesis of an event mean concentration difference of 0, 500 de grees of freedom were calculated with a 7.38 t statistic for TP residential and commercial land use mean values For the two tailed t test, T P p values were in the order of 1x10 13 The p value is much less than the level of significance for the T P analysi s of 0.05, thus the null hypothesis was rejected for residential and commercial EMC values. Again, using a null hypothesis of an event mean c oncentration difference of 0, 344 degrees of freedom were calculated with a 8.04 t statistic for DP residential and commercial areas. For the two tailed t test, DP p values were in the order of 1x10 13 Since p values were less than the level of significance for the DP analysis of 0.05, we can reject the null hypothesis that residential and commercial D P EMC values are similar. Sec ond, event maximum concentrations (EMxCs) for residential and commercial sampling locations were compared for both pollutant constituents. TP in residential areas had an EMxC of 1.91 mg/L. TP in commercial areas had an EMxC of 4.44 mg/L. DP in residential areas had an EMxC of 1.62 mg/L. DP in commercial areas had an EMxC of 1.59 mg/L of samples that were analyzed. Further analysis of the TP plot suggest that the maximum value for commercial might have been some sort of extreme event, however it c orresponds with the same sample event data recorded in the TKN EMxC that had occurred. Both TP and DP values were plotted on the same y axis as some values are not shown in the box and whisker plots.

PAGE 51

40 Last event median concentrations for residential and commercial sampling locations were compared for the two pollutant constituents. TP in residential areas had an event median concentration value of 0.43 mg/L. TP in commercial areas had an event median conce ntration value of 0.17 mg/L. DP in residential areas had an event median concentration value of 0.18 mg/L. DP in commercial areas had an event median concentration value of 0.05 mg/L. Event median concentration for residential sites had a hig her event median concentration for both TP and DP when compared to commercial locations. TP and DP event median concentration values were 2.5 times and 3. 6 times higher in residential areas than in commercial areas, respectively. Overall, the final determination of phosphorus category classification summary is as follows. Higher loading in the form of TP and DP can be seen in residential sampling locations. All phases of the phosphorus statistical analysis provide supporting evide nce nutrients in the form of TP and DP concentrations are higher in residential areas. Both plots, TP and DP, show residential upper and lower bounds greater than commercial sampling locations. Similar to nitrogen, the same axis for both box and whisker plots were used for TP and DP. In relation to both constituent null hypothes es checked, p values were much less than the significance level set, thus no similar relationship between means can be seen between residential and commercial areas

PAGE 52

41 Table 12 : Summary of Non Point Stormwater Statistics (TP and DP) Summary of Non Po int Runoff Data Total Phosphorus (mg/L) Dissolved Phosphorus (mg/L) Residential Commercial Residential Commercial Average 0.51 0.28 0.25 0.09 Standard Deviation 0.33 0.38 0.22 0.16 Sample Size 235 267 192 172 Median 0.43 0.17 0.18 0.05 Max 1.91 4.44 1.62 1.59 Table 13 : Summary of One Tailed and Two Tailed t Test (TP) TP (units of mg/L) Residential Commercial Mean 0.51 0.28 Variance 0.11 0.14 Observations 235 267 Hypothesized Mean Difference 0 Degrees of Freedom 500 t Statistic 7.38 P(T<=t) one tail 1x10 13 t Critical one tail 1.6479 P(T<=t) two tail 1x10 12 t Critical two tail 1.9647 Note: t Test Two Sample Assuming Unequal Variances Table 14 : Summary of One Tailed and Two Tailed t Test (DP) DP (units of mg/L) Residential Commercial Mean 0.25 0.09 Variance 0.05 0.02 Observations 192 172 Hypothesized Mean Difference 0 Degrees of Freedom 344 t Statistic 8.0 4 P(T<=t) one tail 1x10 14 t Critical one tail 1.6493 P(T<=t) two tail 1x10 14 t Critical two tail 1.9669 Note: t Test Two Sample Assuming Unequal Variances

PAGE 53

42 Figure 5 : Comparison of TP for Residential and Commercial Sampling Locations 0 0.5 1 1.5 2 2.5 Residential Commercial Total Phosphorus (mg/L) Total Phosphorus [TP] All Sites

PAGE 54

43 Figure 6 : Comparison of DP for Residential and Commercial Sampling Locations 0 0.5 1 1.5 2 2.5 Residential Commercial Dissolved Phosphorus (mg/L) Dissolved Phosphorus [DP] All Sites

PAGE 55

44 Table 15 : Individual Site Statistics for TP Analysis Site ID Sample Size Average Standard Deviation Max Median Residential Sampling Locations (measured in mg/L) COLAIRIS 46 0.56 0.41 1.91 0.42 CODEGRHE 25 0.52 0.24 1.03 0.53 CODEGRRE 22 0.62 0.42 1.71 0.44 COAUSHCR 46 0.51 0.35 1.83 0.44 CODEORPO 96 0.46 0.28 1.73 0.38 All Residential Sites 235 0.51 0.33 1.91 0.43 Commercial Sampling Locations (measured in mg/L) COACWWL3 19 0.17 0.15 0.54 0.12 COACW6W7 17 0.26 0.10 0.40 0.30 CODEWAWA 66 0.57 0.64 4.44 0.39 COLASHOP 119 0.19 0.16 0.93 0.15 COLAMOPA 46 0.13 0.11 0.50 0.10 All Commercial Sites 267 0.28 0.38 4.44 0.17 Table 16 : Individual Site Statistics for DP Analysis Site ID Sample Size Average Standard Deviation Max Median Residential Sampling Locations COLAIRIS 48 0.18 0.19 1.23 0.13 CODEGRHE 23 0.27 0.13 0.71 0.24 CODEGRRE 19 0.32 0.27 1.33 0.25 COAUSHCR CODEORPO 102 0.26 0.23 1.62 0.17 All Residential Sites 192 0.25 0.22 1.62 0.18 Commercial Sampling Locations COACWWL3 17 0.04 0.03 0.12 0.04 COACW6W7 15 0.18 0.07 0.30 0.16 CODEWAWA 63 0.12 0.21 1.59 0.06 COLASHOP 77 0.06 0.11 0.97 0.04 COLAMOPA 0 All Commercial Sites 172 0.09 0.16 1.59 0.05

PAGE 56

45 D. Metals: Total Copper and Total Zinc As mentioned in the Literature Review, metals and other chemical pollutants are common remnants from human activity and industrial processes. Total Copper and Total Zinc sampling datasets for the ten sampling locations were used to evaluate stormwater runoff in the form of the metals pollutant category. Although additional metals could have been selected, these two metals had prominent data records and are two common metal polluta nts found in urban areas. There were a total of 271 recorded Total Copper samples for the two land use sampling locations, 186 samples for residential and 85 samples for commercial. Additionally, there were a total of 238 recorded Total Zinc samples for t he two land uses types, 155 samples for residential and 83 samples for commercial. Multiple statistical categories were analyzed to determine a final metal category classification summary, which is based on Total Copper and Total Zinc sample data, for resi dential and commercial land uses within Denver, Colorado. Results from the statistical analysis can be seen in Tables 17 19 along with a box and whisker plot for the Total Copper and Total Zinc constituents in Figure 7 and Figure 8. Additionally, individua l site statistics were calculated and can be seen in Table 20 and Table 21. First, event mean concentrations (EMCs) for residential and commercial sampling with a standard d with a standa samples that were analyzed. EMC values for both metal pollutant forms analyzed, Total

PAGE 57

46 Copper a nd Total Zinc, were higher in commercial areas when compared to residential areas. To further understand significance of event mean concentrations for Total Copper and Total Zinc t tests were run on each land use dataset to compare event mean values assum ing unequal variances. Using a null hypothesis of an event mean concentration difference of 0, 106 degrees of freedom were calculated with a 1.87 t statistic for T otal Copper residential and commercial event mean values. For the two tailed t test, T otal C opper p values were 0.06 The p value is slightly greater than the level of significance for the T otal Copper analysis of 0.05, thus the null hypothesis was acc epted for residential and commercial EMC values. Again, using a null hypothesis of an event mean concentration difference of 0, 96 degrees of freedom were calculated with an 1.58 t statistic for Total Zinc residential and commercial area event mean values For the two tailed t test, Total Zinc p values were 0.12 Since p values were greater than the significance level set of 0.05 we can accept the null hypothesis that residential and commercial Total Zinc EMC values are similar. Second, event maximum concentrations (EMxCs) for residential and commercial sampling locations were compared for both metal constituents. Total Copper in c correlation with mean values, further analysis using the data plot suggest both metal constituents, Total Copper and Total Zinc, are likely to have higher extreme events occur at commercial locations.

PAGE 58

47 Last event median concentrations for residential and commercial sampling locations were compared for the two metal constituents. Total Copper in residential areas opper in commercial areas event median concentration value of 72 mg/L. Event med ian concentration differed for each of the metal constituents. Total Copper for commercial sites had higher event median concentrations for residential sites when compared to commercial locations. On the other hand, Total Zinc event median concentrations h ad higher values recorded at commercial sampling sites. This provides additional insight as to why p values were both greater than the significance level set of 0.05 when analyzing the EMC values. Overall, the final determination of metals category classification summary is as follows. Both Total Copper and Total Zinc have similar loading in r esidential and commercial areas and further analysis is needed. All phases of the statistical analysis, event mean, median, and maximum values, in addition to the box and whisker plot, suggest concentrations at commercial areas will be higher when compared to residential areas. However in relation to bot h constituent null hypothes e s checked, two tailed t test p values were greater than the 0.05 significance level set Only minor relationship s each metal constituent EMC mean p values for the two land uses are shown as th e s e null hypothes e s w ere accepted Additionally, further analysis would look to collect more data from commercial locations as residential sampling areas had almost two time s as many sampling events.

PAGE 59

48 Table 17 : Summary of Non Point Stormwater Statistical Analysis (Cu and Zn) Summary of Non Point Runoff Total Copper ( g/L) Total Z inc ( g/L) Residential Commercial Residential Commercial Average 20.1 27.6 104.0 143.1 Standard Deviation 18.2 34.8 87.2 216.9 Sample Size 186 85 155 83 Median 14.3 15.95 80 72.1 Max 130 224 590 1440 Standard Error 2.2 6.2 11.5 39.2 Table 18 : Summary of One Tailed and Two Tailed t Test (Total Copper) Total Copper (units of g/L) Residential Commercial Mean 20.1 27.6 Variance 330.9 1209.1 Observations 186 85 Hypothesized Mean Difference 0 Degrees of Freedom 106 t s tat istic 1.87 P(T<=t) one tail 0.0319 t Critical one tail 1.6594 P(T<=t) two tail 0.0637 t Critical two tail 1.9826 Note: t Test: Two Sample Assuming Unequal Variances

PAGE 60

49 Table 19 : Summary of One Tailed and Two Tailed t Test (Total Zinc) Total Zinc (units of g/L) Residential Commercial Mean 104.0 143.1 Variance 7602.8 47025.5 Observations 155 83 Hypothesized Mean Difference 0 Degrees of Freedom 96 t statistic 1.5 8 P(T<=t) one tail 0.0 6 t Critical one tail 1.660 P(T<=t) two tail 0.1 2 t Critical two tail 1.985 Note: t Test: Two Sample Assuming Unequal Variances

PAGE 61

50 Figure 7 : Comparison of Total Copper for Residential and Commercial Sampling 0 50 100 150 200 250 Residential Commercial Total Copper (g/L) Total Copper [Cu] All Sites

PAGE 62

51 Figure 8 : Comparison of Total Zinc for Residential and Commercial Sampling Locations 0 200 400 600 800 1000 1200 1400 Residential Commercial Total Zinc (g/L) Total Zinc [Zn] All Sites

PAGE 63

52 Table 20 : Individual Site Statistics for Total Copper Analysis Site ID Sample Size Average Standard Deviation Max Median Residential Sampling Locations (measured in g/L) COLAIRIS 54 18.7 11.8 53.6 14.7 CODEGRHE 0 CODEGRRE 0 COAUSHCR 54 30.1 26.3 130.0 24.5 CODEORPO 78 14.1 10.8 73.6 10.9 All Residential Sites 186 20.1 18.2 130.0 14.3 Commercial Sampling Locations (measured in g/L) COACWWL3 0 COACW6W7 0 CODEWAWA 42 45.4 42.1 224.0 33.1 COLASHOP 43 10.2 7.9 39.5 8.4 COLAMOPA 0 All Commercial Sites 85 27.6 34.8 224.0 16.3 Table 21 : Individual Site Statistics for Total Zinc Analysis Site ID Sample Size Average Standard Deviation Max Median Residential Sampling Locations (measured in g/L) COLAIRIS 46 126.3 91.3 468.0 97.4 CODEGRHE 0 CODEGRRE 0 COAUSHCR 53 115.2 107.6 590.0 100.0 CODEORPO 56 75.1 46.1 240.0 65.6 All Residential Sites 155 104.0 87.2 590.0 80.0 Commercial Sampling Locations (measured in g/L) COACWWL3 0 COACW6W7 0 CODEWAWA 40 247.1 273.4 1440.0 167.5 COLASHOP 43 46.4 50.1 215.0 34.4 COLAMOPA 0 All Commercial Sites 83 143.1 216.9 1440.0 72.1

PAGE 64

53 Table 22 : Comparison to 1983 NURP S tudy (Residential) Pollutant Units US EPA 1983 This Study 2015 Median COV Median COV TSS mg/ L 101 0.96 12 1 1.17 TKN mg/ L 1.9 0.73 2.85 0.70 NO2+NO3 mg/ L 0.736 0.83 0.91 0.79 Total Phosphorus mg/ L 0.38 0.69 0.43 0.65 Diss. Phosphorus mg/ L 0.14 0.46 0.1775 0.89 Total Copper g/ L 33 0.99 14 0.91 Total Zinc g/ L 135 0.84 80 0.84 Table 23 : Comparison to 1983 NURP Study (Commercial) Pollutant Units US EPA 1983 This Study 2015 Median COV Median COV TSS mg/ L 69 0.85 66 1.72 TKN mg/ L 1.1 8 0.43 2 0.96 NO2+NO3 mg/ L 0.572 0.48 0.58 0.79 Total Phosphorus mg/ L 0.201 0.67 0.17 1.37 Diss. Phosphorus mg/ L 0.8 0.71 0.05 1.75 Total Copper g/ L 29 0.81 16 1.72 Total Zinc g/ L 226 1.07 7 2 0.96

PAGE 65

54 4.1.2 Correlation Analysis The second statistical analysis provides correlations among different constituents for individual sampling events. These plots, which use a linear regression analysis for correlations, were developed for each of the three stormwater pollutant categories, sediments, nutrients, and metals. Regression was used to evaluate measured vs. predicted relationships between constituents and to aid in understanding trends and associated errors of these linear relationships. Only relevant l inear trend lines for measured and predicted values were provided and can be seen on Figures 9 11 within this section. Additionally, the linear estimate (L INEST ) function in Excel was used to calculate a statist ical array that was used to analyze relationships of the linear regression for each correlation. The developed plots, along with the statistical analyses evaluation were used to show positive, negative or n ull linear correlation s among different constituents within these pollutant categories. Raw data that was used to develop these correlations is provided in the Appendix. A. Constituent vs. Constituent Correlations between similar constituents were analyzed using linear regression. Three categories, which include nitrogen, phosphorus, and metals, were used to create co nstituent vs. constituent correlations as seen in Figures 9 11 The LINEST function was used to evaluate the relationships of these regressions. First, the nitrogen category, which uses TKN vs. NO 2 +NO 3 had R squared values of 0.004 for residential and 0.22 for commercial. Both land use categories have positive linear slopes, however, from the LINEST evaluation, the slope of the TKN vs. NO 2 +NO 3 regression was not significantly different from zero, indi cating a lack of correlation.

PAGE 66

55 When compared to the residential LINEST results, commercial results show less of a standard error o n the slope however, they are still not significant The final plot of the nitrogen regression analysis is shown in Figure 9. Two samples, one commercial and one residential, were not shown on the chart for plotting reasons. These values had a TKN greater than 14 mg/L or NO2+NO3 EMC values greater than 4.0 mg/L. Overall, n o significant correlations can be seen between the two nitrogen const ituents. Second, the phosphorus category, which uses Total vs. Dissolved Phosphorus, had R squared values of 0.30 for residential and 0.36 for commercial. Both land use categories have positive linear slopes and from the LINEST evaluation, these slopes are significant Both residential and commercial have standard errors within ten percent of the slope. With these LINEST results and analysis of the plot, we can see there are in fact p ositive correlations that are significant with respect to the standard error of the mean for both residential and commercial areas. Last, the metal category, which uses Total Zinc vs. Total Copper, had R squared values of 0.33 for residential and 0.93 for commercial. Both la nd uses, residential and commercial, had positive slopes. From the LINEST evaluation, both slopes are significant. The standard error of the residential slope is about 11%, while the standard error of the commercial slope is about 3%. Results for each land use type using the calculated LINEST array can be seen in Table s 2 4 26 With the LINEST results and analysis of Figure s 9 11 p ositive correlations between metals can be seen for both land uses, especially commercial areas.

PAGE 67

56 Table 24 : LINEST Results for TKN vs. NO2+NO3 TKN vs. NO2+NO3 LINEST Statistics Variable Residential Commercial Slope m 0.1703 2.1103 Standard Error (Slope) se slope 0.1978 0.2743 Intercept b 3.2914 1.0812 Standard Error (Intercept) se intercept 0.279 0.2452 Coefficient of Determination R 2 0.0041 0.2232 Standard Error (y estimate) se y 2.43 2.2 0 F Statistic F 0.742 59.177 Degrees of Freedom d f 182 206 Regression Sum of Squares ss reg 4.36 286.19 Residual Sum of Squares ss resid 1070.3 996.2 Table 25 : LINEST Results for TP vs. DP TP vs. DP LINEST Statistics Variable Residential Commercial Slope m 0.8149 1.7019 Standard Error (Slope) se slope 0.0935 0.1727 Intercept b 0.3074 0.1781 Standard Error (Intercept) se intercept 0.0305 0.0310 Coefficient of Determination R 2 0.3029 0.3636 Standard Error (y estimate) se y 0.27 0.35 F Statistic F 76.0 97.1 Degrees of Freedom d f 175 170 Regression Sum of Squares ss reg 5.65 12.07 Residual Sum of Squares ss resid 13.0 21.1

PAGE 68

57 Table 26 : LINEST Results for Total Zinc vs. Total Copper Total Zinc vs. Total Copper LINEST Statistics Variable Residential Commercial Slope m 2.6146 5.9862 Standard Error (Slope) se slope 0.3038 0.1906 Intercept b 47.2118 24.3649 Standard Error (Intercept) se intercept 8.8602 9.1446 Coefficient of Determination R 2 0.3291 0.9337 Standard Error (y estimate) se y 71.87 58.34 F Statistic F 74.1 986.6 Degrees of Freedom d f 151 70 Regression Sum of Squares ss reg 382579 3358023 Residual Sum of Squares ss resid 779948 238259

PAGE 69

58 Figure 9 : Constituent Correlations (TKN and NO 2 +NO 3 ) y = 2.1103x + 1.0812 R = 0.2232 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 TKN (mg/L) NO 2 +NO 3 (mg/L) TKN vs. NO 2 +NO 3 Correlations using Regression Analysis All Sites Residential Commercial Linear (Commercial)

PAGE 70

59 Figure 10 : Constituent Correlations ( TP vs. DP ) y = 0.8149x + 0.3074 R = 0.3029 y = 1.7019x + 0.1781 R = 0.3636 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Total Phosphorus (mg/L) Dissolved Phosphorus (mg/L) TP vs. DP Correlations using Regression Analysis All Sites Residential Commercial Linear (Residential) Linear (Commercial)

PAGE 71

60 Figure 11 : Constituent Correlations (Total Zinc vs. Total Copper) y = 2.6146x + 47.212 R = 0.3291 y = 5.9862x 24.365 R = 0.9337 0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 450.00 500.00 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 Total Zinc ( g/L) Total Copper ( g/L) Total Zinc vs. Total Copper Correlations using Regression Analysis All Sites Residential Commercial Linear (Residential) Linear (Commercial)

PAGE 72

61 B. TSS vs. Nutrients Evaluation of the different nutrients with respect to TSS was selected to determine if relations hips could be determined between the two pollutant catego ries, sediments and nutrients. In order to account for hig h magnitudes of TSS loading, a log scale was used for the y axis for all plots when comparing the TSS values to the different pollutant categories The first TSS vs. Nutrients correlation compares TSS and TKN for the two sampling land uses. R 2 values were 0.21 for residential and 0.28 for commercial. Both land use categories have positive linear slopes and from the LINEST evaluation, both slopes are significant. Both residential and commercial have standard errors on the slopes equaling 16% and 11% respectively. With these LINEST results and analysis of the plot, we can see there are in fact positive correlations that are significant with respect to the standard error of the mean for both residential and commercial areas. Results for each land use type using the calculated LINEST array can be seen in Table 27 along with the Regression plot in Figure 12. The second correlation TSS vs. Nutrients compare s TSS vs. NO 2 +NO 3 for the two sampling land uses. R 2 values were 0.02 for residential and 0.001 for commercial. Correlation for both residential and commercial sampling locations were similar and classified as no correlation, which corresponds with R 2 valu es near 0. Slopes for the two land use categories vary greatly. Residential shows a negative slope, while commercial shows a positive slope for the TSS vs. NO2+NO3 evaluation. From the LINEST evaluation, the standard error of the slope in relation to the residential sampling location proves there is no correlation between the two constituents. Both residential and commercial standard error on the slopes are greater than the slope itself, which shows

PAGE 73

62 these results are not significant. Results for e ach land use type using the calculated LINEST array can be seen in Table 28 along with the Regression plot in Figure 13 The last TSS vs. Nutrients correlation compares TSS vs. Total Phosphorus for the two sampling land uses. R 2 values were 0.28 for resid ential and 0.51 for commercial. Both land use categories have positive linear slopes and from the LINEST evaluation, the st andard errors on these slopes are significant. Both residential and commercial have standard errors on the slopes equaling 9% and 6%, respectively. With these LINEST results and analysis of the plot, we can see there are in fact positive correlations that are significant with respect to the standard error of the mean for both residential and commercial areas. To further this, stronger p ositive correlations are seen between the Total Zinc and Total Copper for commercial areas, when compared to residential areas. Results for each land use type using the calculated LINEST array can be seen in Table 29 along with the Regression plot in Figur e 14.

PAGE 74

63 Table 27 : LINEST Results for TSS vs. TKN TSS vs. TKN LINEST Statistics Variable Residential Commercial Slope m 39.6 79.8 Standard Error (Slope) se slope 6.38 8.54 Intercept b 63.2 14.0 Standard Error (Intercept) se intercept 26.66 30.45 Coefficient of Determination R 2 0.17 0.30 Standard Error (y estimate) se y 209.8 305.5 F Statistic F 38.6 87.3 Degrees of Freedom d f 187 207 Regression Sum of Squares ss reg 1.70E+06 8.15E+06 Residual Sum of Squares ss resid 8.23E+06 1.93E+07 Table 28 : LINEST Results for TSS vs. NO2+NO3 TSS vs. NO2+NO3 LINEST Statistics Variable Residential Commercial Slope m 16.03 25.30 Standard Error (Slope) se slope 19.36 46.28 Intercept b 227.5 208.1 Standard Error (Intercept) se intercept 26.3 42.2 Coefficient of Determination R 2 0.003 0.002 Standard Error (y estimate) se y 245.3 371.2 F Statistic F 0.69 0.30 Degrees of Freedom d f 222 197 Regression Sum of Squares ss reg 4.13E+04 4.12E+04 Residual Sum of Squares ss resid 1.34E+07 2.71E+07

PAGE 75

64 Table 29 : LINEST Results for TSS vs. TP TSS vs. TP LINEST Statistics Variable Residential Commercial Slope m 432.2 668.3 Standard Error (Slope) se slope 38.7 36.9 Intercept b 10.9 10.8 Standard Error (Intercept) se intercept 23.9 17.2 Coefficient of Determination R 2 0.4 0.6 Standard Error (y estimate) se y 195.9 221.9 F Statistic F 124.4 328.6 Degrees of Freedom d f 222 255 Regression Sum of Squares ss reg 4.78E+06 1 .62E+07 Residual Sum of Squares ss resid 8.52E+06 1 .26E+07

PAGE 76

65 Figure 12 : TSS vs. Nutrients Correlations (TKN) y = 39.622x + 63.151 R = 0.171 y = 79.8x + 14.041 R = 0.2968 1.00 10.00 100.00 1000.00 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 TSS (mg/L) TKN (mg/L) TSS vs. TKN Correlations using Regression Analysis All Sites Residential Commercial Linear (Residential) Linear (Commercial)

PAGE 77

66 Figure 13 : TSS vs. Nutrients Correlations ( NO 2 +NO 3 ) 1.00 10.00 100.00 1000.00 10000.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 TSS (mg/L) NO2+NO3 (mg/L ) TSS vs. NO2+NO3 Correlations using Regression Analysis All Sites Residential Commercial

PAGE 78

67 Figure 14 : TSS vs. Nutrients Correlation (Total Phosphorus) y = 432.2x 10.939 R = 0.3592 y = 668.28x + 10.841 R = 0.5631 1.00 10.00 100.00 1000.00 10000.00 0.00 1.00 2.00 3.00 4.00 5.00 TSS (mg/L) Total Phosphorus (mg/L)) TSS vs. TP Correlations using Regression Analysis All Sites Residential Commercial Linear (Residential) Linear (Commercial)

PAGE 79

68 C. TSS vs. Metals Both metals, copper and zinc, were evaluated with respect to TSS to determine relations between the different constituents. Similar to before, the TSS uses a logarithmic y axis to account for orders of magnitude The first comparison uses TSS vs. Total Co pper. R 2 values for this correlation were 0.29 for residential and 0.59 for commercial. Both land use categories have positive linear slopes and from the LINEST evaluation both slopes are significant. Both residential and commercial have standard errors on the slopes equaling 19% and 7%, respectively. With these LINEST results and analysis of the plot, we can see there are in fact positive correlations that are significant with respect to the standard error of the mean for both residential and commercial areas. To further this, stronger positive correlations are seen between the TSS and Total Copper for commercial areas, when compared to residential areas. Results for each land use type usi ng the calculated LINEST array can be seen in Table 30 along with the Regression plot in Figure 15. The second comparison uses TSS vs. Total Zinc. R 2 values for this correlation were 0.29 for residential and 0.77 for commercial. Both land use categories ha ve positive linear slopes and from the LINEST evaluation, the standard errors on these slopes are significant. Both residential and commercial have standard errors on the slopes equaling 11% and 6%, respectively. With these LINEST results and analysis of t he plot, we can see there are in fact positive correlations that are significant with respect to the standard error of the mean for both residential and commercial areas. To further this, stronger positive correlations are seen between the TSS and Total Zi nc for commercial areas, when compared to residential areas. Results for each land use type using the calculated LINEST array can be seen in Table 31 along with the Regression plot in Figure 16.

PAGE 80

69 All equations for each of the two metals are displayed on the plots for each of t he land use types Significant standard errors on the slopes and strong positive correlations were seen in commercial areas as R 2 values were 0.59 for TSS vs. Total Copper and 0.77 for TSS vs. Total Zinc, thus p ositive correlations can be seen between both metals. Table 30 : LINEST Results for TSS vs. Total Copper TSS vs. Total Copper LINEST Statistics Variable Residential Commercial Slope m 5.2 10.9 Standard Error (Slope) se slope 1.0 0.8 Intercept b 109.1 10.0 Standard Error (Intercept) se intercept 25.8 33.6 Coefficient of Determination R 2 0.13 0.71 Standard Error (y estimate) se y 229.9 241.7 F Statistic F 26.2 201.6 Degrees of Freedom d f 168 82 Regression Sum of Squares ss reg 1.38E+06 1.18E+07 Residual Sum of Squares ss resid 8.88E+06 4.79E+06 Table 31 : LINEST Results for TSS vs. Total Zinc TSS vs. Total Zinc LINEST Statistics Variable Residential Commercial Slope m 1.64 1.88 Standard Error (Slope) se slope 0.18 0.12 Intercept b 50.5 49.5 Standard Error (Intercept) se intercept 24.8 31.7 Coefficient of Determination R 2 0.37 0.79 Standard Error (y estimate) se y 191.2 216.2 F Statistic F 81.6 263.0 Degrees of Freedom d f 138 69 Regression Sum of Squares ss reg 2.98E+06 1.23E+07 Residual Sum of Squares ss resid 5.04E+06 3.23E+06

PAGE 81

70 Figure 15 : TSS vs. Metals Correlation (Total Copper) y = 5.1941x + 109.13 R = 0.1348 y = 10.94x + 10.037 R = 0.7108 1.0 10.0 100.0 1,000.0 10,000.0 0 50 100 150 200 TSS (mg/L) Total Copper (mg/L)) TSS vs. Total Copper Correlations using Regression Analysis All Sites Residential Commercial Linear (Residential) Linear (Commercial)

PAGE 82

71 Figure 16 : TSS vs. Metals Correlation (Total Zinc) y = 1.6365x + 50.52 R = 0.3716 y = 1.8781x + 49.508 R = 0.7921 1.00 10.00 100.00 1000.00 10000.00 0.00 200.00 400.00 600.00 800.00 1000.00 1200.00 1400.00 1600.00 TSS (mg/L) Total Zinc (mg/L)) TSS vs. Total Zinc Correlations using Regression Analysis All Sites Residential Commercial Linear (Residential) Linear (Commercial)

PAGE 83

72 4.2 Geospatial Land Cover Results The geospatial analysis using the process listed in the methods section proved to be a great tool for analyzing land use over the large urban area of Denver, Colorado. In 2011, developed land use accounted for 49.6% of land cover in the form of Low Intensity, Medium Intensity, and High Intensity NLCD classifications. Developed Low Intensity, which includes land use ranging from 20% to 49% imperviousness and is used as a residential in this study, makes up 522 km 2 of the total 1964 km 2 study area. Developed Medium Intensity, which includes land use ranging from 50% to 79% imperviousness and is used as a residential in this study, makes up 327 km 2 of the total 1964 km 2 study area. Developed High Int ensi ty, which rang es from 80% to 100% imperviousness and is used as a commercial land use makes up 125 km 2 of the total 1964 km 2 study area. When compared to the 2001 land cover dataset, land cover has slightly increased for all three developed land cover classes, Low Intensity, Medium Intensity, and High Intensity. Developed Low Intensity has increased by 0.4%, Developed Medium Intensity h as increased by 2.9%, and Developed High Intensity has increased by 1.3%; all with respect to the radial study boundary. Results for the 2001, 2006, and 2011 land cover can be seen in the Table 32 and Figures 17 19 This geospatial analysis, which was applied to the 1964 km 2 study area for Denver, Colorado, was used to show developed land use increases from urbanization for 2001, 2006, and 2011. Further application of this geospatial analysis method that was used will be provided in the next section in relation to the geospatial assessment matrix that was developed This application uses the developed non point watershed loading matrix to analyze the sub basins within the

PAGE 84

73 Lakewood Gulch watershed and identify residential and commercial areas contribu ting to high urban pollutant runoff. Table 32 : NLCD Analysis of Developed Areas within 25 km Radial Boundary NLCD Class ID Classification Description for Developed Areas (From NLCD Legend) 2001 2006 2011 Developed Land Use Chang e (2001 to 2011) Area (km 2 ) % of Total Area Area (km 2 ) % of Total Area Area (km 2 ) % of Total Area Area (km 2 ) % of Total Area 22 Developed, Low Intensity 515.3 26.2% 524.4 26.7% 522.4 26.6% 7.1 0.4% 23 Developed, Medium Intensity 269.5 13.7% 310.4 15.8% 327.1 16.7% 57.6 2.9% 24 Developed High Intensity 100.2 5.1% 115.4 5.9% 125.3 6.4% 25.1 1.3% Other 1,079 54.9% 1,014 51.6% 989 50.4% TOTALS 1,964 100 % 1,964 100% 1,964 100% Note: Based on site classifications from the NLCD Legend, NLCD Class 22 and 23 were used as residential areas and NLCD Class 24 was used as the commercial/industrial area in relation non point stormwater data.

PAGE 85

74 Figure 17 : Developed Land Cover Map for 25 km Radius around Denver, C O (2001)

PAGE 86

75 Figure 18 : Developed Land Cover Map for 25 km Radius around Denver CO (2006)

PAGE 87

76 Figure 19 : Developed Land Cover Map for 25 km Radius around Denver, CO (2011 )

PAGE 88

77 4.3 Geospatial Non Point Stormwater Assessment Matrix 4.3.1 Application of Matrix to Lakewood Gulch Study Area As mentioned previously, the Lakewood Gulch tributary of the South Platte River is classified by the U.S. EPA as impaired in two categorical uses including Aquatic Life Warm Water and Recreational Primary Contact Lakewood Gulch is located just west of central Denver. This watershed and associated sub basins w as selected in response to the discussion and determination that the study area is very ne ar fully developed and no new developments are planned Without implementation of any BMP, the Lakewood Gulch would experience excessive loading over the watershed from the three non point land uses as listed in the matrix With further analysis using this matrix, eight sub basins that were delineated by UDFCD using geoprocessed digital elevations maps were compare d to one another to identify locations within this watershed that could potential be key contributors to the urban stormwater impairments in the Lakewood Gulch drainageway. Thes e results can be seen in Tables 28 29 and in Figures 20 21 From this analysis and application of the matrix on a sub basin level, Sub Basin 3, Sub Basin 5, and Sub Basin 7 are the key locations within the watershed where action should be implemented. Sub Basin 5, which has the largest area for all three land uses (Residential A, Residential B, and Commercial), would provide the highest pollutant loadings in the form of non point runoff. If enough research is collected for the areas shown in grey in Figure 33 and Figure 34 a complete non point watershed load ing profile can be developed for the study watershed draining to Lakewood Gulch.

PAGE 89

78 With a complete analysis, Sub Basin 5 would serve a top priority for implementation of BMP sites. Use of LID such as rain gardens for small residential areas, implementation of large vegetated water quality detention basins in large residential neighbors, and selection of soil media below permeable pavements to remove high loads of metals; all result in efforts that can be made to reduce non point runoff pollutants that are d irectly carried into streams. Table 33 : Lakewood Gulch Geospatial Analysis Example using Polygons NLCD Class ID NLCD Classification Description for Developed Areas Land Use Type Variable Area (km 2 ) % of Total Area 22 Developed, Low Intensity RES. 1 Land Use 1 LU1 20.5 40.3% 23 Developed, Medium Intensity RES. 2 Land Use 2 LU2 10.3 20.2% 24 Developed High Intensity COM. Land Use 3 LU3 4.7 9.1% Other N/A N/A 15.6 31.4%

PAGE 90

79 Table 34 : Geospatial Summary for Sub Basins for Lakewood Gulch Matrix Application Sub Basins Sub Basin Area (km 2 ) % of Total Area Land Use 1 Residential A Land Use 2 Residential B Land Use 3 Commercial Developed Land Cover Totals Area (km 2 ) % of Sub Basin Area (km 2 ) % of Sub Basin Area (km 2 ) % of Sub Basin Total Area (km 2 ) % Developed 1 3.5 6.8% 1.7 48.6% 0.8 22.5% 0.2 6.1% 2.7 77.2% 2 5.1 9.9% 2.7 52.5% 0.8 14.9% 0.2 4.2% 3.6 71.5% 3 7.4 14.5% 3.2 43.0% 1.6 22.1% 0.7 9.3% 5.5 74.4% 4 4.2 8.3% 1.8 42.3% 0.9 22.0% 0.5 12.3% 3.2 76.6% 5 9.7 19.0% 3.9 39.9% 2.3 23.3% 1.5 15.6% 7.6 78.8% 6 5.6 10.9% 2.7 48.6% 0.8 15.0% 0.1 2.5% 3.7 66.2% 7 9.4 18.5% 2.7 28.7% 1.6 17.4% 0.9 9.1% 5.2 55.2% 8 6.3 12.3% 2.0 31.4% 1.5 23.3% 0.5 8.5% 4.0 63.2% Average 6.4 13% 2.6 41.9% 1.3 20.1% 0.6 8.4% 4.4 70.4% Entire Watershed 51.1 100% 20.5 40.2% 10.3 20.1% 4.7 9.1% 35.5 69.6% Figure 20 : Comparison of Sub Basin TSS Loading 1 2 3 4 5 6 7 8 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 TSS TSS ( mg) Sub Basin Loading Comparison for TSS Using 2 yr rainfall event

PAGE 91

80 Figure 21 : Lakewood Gulch Watershed Map

PAGE 92

81 Figure 22 : Lakewood Gulch Watershed Map with Delineated Sub Basins

PAGE 93

82 4.3.2 Validation of M atrix with Mixed Land Use Location V alidation of the matrix required returning to the NSQD to locate a mixed use sampling location from which predicted and measured values could be compared. This sampling location followed the similar site selection process, however, 100% residential and 100% commercial, industrial and institutional were not used as a mixed land use sampling location criteria U sing this process, the North Avenue at Den ver Federal Center was selected for validation as it provided a complete sampling re cord for two years. This 0.28 km 2 sampling area is composed of 33% residential, 30% commercial and 37% open space land use There were twenty sampling events collected at the North Avenue location from 1980 to 1981. The database for these locations provides the storm event runoff that was used in the matrix for the depth of runoff variable for each event. The relationship betwe en predicte d and measured values is shown in Figure 23.

PAGE 94

83 Figure 23 : Validation of Matrix with Mixed Use Sampling Location 0.0 1.0 2.0 3.0 4.0 5.0 6.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 Predicted Values (mg) Measured Values (mg) Matrix Validation using Data Collected from North Avenue at Denver Federal Center Predicted vs. Measured Values

PAGE 95

84 5 Conclusion T he following study provides a straightforward methodology and analysis process for applying regional non point storm water quality data to associate land cover datasets to analyze impacts from urbanization in developing areas. Research on unknown sources of contaminations continue to drive the U.S. EPA to set stricter regulations as amendments to the Clean Water Act look to restore natural streams to their once glorified, beneficial, and fishable state. Although efforts on reviving damaged or lifeless ecosy stems remain costly and time extensive, growing surface water impairments continue to drive urban planners, environmentalists, and water resource engineers to look for new methods to treat urban impacts from decades of urban pollution. The following study provides residential and commercial urban stormwate r quality analysis that support to final determination summary for eac h urban land use constituent. The regional stormwater quality analysis results provide general statistics, r egression analysis of correlations, and t t ests to determine p values for associated non equal variances These results are applied to land cover classification s to develop geospatial analysis process to determine land use percentages for a given study area. Once all data is collected the regional matrix can then be applied for the non point stormwater assessment analysis. Resu lts from this matrix application are used to evaluate and identify urban areas in need of BMP implementation strategies to reduce impacts felt from developed urban. Overall, t he following methods used in this study help understand impacts of different lan d uses with respect to regional water quality data. Using a geospatial approach to evaluate collected runoff data, identification and prioritization of watershed

PAGE 96

85 and sub basins can be determined to locate areas contributing to high pollutant loads to nearb y impaired surface waters. Residential and commercial developments act as key factors limiting these impairments to be restored and repaired. With proper planning, evaluation and selection of optimal BMP designs and strategies, installments and retrofits of efficient stormwater treatment systems can play key factor moving forward as populations and developments continue to grow. Learning to unde rstand impacts prior to installments can be a key factor for evaluating stormwater runoff. Not only to reduce extreme flows during heavy storm events with the use of detention basins but also to produce treatment during minor events that work to protect and restore principles of the urban regime. I mplementation of water quality treatment features and integrated prevention plans will continue to expand as protection of natural lands and restoration of the urban regime will continue to remain a key object ive for engineers, planners, and environmentalists who develop design and standards for innovative sustainable engineering process moving forward.

PAGE 97

86 REFERENCES Island Press. Washington, DC. Point Source Pollution Model. U.S. Department of Agriculture: Natural Resources Conservation Service. < http://go.usa.gov/KFO >. (1 June 2015). Carey, R., Hochmuth, G., Martinez, C., B oyer, T., Dukes, M., Toor, G., and Cisar, J. (2013). Evaluating nutrient impacts in urban watersheds: Challenges and research opportunities. Environmental Pollution 173 138 149. Angeles: Addressing Urban Runoff and Water Supply through Low Impact Developmen (Doctoral Dissertation). Los Angeles, CA. Water, Air and Soil Pollution, 186(1 4), 351 363. Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and Wickham, J., (2011). Photog rammetric Engineering and Remote Sensing, 77(9), 858 864. Journal of Irrigation and Drainage Engineering 135(5), 671 675. Hathaway, J., Tucker, R., Spooner, J., and Hunt, W. first flush effect for nutrients in stormwater runoff from two small urban Water, Air, & Soil Pollution 223(9), 5903 5915. Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., M cKerrow, A., VanDriel, J.N., and Wickham, J. (2007). Completion of the 2001 National Land Cover Database for the Conterminous United States Photogrammetric Engineering and Remote Sensing 73(4), 337 341. Homer, C. G., Dewitz, J. A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N. D., Wickham, J. D., and Megown, K., (2015) National Land Cover Database for the conterminous United States Representing a Photogrammetric En gineering and Remote Sensing 81(5), 345 354. practices for non Journal of Landscape and Urban Planning 104(3 4), 364 372. Lee Water Research 34(6), 1773 1780.

PAGE 98

87 Stormwater and Urban Systems Mo deling Proceedings, Monograph 14. 287 326. point pollution: a screening model Journal of Environmental Management. 74(1), 1 9. National Research Council (1999) National Academies Press. Washington, D.C. National Research Council. (2009a) Urban Stormwater Management in the United States The National Academies Press. Washington, D.C. Natio nal Research Council. (2009b) Assessing the TMDL Approach to Water Quality Management. National Academies Press. Washington, D.C. Journal of America n Water Works Association 107(4), 96 97. Pitt, R., Maestre, A., Morquecho, R., Brown, T., Swann, C., Cappiella, K., and Schueler, National conference on urban stormwater: enha ncing the programs at the local level. (February 2003) Center for Watershed Protection. Ellicott, City, MA. ng Requirements U.S. EPA Office of Wetlands, Oceans and Watersheds. (15 May, 2015) Seattle Public Utilities (2009). "Quality Assurance Project Plan NPDES Phase I Municipal Stormwater Pe Stormwater Treatment Best Management Practices Evaluation. Seattle, WA. Environmental Science & Technology 37(14), 3039 3047. Water Environment Research. 87(2), 169 178. luation of the quality of stormwater in Denver, Colorado, 2006 2010: U.S. Geological Survey Open File Report 2012 U.S. Geological Survey Reston, VA. U.S. Environmental Protection Agency, EPA 440/5 87 001.

PAGE 99

88 U.S. EPA (2007a) Developing Your Stormwater Pollution Prevention Plan: A Guide for Construction Sites. U.S. Environmental Protection Agency. Washington, DC. Nonpoint Source Control Branch Washington, DC. Tracking & Environmental Results. < http://ofmpub.epa.gov/waters10/attains_state.control?p_state=CO >. (15 May 2015) Urban Drain age and Flood Control District. (2013) Urban Storm Drainage Criteria Manual, Volume 3 Best Management Practices Denver, CO. Vogelmann, J., Howard, S., Yang, L., Larson, C., Wylie, B., and Van Driel, J. (2001) Data Set for the conterminous Photogrammetric Engineering and Remote Sensing. (67), 650 662. Walsh, C., Fletcher, T., and PLoS ONE. 7(9): e45814. doi:10. 1371/journal.pone.0045814 Ecological Engineering 18(4), 407 414. Management Pract www.bmpdatabase.org > (5 March 2015). Wright Water Engineers, Inc., Geosyntec Consultants, Pitt, R. and Roesner, L. (2013) Regulation 85 Data Gap Analysis Report. Urban Drainage a nd Flood Control District. Denver, Colorado. Lake Line Magazine. Urban Drainage and Floo d Control District, Denver, CO.

PAGE 100

89 APPENDIX A.1 Raw Data The dataset provided below is the final data set that was used for the statistical and correlative analyses. This dataset includes the sampling site locations, land use types, locations and all constituent data from the stormwater databases and personal co mmunication. The datasets are separated by residential and commercial. Residential ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L) TKN (mg/) N0 2 +NO 3 (mg/L) TP (mg/L) DP (mg/L) Total Copper ( g/L) Total Zinc ( g/L) COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 7/9/1996 4.00 2.00 6.00 0.19 4.00 40.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 4/23/1995 9.00 0.50 1.63 0.08 5.00 2.50 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 5/9/1996 13.00 1.00 1.12 0.12 5.00 2.50 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 6/1/1991 14.00 0.25 5.00 30.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/29/1990 18.00 1.80 0.32 10.00 50.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 6/22/1991 20.00 0.37 30.00 60.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/21/1991 30.00 0.34 100.00 30.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 9/20/1995 40.00 1.10 0.76 0.22 9.00 40.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/12/1992 44.00 10.00 190.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 10/4/1995 59.00 1.60 1.78 0.25 5.00 80.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 9/19/1995 61.00 3.20 1.62 0.51 10.00 2.50 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 5/29/1990 72.00 1.50 0.39 10.00 50.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 8/3/1991 72.00 0.16 50.00 150.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 5/16/1995 76.00 2.00 3.23 0.18 11.00 60.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/2/1992 88.00 30.00 200.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 6/28/1995 103.00 0.30 8.32 0.20 5.00 2.50 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 6/19/1990 122.00 3.00 0.54 60.00 10.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 10/22/1995 124.00 10.10 1.77 1.04 20.00 130.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 9/6/1996 128.00 3.90 1.85 0.45 23.00 110.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 6/22/1994 140.00 10.50 3.16 1.03 12.00 130.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 5/25/1996 150.00 3.20 1.71 0.46 39.00 140.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 4/18/1995 153.00 3.70 0.93 0.53 5.00 2.50 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/5/1990 164.00 2.30 0.37 50.00 110.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 7/13/1995 175.00 11.60 1.68 1.00 130.00 590.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 8/18/1995 182.00 4.40 2.85 0.58 17.00 110.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 9/29/1995 203.00 3.00 2.16 0.10 23.00 150.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 6/22/1996 221.00 12.20 0.08 0.61 20.00 220.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/8/1990 292.00 2.00 0.47 30.00 130.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/15/1992 294.00 80.00 240.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 4/10/1995 306.00 1.80 0.55 0.43 5.00 2.50 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 6/13/1995 324.00 6.20 1.82 0.75 26.00 200.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 8/17/1990 352.00 2.50 1.08 30.00 60.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 4/23/1995 389.00 4.60 1.68 1.83 5.00 2.50 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 8/23/1996 438.00 6.90 1.81 0.81 31.00 250.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 6/15/1996 501.00 5.70 1.32 0.71 5.00 160.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 9/11/1996 521.00 1.40 1.04 0.11 16.00 170.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/14/1990 656.00 4.10 0.87 50.00 200.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/19/1994 999.00 7.30 1.45 1.16 68.00 420.00

PAGE 101

90 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L) TKN (mg/) N0 2 +NO 3 (mg/L) TP (mg/L) DP (mg/L) Total Copper ( g/L) Total Zinc ( g/L) COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/7/1990 2.70 0.47 10.00 60.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/28/1990 0.80 0.36 40.00 50.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 8/4/1990 2.70 0.71 40.00 190.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 9/5/1990 2.90 0.39 40.00 100.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 9/19/1990 1.80 0.44 30.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 4/13/1991 0.12 20.00 130.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 4/30/1991 0.34 20.00 60.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 5/23/1991 0.54 30.00 140.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/10/1991 0.47 60.00 50.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/23/1991 0.25 50.00 40.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 6/1/1992 60.00 140.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 6/5/1992 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/17/1992 30.00 80.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/20/1992 30.00 100.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 7/22/1992 10.00 80.00 COAUSHC R Shop_Creek_Wetland Pond_1990 94_S1 39.629 1 104.7415 RE 8/23/1992 80.00 80.00 COAUSHC R Shop_Creek_Wetland Pond_1995 97_S1_95 97 39.629 1 104.7415 RE 8/7/1996 5.50 2.33 0.83 30.00 280.00 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 9/21/2006 7.50 4.45 2.46 0.26 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 5/10/2006 8.30 2.50 0.78 0.19 0.26 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 4/7/2006 28.80 2.60 0.63 0.23 0.22 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 7/14/2009 37.50 1.32 0.24 0.37 0.17 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 5/30/2001 45.00 0.63 0.71 0.19 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 5/2/2008 46.50 2.22 0.69 0.23 0.37 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 8/19/2004 56.00 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 8/27/2004 56.00 1.04 0.24 0.12 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 8/20/2000 78.00 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 6/16/2002 83.00 1.26 0.72 0.71 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 4/17/2007 95.85 5.80 1.91 0.40 0.24 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 9/12/2006 112.50 4.10 1.23 0.17 0.19 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 9/19/2001 128.00 3.46 0.53 0.35 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 6/17/2003 150.00 0.51 0.50 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 5/13/2004 157.00 0.54 0.45 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 4/27/2009 172.00 4.48 1.03 0.45 0.24 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 6/6/2007 194.00 0.80 0.65 0.22 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 8/6/2007 200.00 2.90 1.47 0.67 0.37 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 3/16/2000 210.00 0.79 0.22 0.17 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 5/2/2007 233.00 7.50 2.08 0.72 0.31 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 9/15/2005 239.00 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 6/26/2004 276.00 1.41 0.64 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 5/13/2002 279.00 1.91 0.64 0.35 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 5/18/2001 300.00 0.94 0.25 0.13 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 6/4/2002 308.00 1.36 0.71 0.29 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 4/3/2008 422.00 3.36 1.48 0.72 0.16 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 6/4/2008 454.00 4.20 0.65 0.73 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 6/10/2004 680.00 1.01 1.03 0.22 CODEGRH E Grant_Heron 39.619 7 105.0582 RE 5/16/2003 814.00 1.27 0.92 0.22 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 9/12/2006 7.00 4.00 2.09 0.16 0.17 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 7/14/2009 14.30 1.19 0.04 0.35 0.26 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 7/28/2009 27.50 1.56 0.01 0.61 0.46 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 9/21/2006 33.70 0.43 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 8/19/2004 53.60

PAGE 102

91 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L) TKN (mg/) N0 2 +NO 3 (mg/L) TP (mg/L) DP (mg/L) Total Copper ( g/L) Total Zinc ( g/L) CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 8/27/2004 53.60 0.88 0.25 0.15 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 6/26/2004 65.30 0.88 0.40 0.17 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 5/10/2006 73.35 3.70 1.24 0.30 0.20 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 3/16/2000 81.00 0.95 0.40 0.31 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 4/17/2007 85.00 4.10 1.48 0.27 0.22 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 5/2/2008 88.00 2.24 0.48 0.43 0.34 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 4/7/2006 169.50 5.30 1.02 0.40 0.25 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 5/23/2009 253.80 2.31 1.04 0.50 0.10 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 4/27/2009 257.00 5.90 1.55 0.86 0.46 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 8/6/2007 258.00 0.45 1.49 0.64 0.37 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 5/13/2004 288.00 0.95 0.83 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 6/6/2007 322.50 0.40 0.85 0.17 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 5/2/2007 525.00 7.20 1.76 1.30 0.24 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 5/18/2001 530.00 1.40 0.11 0.14 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 6/16/2002 602.00 0.47 1.51 1.33 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 4/3/2008 616.00 3.45 1.60 0.82 0.44 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 6/4/2008 662.00 2.96 0.53 0.45 CODEGRRE Grant_Reflect 39.618 4 105.0594 RE 5/16/2003 1210.00 1.15 1.71 0.31 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/23/2009 5.30 0.991 1.73 1.62 9.10 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/21/2006 6.50 2.70 2.13 0.47 0.53 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/20/2011 7.00 1.60 0.70 0.31 0.23 5.20 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/21/2006 8 2.7 2.13 ** 0.528 6.00 30.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/5/2005 9 1.9 0.37 0.251 0.237 3.00 30.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/10/2010 9 3.2 0.47 0.47 0.41 4.30 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/17/2008 9.20 2.01 1.14 0.21 0.17 73.60 12.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/6/2007 10.05 4.60 2.12 0.36 0.32 15.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/26/2006 12 3.6 0.81 ** 0.47 8.00 40.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/10/2006 12.00 3.60 0.81 0.51 0.47 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/28/2006 13.00 1.60 2.65 0.37 0.31 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/2/2010 16.00 2.80 1.22 0.52 0.43 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/20/2009 17.4 1.023 0.967 0.86 8.40 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/7/2006 17.50 1.10 0.54 0.21 0.26 14.00 80.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/12/2011 18.00 1.10 1.03 0.19 0.12 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/12/2010 19.00 1.50 1.05 0.21 0.14 5.40 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/13/2010 21.00 1.00 0.67 0.14 0.08 5.10 28.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/3/2009 22.20 1.75 0.63 0.15 0.09 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/11/2008 23.40 0.99 0.77 0.12 0.08 4.80 18.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/3/2008 24.00 1.71 1.11 0.24 0.16 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/12/2010 25 1.5 1.07 0.24 0.18 5.50 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/11/2005 26 1.3 0.6 0.209 0.154 7.00 60.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/2/2007 26 2.80 0.58 0.28 0.117 8.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/26/2010 26 1.3 0.31 0.14 0.1 3.10 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/13/2009 27 0.185 0.382 0.159 7.80 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/17/2007 28.00 3.00 1.05 0.38 0.17 6.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/23/2010 32.00 1.20 0.53 0.37 0.32 4.10 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/8/2010 33.00 4.00 0.95 0.75 0.64 8.50 26.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/9/2010 33 4 0.95 0.75 0.63 9.60 37.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/4/2010 34.00 3.80 0.64 0.26 0.05 8.60 45.30 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/15/2008 34.20 1.36 0.66 0.23 0.17 20.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/25/2009 34.50 0.27 0.74 0.54 11.10 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/31/2005 35 0.61 0.204 0.169 9.00 50.00

PAGE 103

92 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L) TKN (mg/) N0 2 +NO 3 (mg/L) TP (mg/L) DP (mg/L) Total Copper ( g/L) Total Zinc ( g/L) CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/27/2009 35.20 0.79 0.37 0.16 5.80 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/22/2010 38.00 1.60 0.94 0.20 0.12 7.00 111.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/14/2011 39.00 1.40 0.62 0.20 0.09 5.20 31.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/11/2011 40.00 2.90 0.55 0.41 0.30 26.70 34.80 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/14/2011 42.00 1.20 0.45 0.13 0.04 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 3/16/2000 44.00 0.75 0.38 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/25/2009 49 0.785 0.443 0.3 15.60 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/25/2009 49.00 0.44 0.30 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/24/2011 52.00 2.70 1.14 0.37 0.26 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/28/2011 52 2.7 1.14 0.37 0.26 9.00 40.40 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/14/2011 52.00 1.70 0.42 0.27 0.17 4.50 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/18/2011 58.00 1.80 0.65 0.22 0.12 5.20 23.50 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/20/2010 59.00 2.40 1.38 0.27 0.08 10.60 84.20 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/26/2005 64 2.4 0.72 0.194 0.103 20.00 120.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/11/2010 64.00 2.50 0.59 0.28 0.17 7.50 35.30 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/12/2006 66.15 4.80 0.31 0.28 0.12 33.00 130.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/16/2002 67.00 4.50 1.60 0.54 0.56 13.00 68.40 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/8/2008 71.00 1.78 0.62 0.29 0.18 10.00 70.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/26/2009 72.00 1.5 0.25 0.13 12.00 63.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/9/2009 75.00 1.122 0.76 0.41 17.40 62.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/3/2002 85 2.1 1.06 0.45 0.317 5.00 38.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/27/2010 88.00 3.50 1.94 0.32 0.05 10.40 68.10 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/16/2009 90.60 0.506 0.31 0.16 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/21/2010 94.00 13.10 2.13 0.82 0.55 34.10 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/11/2010 94.00 1.50 1.07 0.32 0.18 5.50 29.80 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/14/2001 98 3.4 1.17 0.154 0.164 10.00 37.70 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/7/2007 103 7.3 0.8 ** 0.154 31.00 80.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/6/2011 103.00 3.20 0.26 0.41 0.06 10.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/28/2010 104.00 3.50 1.37 0.89 0.78 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/6/2010 104 3.5 1.37 0.89 0.78 6.60 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/15/2004 107 2.4 0.99 0.323 0.483 16.00 70.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/6/2007 111.50 0.80 0.62 0.11 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/27/2010 120.00 3.10 0.61 0.30 0.04 11.60 63.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/26/2008 123.80 2.09 1.10 0.48 0.09 26.20 150.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/17/2001 124 2.4 1.28 0.508 0.324 14.00 61.50 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/19/2001 124.00 1.28 0.51 0.32 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/10/2006 128.00 3.60 1.15 0.36 0.22 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/11/2002 132 5.9 1.55 0.498 0.298 15.00 56.70 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/15/2007 138.00 7.12 2.21 0.89 30.00 113.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/22/2010 140.00 2.50 0.81 0.30 0.06 12.40 81.30 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/10/2010 144.00 7.10 0.06 0.42 0.04 15.90 96.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/23/2009 146.30 2.72 1.08 0.52 0.11 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/4/2004 152 5.8 0.199 0.51 0.157 25.00 130.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 4/26/2011 176.00 3.90 0.48 0.63 0.36 13.30 70.20 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/7/2011 180.00 1.40 0.48 0.43 0.08 9.70 52.70 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/23/2010 182.00 2.00 0.70 0.43 0.15 16.70 80.70 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/31/2009 208.00 0.95 0.37 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/1/2009 208 1.066 0.953 0.365 30.60 170.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 9/13/2002 209 2.4 0.99 0.304 0.143 16.00 59.40 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/17/2003 216.00 2.80 0.91 0.40 0.39 7.00 128.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/19/2010 221.00 5.20 1.22 0.69 0.15

PAGE 104

93 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L) TKN (mg/) N0 2 +NO 3 (mg/L) TP (mg/L) DP (mg/L) Total Copper ( g/L) Total Zinc ( g/L) CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/26/2010 226.00 9.40 0.06 0.68 0.06 21.30 119.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/26/2004 252.00 1.76 0.94 0.23 15.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/4/2008 266.00 1.33 0.77 0.31 0.07 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/27/2004 320.00 0.88 0.52 0.18 17.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/28/2004 337.00 1.02 0.53 0.19 7.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/8/2010 338.00 9.10 0.71 0.83 0.08 23.30 151.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/14/2010 342.00 3.00 0.85 0.35 0.08 12.20 79.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/11/2001 398 2.8 0.85 0.584 0.61 25.00 126.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/18/2001 415.00 1.17 0.18 0.14 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 7/28/2003 486 4.1 1.59 0.56 0.151 20.00 140.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/30/2001 490.00 0.53 0.67 0.22 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/31/2007 494 11.6 0.16 ** 0.089 20.00 80.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/4/2002 498.00 3.1 0.78 0.729 0.25 24.00 157.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/2/2008 512.00 2.80 2.70 1.25 0.39 70.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/13/2004 513.00 0.76 0.48 0.188 17.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 8/23/2007 648.00 3.10 0.88 0.16 13.00 57.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/16/2003 692.00 5.30 1.52 0.90 0.27 37.00 240.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 6/10/2004 1280.00 1.15 1.12 0.12 27.00 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/5/2006 3.6 0.81 ** 0.164 CODEORPO UDFCD_Orchard_Pond 39.621 1 105.0598 RE 5/25/2007 ** ** COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/23/2012 4.00 0.80 0.81 0.49 9.60 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/25/2012 11.00 2.00 1.80 0.23 5.80 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/22/2013 19 1.40 0.44 0.18 0.14 5.10 22.90 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/12/2012 21.00 1.10 0.49 0.16 4.90 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/19/2011 36.00 3.20 1.66 1.62 1.23 26.70 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/12/2014 39 2.50 1.08 0.31 0.24 10.90 44.70 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/11/2012 40.00 1.10 0.50 0.15 0.09 6.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/5/2014 48 1.60 0.59 0.22 0.12 14.40 69.80 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/25/2013 52.00 3.50 0.82 0.35 0.04 11.80 61.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/9/2013 52 1.10 0.67 0.16 0.07 5.10 21.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/19/2011 64.00 1.10 0.37 0.20 0.08 8.00 43.50 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/11/2012 64.00 3.80 1.08 0.35 13.60 84.30 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/11/2011 70.00 1.50 0.50 0.25 0.16 29.80 49.10 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/15/2013 73.00 1.60 0.96 6.00 36.10 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/23/2014 78 1.90 1.03 0.26 0.09 9.30 51.10 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/30/2014 84 1.80 0.56 0.27 0.13 9.90 71.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/14/2011 105.00 1.80 0.91 0.23 0.06 10.70 73.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 8/11/2012 108.00 3.40 0.76 0.38 8.90 58.80 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/11/2013 121.00 2.50 0.94 0.29 12.00 65.80 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/6/2012 127.00 2.40 0.57 0.39 0.22 12.00 67.60 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 8/29/2014 136 1.70 2.57 0.21 0.09 9.60 48.40 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/29/2013 149.00 4.40 0.90 0.48 0.25 20.90 130.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 6/4/2013 149.00 1.80 0.44 0.27 0.10 10.50 55.60 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/26/2011 158.00 3.10 0.76 0.36 0.05 16.50 102.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/10/2014 158 1.30 0.58 0.21 0.16 7.40 36.70 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/31/2012 161.00 2.80 0.36 0.46 13.70 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 6/16/2013 166.00 6.80 0.56 0.85 0.37 30.70 112.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/27/2011 170.00 2.90 1.12 0.27 0.03 14.80 99.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/14/2011 178.00 3.10 0.99 0.36 0.22 14.20 153.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/11/2012 179.00 0.90 0.23 0.14 7.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/17/2014 193 2.50 0.63 0.3 0.07 16.80 95.50

PAGE 105

94 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L) TKN (mg/) N0 2 +NO 3 (mg/L) TP (mg/L) DP (mg/L) Total Copper ( g/L) Total Zinc ( g/L) COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/27/2013 205.00 3.20 0.03 0.36 0.01 9.20 68.80 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/18/2011 232.00 1.80 0.59 0.38 0.12 17.00 94.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 6/19/2014 261 5.10 0.77 0.65 0.18 19.30 159.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 6/9/2014 265 3.70 0.72 0.61 20.30 146.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 6/19/2011 274.00 4.50 0.63 0.75 0.33 20.10 148.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/9/2013 318.00 3.00 0.68 0.49 0.09 22.40 137.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 8/21/2014 356 3.60 1.02 0.58 0.11 28.10 247.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/9/2014 415 3.20 0.51 0.8 0.13 27.50 190.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/16/2014 430 5.00 0.47 0.72 0.07 30.40 239.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/13/2011 476.00 2.40 0.41 0.55 0.09 23.20 123.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/22/2014 479 4.40 1.33 0.7 0.05 30.00 216.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 8/26/2014 506 2.80 0.81 0.6 0.23 26.40 192.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/31/2014 518 3.10 0.11 0.52 0.08 18.20 116.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 8/11/2013 568.00 4.80 0.04 0.75 0.02 33.00 252.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/5/2012 641.00 5.80 0.73 1.00 0.22 27.80 191.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/13/2013 641.00 3.50 0.33 0.44 14.50 91.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/8/2014 718 6.40 0.80 1.21 0.38 48.90 339.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/7/2011 766.00 1.80 0.65 0.39 11.20 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 8/3/2013 790.00 3.80 0.03 0.78 0.07 23.30 173.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/1/2013 902.00 8.80 1.51 1.76 53.60 468.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 7/12/2011 947.00 3.00 0.55 0.71 0.07 31.00 166.00 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 9/30/2014 986 6.70 1.35 0.87 0.14 44.80 91.90 COLAIRIS 21st_and_Iris_Rain_Garden 39.748 8 105.1066 RE 5/20/2013 1310.00 13.40 0.40 1.91 0.05 45.90 308.00 Commercial ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L ) TKN (mg/) N0 2 +N O 3 (mg/L) TP (mg/L ) DP (mg/L ) Total Coppe r ( g/L) Total Zinc ( g/L) COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 4/10/200 8 302.00 0.50 0.02 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 4/17/200 8 6.40 0.07 0.01 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 5/2/2008 153.60 0.23 0.08 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 5/13/200 8 16.40 0.07 0.03 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 5/15/200 8 32.80 0.08 0.04 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 5/27/200 8 32.50 0.12 0.04 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 6/4/2008 123.00 0.12 0.01 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 7/9/2008 81.50 0 .14 0.04 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 8/12/200 8 137.40 0.35 0.08 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 8/15/200 8 76.60 0.24 0.12 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 9/12/200 8 43.70 0.14 0.07 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 10/6/ 200 8 141.00 0.54 0.12 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 10/13/20 08 11.00 0.16 0.03 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 4/17/200 9 8.60 0.01 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 4/26/200 9 37.30 0.09 0.04 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 5/8/2009 13.00 0.04

PAGE 106

95 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L ) TKN (mg/) N0 2 +N O 3 (mg/L) TP (mg/L ) DP (mg/L ) Total Coppe r ( g/L) Total Zinc ( g/L) COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 5/20/200 9 4.00 0.01 0.01 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 5/25/200 9 89.80 0.11 0.01 COACWW L3 Arapahoe_County_Water_and_Wastewater_Author ity_L3 39.60 05 104.83 79 COM 6/1/2009 95.00 0.17 0.03 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 4/10/200 8 18.60 0.30 0.21 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 4/17/200 8 13.60 0.20 0.16 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 5/2/2008 186.00 0.32 0.19 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 5/13/200 8 13.00 0.15 0.13 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 5/15/200 8 8.40 0.12 0.10 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 6/20/200 8 69.00 0.40 0.21 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 7/9/2008 56.50 0.19 0.14 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 8/7/2008 14.10 0.35 0.24 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 8/12/200 8 51.40 0.32 0.23 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 8/15/200 8 20.80 0.36 0.30 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 9/12/200 8 16.30 0.32 0.29 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 4/17/200 9 57.80 0.39 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 4/26/200 9 40.30 0.17 0.10 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 5/8/2009 34.70 0.32 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 5/25/200 9 20.50 0.17 0.12 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 5/31/200 9 21.70 0.18 0.13 COACW6 W7 Arapahoe_County_Water_and_Wastewater_Author ity_W6W7 39.59 19 104.82 06 COM 6/1/2009 20.50 0.11 0.08 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 8/8/2008 1360.0 0 4.50 0.19 0.90 0.04 31.70 188.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 8/15/200 8 131.00 1.20 0.13 0.04 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 9/11/200 8 5.80 4.20 0.46 0.02 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 4/16/200 9 129.00 3.00 0.96 0.32 0.12 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/22/200 9 4.70 4.70 0.72 0.48 0.14 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/24/200 9 436.00 2.50 0.19 0.38 0.05 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 6/2/2009 116.00 1.80 0.73 0.15 0.03 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 6/13/200 9 104.00 2.20 0.64 0.14 0.03 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 6/25/200 9 235.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 6/26/200 9 230.00 4.10 0.05 0.36 0.09 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/1/2009 91.00 1.80 0.48 0.19 0.10 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/3/2009 55.00 2.00 0.34 0.14 0.06 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/20/200 9 481.00 4.20 0.15 0.65 0.18 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/25/200 9 427.00 1.80 0.18 0.26 0.03 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/26/200 9 80.00 2.30 0.22 0.32 0.15 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/29/200 9 78.00 1.80 0.61 0.10 0.04 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 8/9/2009 141.00 2.40 0.43 0.19 0.07 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 4/22/201 0 886.00 3.70 0.29 0.72 0.09 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 4/23/201 0 301.00 1.40 0.05 0.28 0.03 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/11/201 0 91.00 2.10 0.33 0.34 0.20

PAGE 107

96 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L ) TKN (mg/) N0 2 +N O 3 (mg/L) TP (mg/L ) DP (mg/L ) Total Coppe r ( g/L) Total Zinc ( g/L) CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/14/201 0 202.00 1.60 0.55 0.17 0.01 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 6/12/201 0 147.00 2.00 0.11 0.28 0.06 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/4/2010 427.00 2.80 0.09 0.48 0.04 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/7/2010 384.00 2.80 0.31 0.49 0.17 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/20/201 0 414.00 6.70 0.08 1.32 0.47 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 8/24/201 0 102.00 3.10 0.81 0.37 0.18 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/11/201 1 327.00 6.50 0.99 0.78 0.43 26.20 158.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/14/201 1 64.00 2.00 0.85 0.17 0.03 13.10 52.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/18/201 1 546.00 4.10 0.83 0.74 0.03 54.10 326.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/19/201 1 323.00 2.20 0.41 0.35 0.04 25.00 165.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/24/201 1 1310.0 0 3.90 0.62 0.73 0.05 61.60 402.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 6/13/201 1 1730.0 0 12.40 0.86 2.31 0.08 165.00 896.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 6/20/201 1 571.00 4.40 0.56 0.49 0.05 47.50 276.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/7/2011 4.80 0.07 1.65 0.03 83.40 495.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/10/201 1 221.00 2.30 0.95 0.36 0.05 20.90 72.10 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/11/201 1 74.00 1.40 0.55 0.22 0.07 13.90 52.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/12/201 1 585.00 2.80 0.93 0.70 0.04 33.60 185.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/13/201 1 1540.0 0 3.20 0.13 1.13 0.03 63.40 371.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/14/201 1 190.00 1.10 0.62 0.12 0.03 14.10 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/19/201 1 238.00 3.40 1.16 0.44 0.06 27.00 135.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/26/201 1 180.00 3.00 0.84 0.36 0.06 24.60 111.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 8/1/2011 355.00 4.10 0.59 0.64 0.07 38.40 215.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 9/6/2011 69.00 11.30 3.44 0.77 0.56 74.60 155.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 9/7/2011 59.00 3.00 0.90 0.24 0.14 26.00 55.20 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 9/14/201 1 72.00 2.20 0.61 0.21 0.12 16.80 60.80 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/7/2013 877.00 8.50 0.41 1.48 93.40 529.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/11/201 3 77.00 2.60 0.35 0.51 17.50 64.80 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/12/201 3 160.00 4.50 0.35 0.44 34.30 92.40 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/27/201 3 280.00 4.00 0.40 0.13 31.20 125.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 8/12/201 3 25.00 3.50 0.38 0.25 24.30 34.80 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 9/11/201 3 11.00 1.10 0.12 0.05 7.50 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 9/22/201 3 68.00 2.70 0.26 0.18 17.30 50.40 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/8/2014 2260.0 0 23.90 3.61 4.44 1.59 224.00 1440.0 0 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/9/2014 160.00 2.90 0.58 0.51 0.18 33.90 171.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/12/201 4 87.00 2.30 2.14 0.30 0.08 17.90 87.90 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/22/201 4 1240.0 0 6.80 0.35 1.24 0.06 101.00 649.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 5/23/201 4 1590.0 0 5.60 0.34 1.52 0.03 104.00 624.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 6/19/201 4 649.00 3.00 0.42 0.49 0.09 33.90 196.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/8/2014 696.00 5.40 0.68 0.84 0.07 61.70 375.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/14/201 4 226.00 3.30 0.08 0.41 0.07 32.60 180.00

PAGE 108

97 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L ) TKN (mg/) N0 2 +N O 3 (mg/L) TP (mg/L ) DP (mg/L ) Total Coppe r ( g/L) Total Zinc ( g/L) CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/17/201 4 41.00 2.20 0.51 0.14 0.03 14.30 42.20 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/24/201 4 345.00 3.10 0.87 0.45 0.06 33.50 170.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 7/30/201 4 437.00 2.50 1.10 0.39 0.03 33.60 189.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 8/27/201 4 358.00 2.10 0.47 0.45 0.04 35.00 234.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 8/28/201 4 171.00 1.40 0.55 0.23 0.05 21.20 111.00 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 9/5/2014 15.00 1.60 0.63 0.16 0.08 21.40 76.40 CODEWA WA Denver_Wastewater_Building 39.72 09 105.01 06 COM 9/10/201 4 39.00 1.50 0.53 0.19 0.14 52.90 73.50 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/4/2005 1.00 0.43 0.17 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/9/2005 1.00 0.43 0.17 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/12/200 5 13 0.30 1.72 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/20/200 5 85 1.50 0.40 0.06 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/21/200 5 29 4.00 2.98 0.12 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/3/2005 16 1.70 0.85 0.09 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/4/2005 17 1.80 0.46 0.08 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/9/2005 16 4.50 1.76 0.10 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/10/200 5 5 1.40 1.26 0.08 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/20/200 5 63 3.70 1.61 0.21 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/30/200 6 1.10 0.24 0.10 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/3/2006 2.40 0.80 0.30 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/9/2006 3.20 0.99 0.22 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/3/2006 2.40 0.79 0.14 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/21/200 6 2.60 0.77 0.13 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/10/200 7 40 0.17 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/16/200 7 115 0.20 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/23/200 7 59.5 0.07 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/1/2007 455 0.44 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/7/2007 46 0.12 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/14/200 7 303 0.24 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/29/200 7 165 0.18 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/12/200 7 37 0.15 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/8/2007 297 0.42 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/27/200 7 35 0.20 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/5/2007 130 0.07 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/10/200 7 518 0.37 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/2/2008 27 1.30 0.93 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/14/200 8 26 0.80 0.05 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/27/200 8 156 3.00 0.30 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/6/2008 19 1.70 0.08 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/18/200 8 16 0.80 0.06 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/15/200 8 21 1.00 0.07

PAGE 109

98 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L ) TKN (mg/) N0 2 +N O 3 (mg/L) TP (mg/L ) DP (mg/L ) Total Coppe r ( g/L) Total Zinc ( g/L) COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/16/200 9 65 1.40 0.28 0.16 0.05 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/17/200 9 53 1.20 0.13 0.15 0.04 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/26/200 9 47 1.40 0.31 0.12 0.04 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/23/200 9 1020 4.80 0.73 0.06 0.97 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/25/200 9 29 0.60 0.12 0.05 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/1/2009 137 1.80 0.48 0.15 0.03 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/7/2009 38 2.50 1.15 0.12 0.03 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/11/200 9 41 1.40 0.82 0.09 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/13/200 9 1180 1.70 0.33 0.34 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/14/200 9 539 1.20 0.56 0.22 0.03 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/23/200 9 39 2.60 1.14 0.15 0.05 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/26/200 9 1.30 0.73 0.14 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/26/200 9 190 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/1/2009 156 1.40 0.38 0.14 0.04 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/3/2009 347 2.10 0.69 0.26 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/4/2009 50 1.40 0.32 0.05 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/10/200 9 59 2.30 1.31 0.12 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/20/200 9 465 2.80 0.30 0.58 0.14 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/25/200 9 7 0.70 0.03 0.16 0.08 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/27/200 9 293 1.50 0.45 0.24 0.05 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/29/200 9 20 1.40 0.34 0.12 0.05 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/6/2009 94 2.40 1.27 0.14 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/9/2009 326 1.20 0.42 0.19 0.07 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/17/200 9 164 2.40 0.64 0.16 0.08 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/18/200 9 31 1.60 0.69 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/12/200 9 32 4.40 1.64 0.31 0.19 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/23/200 9 15 1.60 0.78 0.23 0.18 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/21/201 0 620 3.10 0.45 0.49 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/22/201 0 135 0.70 0.15 0.14 0.04 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/23/201 0 163 0.50 0.04 0.11 0.01 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/11/201 0 264 2.40 0.48 0.36 0.08 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/12/201 0 80 1.70 0.46 0.16 0.05 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/13/201 0 39 0.80 0.49 0.12 0.06 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/13/201 0 1940 3.70 1.45 0.41 0.04 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/11/201 0 1510 4.10 0.62 0.65 0.03 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/13/201 0 52 0.50 0.14 0.07 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/26/201 0 149 8.90 1.92 0.33 0.04 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/4/2010 571 5.20 0.68 0.50 0.03 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/6/2010 303 2.20 0.36 0.26 0.03 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/19/201 0 203 5.70 0.90 0.40 0.04

PAGE 110

99 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L ) TKN (mg/) N0 2 +N O 3 (mg/L) TP (mg/L ) DP (mg/L ) Total Coppe r ( g/L) Total Zinc ( g/L) COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/20/201 0 61 2.30 1.27 0.14 0.02 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/22/201 0 127 1.80 0.94 0.17 0.03 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/9/2010 134 2.50 0.88 0.23 0.07 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/18/201 1 125 2.40 0.50 0.26 0.05 11.80 80.90 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/10/201 1 525 3.30 0.22 0.41 0.04 18.30 137.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/10/201 1 30 0.70 0.17 0.06 0.05 6.70 20.60 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/18/201 1 757 2.40 0.41 0.80 0.00 35.00 215.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/19/201 1 42 0.40 0.18 0.05 0.00 39.50 22.70 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/19/201 1 234 1.80 0.21 0.26 0.04 12.20 93.40 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/19/201 1 50 1.20 0.30 0.07 0.00 10.20 45.90 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/7/2011 763 2.10 0.53 0.36 0.03 19.90 117.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/12/201 1 219 0.90 0.20 0.66 0.04 12.00 53.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/13/201 1 243 1.30 0.22 0.15 0.00 7.90 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/19/201 1 184 1.20 0.72 0.23 0.00 9.20 55.80 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/26/201 1 55 1.20 0.53 0.10 0.06 5.40 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/27/201 1 96 2.40 0.80 0.11 0.03 3.60 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/11/201 2 7 0.60 0.18 0.06 3.70 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/25/201 2 8 1.40 0.33 0.13 4.10 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 4/13/201 3 156 1.80 0.37 0.21 0.07 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/7/2013 129 3.40 0.83 0.21 0.08 11.40 63.60 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/8/2013 58 1.30 0.34 0.17 0.08 7.40 34.40 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/20/201 3 35 1.80 0.72 0.13 0.05 8.60 26.20 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/29/201 3 38 1.30 0.32 0.11 0.06 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/1/2013 22 3.20 0.97 0.17 9.60 44.50 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/6/2013 134 2.30 0.53 0.15 11.70 73.20 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/11/201 3 308 2.50 0.63 0.31 15.60 107.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/13/201 3 226 1.00 0.31 0.16 5.40 37.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/9/2013 87 2.00 0.76 0.15 0.06 15.30 75.30 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/10/201 3 7 0.80 1.17 0.02 0.05 2.60 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/22/201 3 16 0.50 0.18 0.05 0.04 1.50 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/12/201 4 56 0.90 0.30 0.12 0.00 4.80 31.60 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/23/201 4 193 1.90 0.81 0.24 0.02 16.30 103.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 5/31/201 4 672 2.80 0.49 0.45 0.03 26.40 191.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/9/2014 192 1.50 0.32 0.26 10.50 64.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 6/19/201 4 64 2.70 0.96 0.24 0.05 8.70 30.80 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/4/2014 18 3.70 0.08 0.14 0.03 10.70 38.80 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/17/201 4 23 1.30 0.41 0.08 0.03 4.40 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/30/201 4 10 0.00 0.31 0.05 0.11 3.10 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 7/30/201 4 9 0.40 0.17 0.02 0.02 2.10 0.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/8/2014 9 1.50 0.65 0.06 0.01 7.00 0.00

PAGE 111

100 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L ) TKN (mg/) N0 2 +N O 3 (mg/L) TP (mg/L ) DP (mg/L ) Total Coppe r ( g/L) Total Zinc ( g/L) COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/21/201 4 13 2.30 1.10 0.15 0.10 8.10 20.80 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/27/201 4 82 0.90 0.51 0.06 0.01 5.80 13.50 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 8/28/201 4 12 0.60 0.33 0.04 0.03 4.10 10.20 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/5/2014 8 0.90 0.54 0.07 0.05 8.20 19.40 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/22/201 4 16 1.00 0.23 0.07 0.06 4.20 26.00 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/29/201 4 66 2.00 0.50 0.11 0.02 8.40 44.30 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 9/30/201 4 106 1.30 0.45 0.11 0.02 9.90 50.60 COLASHO P Lakewood_Shops 39.87 48 105.16 30 COM 10/10/20 14 47 1.30 0.38 0.07 0.06 6.10 48.50 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 6/4/2005 8 0.70 0.70 0.02 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 6/9/2005 1.00 0.43 0.17 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 6/10/200 5 63 3.70 1.61 0.21 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 6/12/200 5 85 1.50 0.40 0.06 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 6/21/200 5 13 1.72 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 8/4/2005 16 1.70 0.85 0.09 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 8/10/200 5 16 4.50 1.76 0.10 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 8/23/200 5 5 1.40 1.26 0.08 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 4/28/200 6 16 2.60 0.77 0.13 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 6/24/200 6 111 2.40 0.80 0.30 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 8/3/2006 17 2.40 0.79 0.14 COLAMOP 6 UDFCD_Modular_Porous_Pavement_05_to_06 39.88 33 105.20 00 COM 9/22/200 6 16 2.60 0.77 0.13 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 4/16/199 5 32 3.10 1.29 0.11 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 4/17/199 5 36 0.70 0.91 0.09 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 4/29/199 5 27 1.50 0.67 0.13 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 5/16/199 5 11 0.60 0.15 0.01 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 6/17/199 5 22 1.10 0.17 0.04 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/13/199 5 88 2.80 1.04 0.18 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/18/199 5 175 2.50 1.14 0.12 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 4/27/199 6 10 1.10 0.70 0.01 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 5/29/199 6 41 13.00 0.49 0.11 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/13/199 6 450 9.90 2.88 0.49 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/16/199 6 126 9.90 1.13 0.17 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/27/199 7 182 4.30 2.03 0.12 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/28/199 7 57 1.40 0.36 0.04 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/1/1997 18 1.60 0.98 0.08 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/4/1997 18 0.90 0.40 0.03 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/17/199 7 25 1.00 1.50 0.11 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 9/22/199 7 14 0.80 0.95 0.03 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 5/4/1998 199 2.30 0.95 0.50 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 5/8/1998 218 1.30 0.63 0.38 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 5/22/199 8 52 2.20 0.66 0.14

PAGE 112

101 ID STATION NAME Lat. Long. Land Use Type Date of Event TSS (mg/L ) TKN (mg/) N0 2 +N O 3 (mg/L) TP (mg/L ) DP (mg/L ) Total Coppe r ( g/L) Total Zinc ( g/L) COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 6/4/1998 11 2.10 0.63 0.09 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 6/14/199 8 37 1.80 0.33 0.18 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/8/1998 20 2.30 1.05 0.26 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/22/199 8 22 2.20 0.85 0.17 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/23/199 8 21 1.20 0.43 0.14 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/30/199 8 19 1.20 0.63 0.07 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/10/199 8 26 0.90 1.20 0.07 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 9/1/1998 25 1.00 0.52 0.20 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/15/200 4 4 1.10 1.82 0.08 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/16/200 4 17 1.10 1.01 0.10 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 7/22/200 4 4 0.70 2.22 0.05 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/18/200 4 46 1.10 0.59 0.08 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 8/27/200 4 4 0.70 1.29 0.07 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 9/27/200 4 12 0.70 1.07 0.06 COLAMOP 4 UDFCD_Modular_Porous_Pavement_94_to_04 39.88 33 105.20 00 COM 10/5/200 4 5 1.00 0.49 0.05 A.2 Calculations Assessment Matrix Steps and Calculations for Entire Lakewood Gulch Watershed Step 1: Set up matrix with selected constituent and various land cover categories. Seven constituents were used in this example as they correlate with the two land use water quality results for Denver, Colorado. These seven constituents were analyzed using event mean con centrations and can be found in Chapter 3. Land cover was evaluated using geospatial analysis methods to determine different categories associated with collected non point stormwater quality data. This process can be found in Chapter 4.

PAGE 113

102 Step 2: Evaluate variables with analyzed water quality data results and land cover percentages. As mentioned in Chapter 5, a 2 year, 1 hour rainfall depth was used over the watershed area. Step 3: Using Microsoft Excel, MATLAB or some other matrix analysis program, this matrix can be evaluated. Additionally, depending on size of the matrices, manual calculations can also be used. The results for the Lakewood Gulch watershed are shown below for a given 2 year, 1 hour storm event.