I-58_SandyCreekWatershedStudy_FinalReportDRAFT_20190520

Sandy Cr. Watershed Study Final Report
City of Durham, Stormwater & GIS Services, Water Quality Unit, Project #16-001
Prepared by Susan Gale, Christine Cailleret, Shannon Hauschild, and David Milkereit
Month DD, 2019
Table of contents
Introduction.................................................................................................................................................. 4
Background..............................................................................................................................................
4
Objectives.................................................................................................................................................
5
Study area and site descriptions..................................................................................................................
5
Overview of the Sandy Creek watershed.................................................................................................
5
Study area and monitoring sites..............................................................................................................
7
Methods.....................................................................................................................................................
10
Hydrology, channel morphology, and instream habitat........................................................................
10
Waterquality.........................................................................................................................................
13
Sedimentquality and toxicity................................................................................................................
17
Results........................................................................................................................................................
19
General conditions summary..................................................................... Error! Bookmark not defined.
Summaryof known illicit discharges......................................................................................................
19
Summary of monitoring events.............................................................................................................
20
Hydrology, channel morphology, and instream habitat........................................................................
21
Waterquality monitoring......................................................................................................................
29
Sedimentquality and toxicity.....................................................................................................................
45
Discussion.......................................................................................................Error! Bookmark not defined.
Recommendations??......................................................................................Error! Bookmark not defined.
References..................................................................................................................................................
48
Appendix 1. Soil moisture historic data summaries...................................................................................
49
Appendix 2. Summary of rain events.........................................................................................................
50
Appendix 3. Stream stage logger results....................................................................................................
52
Appendix 4. Ambient water quality results................................................................................................
53
Appendix 5. Baseflow water quality results...............................................................................................
54
Appendix 6. Storm sampling summaries....................................................................................................
56
Appendix 7. Storm water chemistry results...............................................................................................
60
Appendix 8. Specific conductance time series...........................................................................................
62
Appendix 9. Sandy Creek Watershed Existing Data Summary...................................................................
64
Tables
1

Table 1 Percentage of National Land Cover Database (NLCD) categories in the Sandy Creek watershed................7
Table 2 Descriptions of monitoring sites....................................................................................................................9
Table 3 Soil moisture criteria used to determine relative soil moisture levels (wet, moderate, and dry) ............. 14
Table4 Pesticides analyzed..................................................................................................................................... 16
Table 5 Chemical analyses for stream sediment..................................................................................................... 17
Table 6 Summary of data collections...................................................................................................................... 20
Table 7 Habitat assessment sub -scores and total score for each core site............................................................ 24
Table 8 Stream discharge measurements, means, and standard deviations (SD) at core study sites .................... 27
Table 9 Estimates of baseflow discharge (cfs) for each study site.......................................................................... 28
Table 10 Summary of blank QC samples with concentrations above reporting limit ............................................. 30
Table 11 Summary of storm events sampled.......................................................................................................... 32
Table 12 Completed storm events sampled by study site...................................................................................... 32
Table 13 Percentiles for specific conductance results for core monitoring sites .................................................... 44
Table 14 Pesticides and associated reporting limits for contract laboratory.......................................................... 45
Table 15 Results from chemical analyses of stream sediment, TECs, PECs, PEC-Q's, and Incidence of Toxicity.... 46
Table 16 Results from physical analyses of stream sediment................................................................................. 46
Table 17 Results from baseflow sampling, 2/23/2018............................................................................................ 54
Table 18 Results from baseflow sampling, 6/6/2018.............................................................................................. 55
Table 19 Water chemistry results for Storm Events 1 and 2................................................................................... 60
Table 20 Results from storm sampling equipment and field blanks....................................................................... 61
Figures
Figure 1 Overview of the New Hope Creek watershed..............................................................................................4
Figure 2 Sandy Creek watershed and major tributaries.............................................................................................4
Figure 3 Hydric and non-hydric NRCS soil mapping units for Sandy Creek watershed study area ............................6
Figure 4 National Land Cover Database 2011 (NLCD2011) land use within the Sandy Creek watershed study area.
.................................................................................................................................................................................... 6
Figure 5 Sandy Cr. study area and location of core, ambient, synoptic nutrient, and pesticide/herbicide
monitoringsites..........................................................................................................................................................8
Figure 6 Examples of normal (a) and deformed (b-f) menta ("teeth") of Chironomus midge larvae ..................... 19
Figure 7 Drought conditions during the study period............................................................................................. 21
Figure 8 Temperature and precipitation recorded in 2018 at Raleigh -Durham Airport compared to climate
normals. Obtained from www.weather.gov/rah/2018krdu, 4/9/2019.................................................................. 22
Figure 9 Daily precipitation at USGS rain gage at Maureen Joy Charter School ..................................................... 22
Figure 10 Average daily soil moisture at ECONET site DURH from 2/1/2018 — 12/31/2018.................................. 23
Figure 11 Stream cross sections from core monitoring sites.................................................................................. 26
Figure 12 Distributions of Relative Percent Differences (%) for duplicate samples by parameter. Results rejected
due to blank contamination were excluded from analysis..................................................................................... 31
Figure 13 Summary of stream stage (ft.), precipitation (in.), and subsamples composited for analysis for Storm
Events 1A & 113 (5/16/2018 - 5/18/2018)............................................................................................................... 34
2

Figure 14 Summary of stream stage (ft.), precipitation (in.), and subsamples composited for analysis for Storm
Event2(6/26/2018)................................................................................................................................................ 34
Figure 15 Nutrient concentrations for all sites for Storm 1 (Dry AMC) and Storm 2 (Moderate AMC) .................. 37
Figure 16 Alkalinity, cation, anion, and TSS concentrations for all sites for Storm 1 (Dry AMC) and Storm 2
(Moderate AMC)...................................................................................................................................................... 37
Figure 17 Metal concentrations for all sites for Storm 1 (Dry AMC) and Storm 2 (Moderate AMC) ...................... 38
Figure 18 Histograms of specific conductance results............................................................................................ 44
Figure 19 Results from sediment particle size analyses.......................................................................................... 47
Figure 20 Time series of soil moisture from NC ECONET site DURH, 2008-2017 (period of record) ...................... 49
Figure 21 Distributions of soil moisture (m3/m3) at ECONET site DURH, 2012 — 2017.......................................... 49
Abbreviations
3

Introduction
Background
The City of Durham Stormwater & GIS Services Division is developing watershed management implementation
plans (WIPs) for each of the major watersheds within the City's jurisdictional limits. The WIP for New Hope Creek
Watershed is currently in development. The goal of the WIP process is to identify strategies for improving water
quality within the watershed under study, and to propose specific projects for implementation (such as
stormwater control measures or other strategies) that will support the strategies identified. The Water Quality
Unit (WQU) completes watershed characterization studies to support the WIP efforts, which often include
hydrologic or water quality modeling efforts. Secondarily, these watershed studies allow wider monitoring (both
geographically and parametric coverage) to assist the WQU with better characterization of the City's aquatic
resources and identify areas of potential concern.
The larger New Hope Creek watershed (Figure 1) originates in neighboring Orange Co. Very little of the
mainstem of New Hope Creek is located within the jurisdictional boundaries of the City of Durham. There are
two major tributaries to New Hope Cr. with portions of their watersheds within the City boundary: Mud Cr. and
Sandy Cr. Very little of the watershed area of Mud Cr. is located within the City boundary and these areas are
fairly discontiguous. The Sandy Creek subwatershed is entirely located within the City's jurisdiction, and so was
selected for monitoring to support the New Hope Creek WIP, as it would likely provide better opportunities for
subwatershed-scale enhancements.
Watershed boundaries Study area °
Headwaters New Hope Creek + Streams E
Interstate Fork i' 70
US Highway e
INC Highway er Mountain 147
r
Ramp ' Creek -Little l`!I � 147 yyy
City Limits
- % River
County Bounda
Sevenmile Stony f Crooked
Creek -Eno Creek -Eno Creek -Eno
River River ; River 15
751
U
W rb
Cre,'k�� m
Cane
Creek
re'dk�
bird 1
Fork Creek
i
Collins Little 15
Creek Creek
University w
Lake �� �` 751
Morgan
- ------ ----Creek -i-- - --- --- J
files it
Figure 1 Overview of the New Hope Creek watershed Figure 2 Sandy Creek watershed and major tributaries
4

The monitoring data collected for each watershed characterization study is somewhat unique. This is due to
differences in watershed characteristics and potential water quality impacts, which in turn can result in different
modeling approaches used to identify the most promising projects for water quality improvement during the
WIP process. Data gaps within the Sandy Creek watershed were identified through a review of existing water
quality data that was completed prior to the design of the Sandy Creek Watershed Characterization study.
Findings were summarized in a prior document (COD SW, 2017).
Objectives
Monitoring for the Sandy Creek Watershed Study (project 16-001) was initiated by the WQU in early 2018.
Details on project objectives, study design, and schedule were previously described in the project's Quality
Assurance Project Plan (Durham SW, 2018b), but the three primary objectives identified for this study were:
1. Characterization of nutrient concentrations, loading, and sources.
2. Assessment of potential effects of instream chemical, physical, and hydrological conditions on aquatic
life uses.
3. Identify potential sources of pesticides within the Sandy Creek watershed that were previously identified
during US Geological Survey (USGS) monitoring conducted in 2014.
To meet these objectives, monitoring of multiple stream characteristics were conducted, including: hydrology,
channel morphology, instream habitat, water quality, sediment quality, and toxicity indicators of instream biota
(midge deformiteies). Data collections were conducted from February through December 2018. An interim
report completed in September 2018 (COD SW 2018a) summarized data collections from the first six months of
the project (February through July). This document expands on that report to include data collected during the
entire study.
Study area and site descriptions
Overview of the Sandy Creek watershed
The Sandy Creek watershed is entirely located within the jurisdictional boundary of the City of Durham, with its
headwaters located roughly along the Durham Freeway (NC 147). Much of the eastern boundary of the
watershed runs along Chapel Hill Rd. and Old Chapel Hill Rd. The western boundary follows portions of US
15/501 and Garrett Rd. The confluence of Sandy Cr. with New Hope Cr. is just southwest of Garrett Rd. The
stream network flows roughly SSW from its headwaters and drains to New Hope Creek, which then flows to
Jordan Lake. The Sandy Creek watershed is located in the Cape Fear River Basin and is part of the U.S. Geological
Survey (USGS) 12-digit Hydrologic Unit 030300020601 (Headwaters New Hope Creek). Sandy Cr. is the only
named waterbody, though there are two significant tributaries, referred to as "Tributary A" and "Tributary D" in
COD GIS data. Tributary A is actually somewhat of a misnomer, as it does not drain to Sandy Cr., but flows
through a wetland complex west of Garrett Rd. and then drains directly to New Hope Cr. Tributary D is the
largest tributary, originating in the northeast corner of the watershed and flows into Sandy Cr. near the
intersection of W. Cornwallis Rd. and US 15/501.
5

The N.C. Department of Environmental Quality (NCDEQ) includes only the mainstem of Sandy Cr. in its statewide
inventory of streams (index # 16-41-1-11) and it has been assigned the stream classification WS-V; NSW, which
protects it for water supply uses (WS-V) and requires additional management strategies due to its nutrient -
sensitive water (NSW) status. NCDEQ has not made any site -specific use attainment determinations for Sandy
Cr. since it is not monitored by the state. Sandy Cr. (like all waters of the state) is considered to be impaired due
to high mercury levels in fish tissue. Since the fish consumption advisory is a statewide impairment applying to
all waterbodies of NC, this impairment issue is outside of the scope of this watershed assessment or the WIP.
The entire watershed is located in the Triassic Basins (45g) Level 4 ecoregion, which is characterized by
landscape slopes that are often lower than surrounding ecoregions. Soils also tend to be clayey with low
permeability, which often leads to naturally depressed baseflows in streams (Griffith 2002).
According to the National Resources Conservation Services' (NRCS) Web Soil Survey, the predominant soil map
units by area within the watershed include: White Store sandy loam (43.6% of area, including all slope classes)
and Mayodan sandy loam (14.2%, including all slope classes). Urban -influenced soil map units (Iredell-Urban
land complex, Mayodan-Urban land complex, Urban land, and White Store -Urban land complex) make up
another 22.2% of the area. Hydric soils make up 12.7% of the area, though they are primarily located along
Sandy Cr. and its tributaries (Figure 3). Most of the non-riverine/non-pond National Wetland Inventory (NWI)
features that are present in the watershed are co -located with the areas of hydric soils in the lower end of the
study area, though NWI also shows a large (7 ac.) forested/shrub wetland feature along Sandy Cr. Tributary D
northeast of NC 751 (Academy Rd.). There are two isolated areas of the hydric soil map unit Iredell loam, 2-6%
slope at the interchange of NC 147 and US 15/501bypass and on the Duke University Hospital complex.
Figure 3 Hydric and non-hydric NRCS soil mapping units for Figure 4 National Land Cover Database 2011 (NLCD2011)
Sandy Creek watershed study area land use within the Sandy Creek watershed study area.
11

The most recent available NLCD data are from 2011, and show that the study area is predominantly Developed,
though it shows sizeable contiguous areas of Forest land classes as well (Figure 4). Percentages for individual
categories as well as aggregated groups are shown below. Duke University owns much of the land in the upper
portion of the watershed (approximately 20-25% of the total study area), including the academic campus,
hospital complex, and the Duke University Golf Club.
Table 1 Percentage of National Land Cover Database (NLCD) categories in the Sandy Creek watershed.
Code
Description
% area
11
Open Water
0%
21
Developed, Open Space
37%
19% Total all Developed
17% classes: 79%
5%
22
Developed, Low Intensity
23
Developed, Medium Intensity
24
Developed, High Intensity
41
Deciduous Forest
9%
Total all Forest
7% classes: 19%
2%
42
Evergreen Forest
43
Mixed Forest
52
Shrub/Scrub
0%
71
Grassland/Herbaceous
0%
81
1 Pasture/Hay
0%
90
Woody Wetlands
1%
According to the NC Division of Energy, Mines, and Land Resources (DEMLR), there are two facilities located
within the watershed boundary that have current NPDES stormwater permits:
Facility
Location
Permit number
Fleet Maintenance
1800 Camden Ave.
NCG080771
Couch Oil Company of Durham
2907 Hillsborough Rd.
NCG080865
City of Durham's Stormwater and GIS Services' spatial data were reviewed to determine potential impacts of
these facilities within the Sandy Creek watershed. The Fleet Maintenance site appears to drain across a natural
ridgeline into another watershed (Ellerbe Creek). The stormwater system at the Couch Oil Company site does
appear to drain to the Sandy Creek watershed, but is routed through a constructed wetland (the Duke University
Smith Warehouse/Maxwell St. Parking Lot, SCM Number 00537). No NPDES wastewater discharges were
identified using NCDEQ lists.
Study area and monitoring sites
Based on the review of existing data and field reconnaissance, the study area was limited to the watershed
upstream of where Sandy Cr. crosses Larchmont Rd. and where Tributary A crosses Martin Luther King Blvd.
(Figure 4). This captures the mainstem of Sandy Cr. and its two primary tributaries. A list of monitoring sites is
provided in Table 2 and shown in Figure 5. There are five basic types of sites, each differentiated by the types of
monitoring completed at each:
• Core (C): Baseflow water quality, cross sections, stream discharge, continuous monitoring of stage and
specific conductance, stormflow water quality, sediment quality, and instream habitat
7

• Ambient (A): Ambient water quality (pre -scheduled, sampled under ambient conditions)
• Pesticide/herbicide (P): Pesticide/herbicide water quality
• Midge deformity (M): Assessment of chronic toxicity of instream biota
• Synoptic nutrient (SN): Synoptic (one-time) sampling of nutrients
Figure 5 Sandy Cr. study area and location of core, ambient, synoptic nutrient, and pesticide/herbicide monitoring sites.
Location of USGS rain and stream gages also shown.
Ej

Table 2 Descriptions of monitoring sites.
Site
Site ID
Stream
Location
Latitude,
DM values
%
Comment
type
Longitude
Drainage
impervious
(dec degrees)
area (acres)
C, M
NH1.6SC
Sandy Cr.
Larchmont Rd.
35.9647,-78.9695
4,236
32.4
Outlet of Sandy Cr. study area
C, M
NH1.8SCTA
Tributary A
Martin Luther King Pkwy.
35.9614,-78.9608
748
54.1
Majority of watershed piped
M, A
NH3.3SC
Sandy Cr.
Cornwallis Rd.
35.9833,-78.9569
USGS stream gage
Upstream of confluence
P
NH3.4SC
Sandy Cr.
35.9843, - 78.9566
Below golf course
g
with Tributary D
Upstream of confluence
P
NH3.4SCTD
Tributary D
35.9836,-78.9560
Below golf course
with Sandy Cr.
C, M, P
NH4.4SCTD
Tributary D
Academy Rd.
35.9908,-78.9400
959
36.8
Headwaters
C, M, P
NH4.7SC
Sandy Cr.
Morreene Rd.
36.0035,-78.9522
563
49.8
Headwaters; Duke campus
P
NHS.OSCTD
Tributary D
Anderson Rd.
35.9941,-78.9324
Headwaters
SN
SCSN01
Sandy Creek
Waterbury St.
35.9684,-78.9731
64
30.4
SN
SCSN03
Sandy Creek
Welcome and Tryon St.
35.9863,-78.9653
51
22.0
SN
SCSN04
Sandy Creek
Evans St
35.9879,-78.9614
51
10.0
SN
SCSN06
Sandy Creek
Kangaroo Dr. post office
36.0149,-78.9504
54
54.3
SN
SCSN07
Sandy Creek
Duke Manor Apartments,
36.0148,-78.9438
LaSalle St.
70
59.4
SN
SCSN10
Tributary D
Circuit Dr.
36.0028,-78.9454
51
47.2
SN
SCSN11
Tributary D
Parking lot at end of
35.9995,-78.9505
78
35.8
Fuqua Dr.
SN
SCSN13
Tributary D
Nasher Museum parking
35.9995,-78.9273
lot off Campus Dr.
51
36.9
SN
SCSN14
Tributary D
Case and Hull St.
36.0029,-78.9242
54
52.5
SN
SCSN15
Tributary D
Campus Dr.
36.0026,-78.9177
53
39.3
SN
SCSN17
Tributary D
Parking lot off Morehead
35.9915,-78.9292
52
58.9
Dr.
SN
SCSN18
Tributary D
Brooks -Pascal and
35.9925,-78.9417
51
56.1
Cameron Dr.
SN
SCSN21
Tributary D
Near Prince and Pierce St
35.9855,-78.9357
82
28.8

Methods
This study collected information on water chemistry, sediment quality, sediment toxicity, aqueous pesticides,
stream discharge, stream stage, channel cross sections, and instream habitat assessments at selected sites on
Sandy Cr. and its major tributaries. Indicators of interest for water chemistry included field measurements (pH,
temperature, dissolved oxygen, specific conductance, turbidity), total suspended solids, total and dissolved
metals with an associated water quality standard or other widely accepted criteria, nutrients, dissolved organic
carbon, and a selection of pesticides (previously identified from a USGS study). Samples were also analyzed for
additional cations and anions to facilitate calculation of site -specific criteria for certain metals. Sediment quality
samples were analyzed for total metals, polynuclear aromatic hydrocarbons (PAH), TOC, and some physical
characteristics (particle size distribution, bulk density). Data loggers were installed at the four core monitoring
sites to record specific conductance at 15-minute intervals. These monitoring efforts were focused on a
relatively small number of locations within the watershed. In order to better characterize nutrient
concentrations throughout the watershed, synoptic sampling for nutrients occurred at 13 smaller headwater
catchments. Methods for all data collections are described below.
In general, data analyses were completed using Microsoft Excel 2010 and SAS JMP 13, unless otherwise noted in
the following sections.
Hydrology, channel morphology, and instream habitat
Precipitation, soil moisture, and drought status
Precipitation data were obtained from the USGS website (https://waterdata.usgs.gov/nwis) for the rain gage at
Maureen Joy Charter School (USGS site 355852078572045), which is located approximately 1000 ft. from the
USGS stream gage site on Sandy Cr. (Durham site NH3.3SC). For the period of 12/9/2018—12/11/2019,
precipitation data were downloaded in December but the USGS no longer provides these results on its website,
suggesting that there were quality concerns with the data. This was likely due to a mixed snow, sleet, and rain
event during this time. However, these results were used to characterize this storm event for interpretation of
stormflow sampling that occurred during this period, but are flagged as "Rejected" in original data files. During
storm sampling events, data from personal weather stations available through Weather Underground
(wunderground.com) were also reviewed to determine relative spatial coverage of rain events across the study
watershed. Annual climate summaries for the Raleigh -Durham area were obtained from the National Weather
Service website (https://www.weather.gov/rah/2018krdu).
Soil moisture data were obtained from the NC CRONOS website (http://www.nc-climate.ncsu.edu/) for the
ECONET weather station located at North Durham Water Reclamation Facility (site DURH). Daily results were
downloaded and stored locally on a regular basis, since online retrievals of daily data are limited to the last 180
days. Hourly data were required for storm events; these data were downloaded as soon as possible after storm
events since they are only available online for the previous seven days. More discussion of the use of soil
moisture data is provided in the Storm section of Water Quality Monitoring Methods.
10

Information on drought status was obtained from the NC Drought Management Advisory Council website
(www.ncdrought.org).
Stream stage
Staff gages and In Situ Level Troll 500 vented pressure transducers were installed at the four core sites
(NH1.8SCTA, NH1.6SC, NH4.4SCTD, and NH4.7SC) in mid -February. Level Troll loggers were programmed to
record pressure (PSI), temperature (°F), and stream stage (ft.) at 15 minute intervals. Loggers were downloaded
every 2-4 weeks using the In Situ Rugged Reader, and files were transferred to the WQU server using the
proprietary Win -Situ software. Loggers were also downloaded during retrieval of storm samples, since the stage
data were needed to determine which stormflow subsamples to process for analysis. Logger data were
warehoused in both the proprietary Win -Situ file format (wsl) and as comma -delimited (csv) files. Staff gage
readings were recorded during downloads and during site visits for other project tasks, such as chemical
monitoring. No attempt was made to install staff gages and loggers at the same elevation (i.e., bottom of the
staff gage even with the bottom of the logger), but both were installed in close proximity within each stream.
While stage measurements from each data source at a given site and time are different, it was expected that a
consistent and stable relationship would exist between the two measures.
During downloads, field staff recorded the staff gage height (ft.); current water level from bed to top of water
(ft.); current water level from the bottom of the logger to the top of water (ft.); current logger readings (time,
logger depth [ft.], pressure [PSI], temperature [°F], and battery status); and current condition of the logger.
Minor adjustments to the logger programs at certain sites were required in March to ensure that data were
being consistently recorded at all locations. At some sites, water level dropped below the bottom of the staff
gage during some site visits. In these cases, a pocket rod was used to record the elevation of the bottom of the
staff gage above the water, which would be recorded as a negative value; i.e., if the water was 0.25 ft. below the
bottom of the staff gage, it would be recorded as -0.25 ft.
At NH1.6SC, the channel shifted significantly within the first few months of the project, which resulted in a much
lower water level than anticipated. The logger had been installed in a small pool that was filled in when this shift
occurred. On 5/24/2018, the Level Troll logger was lowered by 0.79 ft. to ensure that it would stay submerged
during the lower water levels anticipated during the summer and fall. During preparation of results for this
report, all readings after the logger was lowered were corrected by subtracting 0.79 ft. from the raw stage
measurement to allow direct comparison of results collected during the entire study. The staff gage remained at
its original location and elevation to provide a consistent point of reference.
Cross sections
Establishment and surveys of stream cross sections followed general methods described in (Harrelson, 1994)
and described in the project QAPP (Durham SW, 2018b). The cross sections were established at the four core
sites (NH1.8SCTA, NH1.6SC, NH4.4SCTD, and NH4.7SC) on 2/23/2018, except for site NH1.8SCTA (Tributary A at
MLK Blvd), where the cross section was established on 5/1/2018. Cross sections were monumented by driving in
rebar on both sides of the stream, with sufficient set back from the top of bank to capture any changes to top -
of -bank elevation and floodplain deposits as well as to prevent the rebar from being washed away during high
flows. Rebar caps were installed, flagging hung near the pins, and the general locations were recorded in field
notes to facilitate finding the pins on future site visits. Ground level at the pin on stream left (on the left bank
11

when facing downstream) was established as the benchmark (0,0 origin) for each cross section. The only
exception to this process was at NH1.8SCTA, where a wing wall on the left side of the stream necessitated
establishing the benchmark pin several feet higher than the right bank. In this case, the right "pin" was actually
established as a tree and marked with spray paint.
Cross sections were completed during each of the four quarterly baseflow sampling events. A measuring tape
was strung as a tag line above the established pins and levelled by either using a line level or by taking elevation
measurements at the left and right edge of water. Beginning at the left pin, the value on the tape was read and
recorded as the first distance measurement. A measurement of the elevation of the tag line above the
benchmark (i.e., ground level at the pin) was then taken at this distance using a pocket rod or surveyor's rod.
These measurements (distance on tag line, elevation of tag line from ground) were repeated at each break in
slope between the pin and edge of water on both the left and right sides of the stream. Measurements were
taken at key features (pins, top of bank, bankfull elevation, and top of water) and so noted in the field notes.
Since discharge measurements were taken concurrently with cross sectional surveys, the measurements of
distance and water depth from discharge readings were used to develop the cross sectional profile within the
wetted channel. In other words, separate cross section measurements were not taken between the right- and
left -edge -of -water. Instead, the water depth at each point from the discharge measurement was added to the
elevation of the tag line at edge -of -water to get the total elevation of the tag line above the streambed within
the wetted channel.
Raw distance and elevation measurements were entered into Excel for data correction and graphing. The
measurements at the left pin were used to correct all distance and elevation measurements so that the cross-
section measurements always started with a distance and elevation of 0.0 ft. at the left pin. The corrected data
were graphed to create each stream's cross sectional profile for each survey.
Stream discharge
Stream discharge measurements were taken during each of the four quarterly baseflow sampling events at the
four core sites (NH1.8SCTA, NH1.6SC, NH4.4SCTD, and NH4.7SC), as described in the project QAPP (Durham SW,
2018b). The final data review in December indicated poor quality discharge measurements at the established
site. An additional discharge measurement was collected at NH1.8SCTA. In this case, the measurement was
taken downstream of the culvert going under Martin Luther King Boulevard (35.9611,-78.9604). The location
was just below a riffle where the stream transitions to a run. No additional tributaries or other significant flow
inputs were found between this location and the established site. The discharge measurements were taken at
changes in bed slope and at closer intervals where the majority of the flow occurred.
Discharge measurements were generally collected at the same location as the monumented cross section, if
flow conditions allowed. QC checks of the Sontek FlowTracker ADV meter included a Beam Check prior to each
day of monitoring and the automated QC check prior to beginning each discharge method. The 0.6 method was
used for measuring velocity, except for one occasion at one site (NH1.8SCTA), where deeper water and a
downstream culvert required use of measurements at two depths within the water column (0.2 and 0.8) for
some verticals, as recommended by the FlowTracker manufacturer (SonTek/YSI, 2009). Recommendations in the
WQU SOP for discharge measurements (COD SW, 2008) were initially followed, including the requirement for a
12

minimum of 20 verticals equally spaced across the wetted channel. However, these methods often resulted in
poor data quality. The evenly spaced intervals introduced a significant risk that a velocity measurement would
not be made in the thalweg, and so the area of the stream that carries the highest volume of water would not
be included in the total stream discharge calculations. Also, it was found that measurements near the thalweg
would be routinely flagged by the FlowTracker for containing too large of a proportion of the total stream
discharge, which increased the total measurement error. To address these issues, in certain cases, verticals were
spaced further apart in the areas of low velocity (such as near the banks) and more closely spaced in areas of
higher velocity (such as the thalweg), and verticals were also placed at significant changes in bed slope. In
several cases, fewer than 20 verticals were taken.
In certain cases, site conditions were not ideal for obtaining high quality discharge measurements. For example,
at NH1.8SCTA, the reach was essentially a plunge pool between two culverts and there was no area of laminar
flow. For the extra discharge reading taken in January 2019, high quality velocity readings were finally obtained
by measuring discharge downstream. In others, such as at NH1.6SC, water was often very shallow and the bed
consisted of unstable, unconsolidated sands, so boundary errors (due interference with the sensor beams) were
common.
Stream habitat assessments
Stream habitat assessments were completed once at the four core sites (NH1.8SCTA, NH1.6SC, NH4.4SCTD, and
NH4.7SC) using the NC Department of Environmental Quality's instream habitat form. This method is a visual,
qualitative assessment that provides sub -scores for specific instream habitat types (e.g., riffle/pool sequence,
substrate heterogeneity) and bank/near bank condition (e.g., bank stability, canopy cover). The total score
ranges from 0-100, with higher scores indicative of more diverse habitat, stable channels, and good riparian
cover.
Water quality
Illicit discharges
COD SW investigates reports of illicit discharges to the stormwater system and surface waters within the City
boundary, including sanitary sewer overflows (SSOs) from the City -owned system. Reports from 2018 were
reviewed to identify those incidents that occurred within the study watershed and the potentially affected sites.
This information is used to determine if they could have impacted data collections that occurred during the
study.
Baseflow sampling
Baseflow sampling occurred quarterly, concurrently with stream discharge and cross section monitoring.
Sampling only occurred if there had been no significant rain in the preceding three days and site conditions
indicated baseflow conditions. All scheduled water quality samples and in -situ field parameters were collected
at the four core sites (NH1.8SCTA, NH1.6SC, NH4.4SCTD, and NH4.7SC). Sampling was conducted as described in
the Sandy Creek Watershed Study QAPP (2018b), with all samples taken as grab samples and chemically
preserved during collection. Samples for dissolved fractions were filtered in the field prior to chemical
preservation using a peristaltic pump, new tubing, and a 0.45µm capsule filter. All samples were immediately
placed on ice after collection and held at <6°C until transferred to the labs for analysis. Field blanks and
13

duplicate samples were collected during each of the baseflow sampling events at one site. Duplicate samples
were averaged prior to data analysis.
Ambient sampling
Quarterly ambient sampling occurred at NH3.3SC (Sandy Cr. at Cornwallis) following the pre -determined
schedule for the COD SW Ambient Monitoring Program. Samples were therefore taken under a range of flow
conditions. All scheduled water quality samples and in -situ field parameters were collected. All samples were
taken as grab samples and chemically preserved during collection. Samples for dissolved fractions were filtered
in the field prior to chemical preservation using a peristaltic pump, new tubing, and a 0.451Lm capsule filter. All
samples were immediately placed on ice after collection and held at <6°C until transferred to the labs for
analysis. Duplicate samples were collected during the August sampling event, and were averaged prior to data
analysis.
Storm sampling
A total of four storm events were sampled in order to capture the desired range of antecedent soil moisture
conditions (AMC) ("Wet", "Moderate", and "Dry") at each of the four core sites (NH1.8SCTA, NH1.6SC,
NH4.4SCTD, and NH4.7SC). To determine typical soil moisture levels for the Durham area, daily average soil
moisture (m3/m3) records were downloaded from the State Climate Office of North Carolina's NC CRONOS
database (http://www.nc-climate.ncsu.edu/) for site DURH (North Durham Reclamation Facility) for the period
of record (Nov. 2008 — Dec. 2017). In reviewing a time series of average daily soil moisture (Appendix 1), it
appeared that a distinct shift occurred in annual distributions of soil moisture levels in mid-2011. This could be
attributed to drought conditions, as there was a continuous and extended period of abnormally dry and mild
drought conditions that lasted from late 2010 through 2012 (COD SW, 2017). Since that time, maximum soil
moisture levels never rebounded to their pre -drought levels and the annual range of values appeared to be
smaller. Maintenance records for this site were reviewed on the NC CRONOS web site and only standard
maintenance activities were documented, so there is no evidence that changes to instrumentation could be a
cause for this shift. Because of this shift, only results from Jan. 2012 — Dec. 2017 were used to calculate
summary statistics (Appendix 1) to help guide determination of Wet, Moderate, and Dry AMC. For the purposes
of this study, the target ranges for relative soil conditions were based on the 251h and 751h percentiles of the
average daily soil moisture records from 1/1/2012—12/31/2017, and are shown in Table 3.
Table 3 Soil moisture criteria used to determine relative soil moisture levels (wet, moderate, and dry).
Relative soil moisture conditions
Quantile of soil moisture data (2012 — 2017)
Absolute range
Wet
>75th percentile
> 0.42 m3/m3
Moderate
25th — 75th percentile
0.34 — 0.42 m3/m3
Dry
<25th percentile
<0.34 m3/m3
For collection of stormwater samples, automated ISCO Model 6700 and 6712 samplers were outfitted with 24
sample bottle cages and disposable 1L ProPak liners. New intake tubing was used for each sampling event. At
NH4.4SCTD, NH4.7SC, and NH1.6SC, the In Situ Level Trolls installed at the site were attached to the ISCO
sampler to provide stream stage information. At NH1.8SCTA, an older model 6700 ISCO was used that was not
compatible with the In Situ Level Trolls and so required use of the ISCO Model 720 module and submerged
probe to provide stream stage to the sampler. Autosamplers were programmed as described in the QAPP: a set
14

rise in stream stage (usually 0.25ft.) was used to initiate sampling, and subsamples were collected every 15
minutes until all of the available sample containers were full. The volume of each subsample varied depending
on the expected storm length and intensity, but ranged from 100 — 250mL. Multiple subsamples were collected
in each sample bottle. Wet ice was placed in each sampler during final programming before each storm event.
After each storm event, samplers were retrieved as soon as practical and safe and stage loggers were
downloaded. Samples were brought back to the COD SW lab for processing. Stream stage data were reviewed to
identify the timing of the rising limb, peak, and falling limb of the hydrograph to identify which samples should
be composited to represent the entire storm event for each site. Since each site reacted differently to a given
storm event, the number of subsamples required was not necessarily consistent across all sites. A minimum of
4L of sample was needed from each site to ensure sufficient volume for analysis of all parameters.
Once subsamples were identified for each site, they were combined in a clean Teflon -lined sample churn. Churns
were cleaned using a phosphate -free lab detergent, rinsed with tap water, and then copiously rinsed with
deionized (DI) water. The composite sample was churned slowly by one person while another used the spigot on
the churn to fill analytical sample bottles. Samples for dissolved metals and DOC were filtered using silicone
tubing, a peristaltic pump, and a 0.45µm capsule filter. All samples were held at <6°C until pick up by the
analytical lab's courier.
QC samples were taken during most storm sampling events: one set each of a duplicates, equipment blanks, and
field blanks. Duplicates required doubling the subsample volume during sampling to attain a minimum of 8L.
Equipment blanks were prepared in the lab prior to deploying ISCO samplers by rinsing and purging the ISCO
intake tubing three times with deionized water, then taking a series of manual 1L grab samples of DI until
sufficient sample volume was attained. Field blanks were prepared by pouring sufficient DI water into ProPak
liners in the ISCO and leaving them uncapped during the storm event. All QC samples were composited and
processed using the same methods as the environmental samples. Results from duplicate samples were
averaged prior to analysis.
Hardness -based metal criteria calculations
The NC DEQ has water quality standards for some metals monitored in this study, including Cd, Cr, Cu, Pb, Ni, Si,
and Zn. These are based on the water hardness and include separate values for both acute and chronic exposure
for dissolved fractions of each metal. Hardness levels from each site were used to calculate site -specific acute
and chronic values using an Excel spreadsheet provided by NC DEQ (https://deg.nc.gov/nc-stdstable-09222017).
These were used for comparison to concentrations of each metal collected under baseflow and stormflow
conditions. The baseflow concentrations were compared to the chronic criterion and stormflow concentrations
were compared to the acute criterion for each site.
Synoptic nutrients
A study plan for this project task was developed and finalized in June 2018 (COD SW, 2018d). The approach
taken was to delineate headwater catchments of approximately equal size for the Sandy Creek subwatershed
and sample at the outlet of approximately 12-15 of these catchments. A total of 16 sites were identified as
appropriate for sampling during field reconnaissance. Thirteen of these sites were visited in July. At each site,
field parameters (specific conductance, dissolved oxygen, pH, and temperature) were measured and grab
15

samples were collected for analysis of nutrients (NH3, NOx, TKN, and TP). Two field blanks were collected (one
by each sampling team) and one duplicate sample was collected. The remaining three sites were not sampled
due to lack of water in the channel.
In addition to the water quality work, spatial analyses were completed to delineate each catchment and to
characterize land use and total impervious within each watershed. The 20 ft. resolution legacy digital elevation
models (DEMs) were obtained from the NC Floodplain Mapping Program website
(https://sdd.nc.gov/sdd/DataDownload.aspx). The City of Durham vector data representing the stormwater
system and open channels (streams) were then burned into the DEM. This forced flow to be routed along the
actual surveyed channels when the DEM was processed using the ArcHydro tools in ESRI ArcMap to delineate
the catchments for each of the synoptic nutrient monitoring sites. [I CAN'T FIND HIS DESCRIPTION OF
CALCULATING LAND USE AND IMPERVIOUS — WILL NEED TO TOUCH BASE WITH HIM NEXT WEEK ON THIS.]
Conductivity loggers
Onset HOBO U24-001 conductivity loggers were deployed in wells at all core sites in mid -February. The wells
consisted of 2" PVC with 1" holes drilled at approximately 6" intervals to allow good mixing with surrounding
stream water. PVC was attached to fence or sign posts that had been driven into the stream bed, generally in a
pool or other area with lower water velocities. Loggers were suspended in the well using paracord tied to the
inside of the well cap. Loggers were programmed to measure and record uncorrected conductivity every 15
minutes. Sites were visited at approximately two -week intervals to download data and perform logger
maintenance. During site visits, a calibrated Oakton ECTester conductivity meter was used to measure
uncorrected conductivity (µS/cm) and temperature (°C) inside and outside of the well before removing the
logger. Data were downloaded using a HOBO waterproof shuttle. The relative level of fouling of the sensor was
recorded (scale of 1-5), the sensor was cleaned using a cotton swab and Liquinox detergent, and then
thoroughly rinsed. Initially, a second measurement of conductivity and temperature were made before
reinstalling the logger. However, it was found that this second measurement was always very close to the first
measurement (within ±6 µS/cm), so taking the second measurement was discontinued in May.
Table 4 Pesticides analyzed
2,4-D
Acephate
Aminomethylphosphonic
acid (AMPA)
Atrazine
Azoxystrobin
Carbaryl
Carbendazim
Dimethenamid-P
16

HOBOWare Pro software was used to transfer downloaded data from the waterproof
shuttle to the WQU server. The Conductivity Data Assistant in the HOBOWare Pro
software was used to convert the loggers' conductivity measurements to specific
conductance (µS/cm at 25°C), and to correct for sensor fouling and drift using the
field measurements taken at the beginning and end of each deployment period. In
some cases, the factory calibration was used for correction, but this generally
resulted in poor accuracy of the corrected data. Individual corrected files were saved
in the proprietary .hobo file format and also exported as comma -delimited (csv) files.
Pesticides
Sampling for pesticides was proposed for inclusion in this study based on data
collected by the USGS during April — June 2014. The USGS study found relatively high
Diuron
Fipronil
Glyphosate
Imidacloprid
MCPA
Metolachlor
Myclobutanil
Prometon
Propiconazole
Simazine
Sulfometuron-methyl
Tebuconazole
Tebuthiuron
Fri clopyr
levels of certain pesticides in three Durham streams, including at the USGS stream gage location on Sandy Cr.
(corresponding to COD site NH3.3SC). This site is located just downstream of the confluence of the mainstem of
Sandy Cr. and Tributary D, so four sampling locations were selected to try to identify potential geographical
areas and land uses that may be primary sources for these compounds.
Pesticides have been grouped into four major groups by the USGS: herbicides, insecticides, fungicides, and a
mixed group of fumigants, nematicides, and other miscellaneous pesticides (Gilliom 2006). This is an extremely
wide range of complex organic compounds that may require analysis for not only the parent compound but for
one or more degradation products in order to detect the presence of pesticide of interest. When detected,
concentrations also generally occur at extremely low levels. While the USGS has internally developed analytical
methods for detection of these compounds and their various degradation products at very low (ng/L) levels, few
(if any) laboratories have that capability. The USGS sampled for a total of 50 compounds (including both parent
pesticides and degradation products), but this study prioritized pesticides for analysis based on frequency of
detection by USGS, detection by USGS at concentrations above published toxicity limits, and analytical abilities
of the private contract laboratory to be used by the WQU. This resulted in a total of 22 pesticides being targeted
(Table 4). Sampling events were scheduled for a similar seasonal timeframe as the USGS study to enhance
comparability of results from the two efforts.
At each of the four sites monitored, grab samples were collected in amber glass containers and placed on ice.
Field measurements were also collected at each site. Samples were hand -delivered to the current contract
laboratory (Meritech, Inc.), who re -packed samples and shipped them overnight to their subcontract lab (Waters
Agricultural Laboratories) for analysis.
Sediment quality and toxicity
Sediment quality
Table 5 Chemical analyses for stream sediment
17

Sediment samples were collected for both physical and chemical
analysis following the methods outlined in "Standard Operation
Procedures for Sediment Quality Monitoring" (COD SW, 2014;
also see COD SW 2018b). Samples collected for chemical analysis
were collected using the Scoop Grob method and samples for
physical analysis were collected using the Shelby Tube Sediment
Collection method. At each of the four sample sites sediment
samples were collected at six different instream sediment
depositional areas that were combined in -field to create a single
composite sample for both physical and chemical analysis
samples at each sample site. Subsample volumes were
measured and recorded in the field prior to remove each from
the Shelby tube for compositing.
Samples were analyzed for the physical characteristics dry
weight and particle size distribution. Bulk density and porosity
were calculated by COD SW using the dry weight and the total
sample volumes measured in the field. Chemical analyses
included organic carbon, ten metals, and 11 polyaromatic
hydrocarbons (PAHs) (Table 5).
Final sediment chemistry data were analyzed in Excel 2010. For
Organic carbon
Metals
Al
Fe
Pb
Mn
As
Cd
Cr
Ni
Zn
Cu
PAHs
Anthracene
Benzo(a)pyrene
Chrysene
Fluoranthene
Naphthalene
Pyrene
Benzo(a)anthracene
Benzo(g,h,i)perylene
Dibenzo(a,h)anthracene
Indeno1(1,2,3-cd)pyrene
Phenanthrene
each parameter the Probable Effect Concentration Quotient (PEC-Q) was calculated by dividing the raw
analytical results by the corresponding PEC value as described in MacDonald, 2000. This was done in order to
normalize the results by accounting for mixtures within a specific sediment sample. Mean PEC-Q values were
then calculated for all parameters at each sample location. The mean PEC-Q value for each site was then used to
calculate the Incidence of toxicity (%) for each site using the equation also explained in MacDonald (2000).
Midge Deformity Analysis
Midge deformity sampling was performed by the contract biologist with assistance from one COD SW staff
member. Specimens of the midge genus Chironomus were collected by sampling submerged areas of silty
substrates with a D-frame aquatic sweep net. Samples were then field -picked to remove midge specimens from
detritus and transferred to vials of 70% ethanol for preservation until they were identified and analyzed for
deformities by the contract biologist. Each midge specimen was dissected and its head capsule slide mounted
for microscopic examination for visible deformities of the mentum ("teeth"). Each specimen with a deformity
was categorized as Class I, II, or III. Figure 6 shows an example of a normal mentum (1a) as well as common
deformities such as a chipped tooth (1b; a Class I deformity), missing tooth (1c,1e; Class II deformity), fused
teeth (1d; Class II deformity), and a combination of missing and fused (1f; Class III deformity) (from Eaton 2017).
This information was used to calculate the Toxicity Score using the following equation:
(Equation 1) S — (C1+2C2+3c3) X 100
n
where:
Im.

S = toxicity score
Cl, Cz, C3 = number of Class I, Class II, and Class III deformities, respectively
n = total number of larvae
The presence of environmental toxicity is suggested when at least 25% of specimens exhibit deformities and
there is a Toxicity Score (S) of at least 52.
Figure 6 Examples of normal (a) and deformed (b-f) menta ("teeth") of Chironomus midge larvae.
Photo source: Researchgate.net.
Results
Data collections occurred between 2/1/2018—1/9/2019. With the exception of pesticide sampling, all sampling
was completed as planned. Pesticide sampling was discontinued after the first sampling event since all results
were reported as non -detects.
Results in this section are grouped as follows:
• Known illicit discharges that occurred during study period
• Schedule of monitoring events
• Physical assessments (hydrology, channel morphology, and instream habitat)
• Water quality
• Sediment quality
Summary of known illicit discharges
A total of 16 illicit discharges were identified that occurred in the study watershed during the monitoring period.
Of these, four were small SSOs (<1,000 gallons each), three were illicit connections, two were private sewer
breaks/overflows, two were related to insufficient sediment and erosion control, and the remainder were
miscellaneous discharges (yard waste, commercial vehicle washing, etc.). However, this does not include
sediment and erosion control issues on larger properties, as these are under the jurisdiction of Durham County.
This caused issues at one monitoring location (NH1.8SCTA) on several occasions: extremely turbid water was
19

noted during dry weather conditions, and these were referred to the County. These conditions necessitated
rescheduling water chemistry sampling on at least one occasion.
[PHOTO OF NH1.8SCTA WITH AND WITHOUT S&E ISSUES WOULD BE HELPFUL]
Summary of monitoring events
The schedule of completed sampling events is summarized in Table 6. Data collections began with the
installation of data loggers for stream stage and SC in February 2018. Quarterly baseflow monitoring (water
chemistry, stream discharge, cross sections) occurred in February, June, August, and November. During the June
baseflow monitoring efforts, habitat assessments and pesticides sampling were also completed. Ambient water
quality sampling was performed in February, May, August, and November, on separate days from the baseflow
sampling events. Storm event water quality sampling was performed in May, June, September, and December.
Synoptic sampling for nutrients occurred at all sites on a single day in July. Midge deformity sampling was
attempted in early October, with the sediment quality samples taken a month later in November. Data loggers
continued to be deployed through December 2018 until all analytical results had been received from the lab and
reviewed and it was confirmed that scheduled data collections were complete.
All monitoring activities described in the original QAPP and associated task plans were completed, with the
exception of pesticides, collection of specimens for the midge deformity analysis, stream stage loggers, and SC
loggers. Pesticides were initially scheduled to occur three times, but the second and third sampling events were
cancelled after all analytical results from the first round of sampling were reported as non -detects. For the
midge deformity specimens, the contract biologist was unable to find enough specimens of the correct genus to
complete the analysis. For both types of data loggers, there were intermittent technical issues that resulted in
data gaps.
Table 6 Summary of data collections
L
+
to
W
H
to
bA
E
Q
�..1
C
O
C
M
CrCr
E
fa
2V
41
M
y
G1
Y
`A
3
c7
O
ba 'a
v E
wo a
> -a
�, M
41
v
-
3
R
a,
r
c r
C
�,,
c
w
-°1a
D o
m
tou
o
�
E
_o
o
r
M
U
C
3
O
W
W
to
MEA3
N
N
M
E
C
'a
Month
LA "a
Vf E
u -a
U
Vf
co
i
d
Vf
LA
C
G
Feb 2018
X
X
I X
X
X
X
X
Mar
X
Apr
X
X
X
May
X
X
X
X
X
Jun
X
X
X
X
X
X
X
X
X
Jul
X
X
X
X
Aug
X
X
X
I X
I X
X
X
Sep
X
X
X
X
Oct
X
X
X
X
Nov
X
X
X
X
X
X
X
20

Dec X X X X X
Jan 2019 X
Physical monitoring (hydrology, channel morphology, and instream habitat)
This section summarizes data and results collected for precipitation, soil moisture, drought status, stream stage
continuous monitoring, cross sections, stream discharge, and habitat assessments.
Precipitation, soil moisture, and drought status
Drought conditions in Durham County (Figure 7) during the first two months of the study period were rated as
Abnormally Dry (category DO), and were rated Normal for the remainder of the study. Drought conditions,
however, do not capture the extreme weather encountered in 2018: record amounts of rain were recorded in
the area during this year, with the annual total almost 15" above normal at Raleigh -Durham airport (Figure 8).
Rain events were frequent at the USGS rain gage at Maureen Joy Charter School between 2/1/2018 —
12/31/2018: 128 of 334 days had measureable rain events, which were at times heavy (Figure 9). Most notable
was Hurricane Florence, which affected the area from Sept. 12 —17 and over 9.5" of rain were recorded at the
USGS rain gage. The frequent rains also contributed to soil moisture levels remaining relatively high (above the
historic median) throughout much of the study period (Figure 10).
Figure 7 Drought conditions during the study period.
Graph downloaded 2/15/2019 from https://www.ncwater.org/Drought Monitoring/dmhistory/.
21

P,ALEIGH-DURHAM IHTL AIRPORT NC, NC - 2018
LL �
C �
n 33
2a
n
O
-1{1
CC W
�I
30
G
d
1$
Lys
V
'G 9
L�!
im
CA
so
743
60
93
43
33
2a
lfl
0
-1{I
Go
45
30
1$
9
]an Feb Mar Apr May ]un ]ul Aug Sep Oct Nov Dec
F&-:wG Min mcwd Max Hcrmal ___EU4x Hamel _ WUmYc Hanal
Figure 8 Temperature and precipitation recorded in 2018 at Raleigh -Durham Airport compared to
climate normals. Obtained from www.weather.gov/rah/2018krdu, 4/9/2019
USGS 355852D78572D45 RAINGAGE AT MAUREEN JOY CHARTER SCHOOL NR DURHAM
5.0
m
m
r
cs
4.0
m
0
+' 3.0
c
0
y 2.0
.H
a
ca
m
li 1.0
}
J
N
S
d
0.0 LL J� A I
Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2018 2018 2018 2018 2018 2018 2018 2018 2018 2018 2018
— Daily sun precipitation — Period of provisional data
Period of approved data
Figure 9 Daily precipitation at USGS rain gage at Maureen Joy Charter School.
Graph downloaded 2/15/2019 from https://waterdata.usgs.gov/nwis.
22

Figure 10 Average daily soil moisture at ECONET site DURH from 2/1/2018—12/31/2018.
Historic (2012 — 2017) 75th percentile (dashed line), median (dotted line), and 25th percentile (solid black line) shown for reference.
Habitat assessments
Habitat assessments at all four Sandy Creek core sites were completed on 6/6/2018 using the NC DEQ Instream
Habitat Assessment Form/Method (see COD SW, 2018b for details). Sub -scores and total scores are provided in
Table 7. Common issues across all sites were bottom substrate quality and bank stability. The Light Penetration
parameter scored quite high among all core sites, indicating that most sites had good vegetation structure
within their riparian zones. One exception was NH4.4SCTD, which was a relatively recently restored stream
reach, that had a well vegetated riparian areas but the vegetation was primarily very young woody plants,
invasives (such as Japanese privet), and herbaceous plants. This site was expected to score well overall due the
stream channel and floodplain restoration, but a significant factor here was embeddedness (heavy silt deposits
in riffles), which brought down the Bottom Substrate sub -score. The low habitat scores for NH1.8SCTA and
NH1.6SC suggest they are the most physically degraded sites and are the least able to support diverse instream
communities. However, the reach assessed at NH1.8SCTA is extremely short and bounded by culverts on the
upstream and downstream ends, so may not be representative of the daylighted portions of Tributary A
downstream of this monitoring site. Site NH4.7SC had the highest habitat score, in spite of its location in a more
densely developed area of the study watershed. Much of this may be due to its good riparian buffer, variety of
instream habitats, mix of substrate sizes, and presence of decent pool/riffle sequences. This reach appears to
have had native rock placed instream at the top of the reach at the outlet to a large culvert, presumably to
dissipate energy during high flow events. Though artificial, this has likely played a role in creating and
23

maintaining the pool/riffle sequence. The stream banks, though, were extremely incised with raw, actively
eroding banks, though this erosion also appears to be contributing some native rock to the bottom substrate.
Table 7 Habitat assessment sub -scores and total score for each core site.
Possible score for each category and total score are shown in parentheses.
C
a+
T G
O
41
+�+
= R
C
a t
a,
O
+1
> 4'
M
i
o
U
C w
M *-0M
R
i
=
� OJ
N
i
MV
N
C M
MO
L
CA O
a0+ �
a-' of
O O
w l0
Y>
C
r
+' N
L C O
•� 41
a C O
MO
a' O
Site ID
L
U 2 ••
N
i ..
O 3 `i
CO to ..
O
0.
CC .`1..
M C .1..
m M
a ei
J a.
O ri
d' N ..
O .`i..
H
NH1.6SC
4
8
3
4
3
8
10
6
46
NH4.7SC
5
19
8
10
14
5
10
8
79
NH1.8SCTA
2
6
11
4
0
8
8
5
44
NH4.4SCTD
4
16
2
10
16
8
7
10
73
Cross sections
The original study plan called for cross sections to be completed at all four core sites (NH1.8SCTA, NH1.6SC,
NH4.4SCTD, and NH4.7SC) during the first and last quarters of the study. However, within the first few months it
was noted that channels at some sites, particularly NH1.6SC, were extremely dynamic and the sandy streambeds
were significantly reworked during high flow events. It was decided that quarterly cross sections would help
better document changes in the channel. Therefore, four measurements were completed at each of the sites.
Cross sections were completed concurrently with the collection of other baseflow monitoring events, with the
exception of NH1.8SCTA. At this location, the first cross section measurement was not completed until May
2018. Results from each cross sectional measurement at each site are shown in Figure 11.
Site NH1.6SC appeared to have undergone the most physical change since the beginning of the project.
Variability in the stream channel was often observed in between field visits. The stream became quite
entrenched between the August and November measurements, with bed elevation dropping over 1.5 ft. in some
places. It is likely that the extremely high flows from Hurricane Florence, which passed through Durham in mid -
September, were responsible for these. It is likely that the stage logger readings will not be comparable
throughout the duration of this study at site NH1.6SC due to the variability and instability of the channel.
Site NH1.8SCTA showed more stability but also appeared to become slightly more incised between August and
November, particularly the stream channel on the right side of the bar. However, from May to August, the
stream channel seems to have been naturally aggrading and degrading. One complicating factor is the presence
of riprap on the channel bottom at this location. If it shifted or was moved during discharge measurements (for
example, to reduce eddy currents and beam interference for the FlowTracker), this would have affected the
elevation readings. There may be minor comparability issues in regard to the stream stage logger measurements
due to minor shifts in the channel at NH1.8SCTA.
Sites NH4.7SC and NH4.4SCTD underwent the least physical change throughout the project. All notable changes
at both of these sites occurred between August and November, again corresponding to Hurricane Florence.
24

NH4.7SC experienced some erosion on the right bank and became slightly incised. The stream channel at
NH4.4SCTD remained very stable throughout the duration of this project, and the only distinct notable change in
the cross section was due to sediment deposition on the right bank. This is the study reach that had previously
undergone restoration and so this deposition is a sign of its increased connection with its floodplain and
indicative of a higher functioning stream/floodplain system that seen at the other study sites.
25

0.00 5.00 10.00 15.00 20.00
Distance (ft)
25.00 30.00
35.00
40.00
45.00 50.00
-1.00
1.00
$
e
300
0
m
7
d
5.00
W
7.00
900 NH1.6S[
0.0 5.0 10.0 15.0 20.0
Distance (ft)
25.0 30.0
35.0
40.0
45.0 50.0
-1.00
1.00
$
0
m
3.00
5.00
a?i
W
7
9.ao
�NHIMLWAII
0.00 5.00 10.00 15.00 20.00
Distance (ft)
25.00 30.00
35.00
40.00
45.00 50.00
-1.00
_____
1.00
$
0
3.00
m
m
W
5.00
7.00
900
NH4.4SCTD
0.00 5.00 10.00 15.00 20.00
Distance (ft)
25.00 30.00
35.00
40.00
45.00 50.00
-1.00
1.00
$
3.00
0
m
5.00
a?i
W
7.00
9.aa
NH4.75C
Figure 11 Stream cross sections from core monitoring sites.
�2/23/2018
--p-5/1/2018
+6/6/2D18
-� 8/16/ 2018
--)(---11/29/2018
Average BankfuH
tStage Logger
26

Stream discharge
Stream discharge measurements were collected concurrently with other baseflow sampling. Four
measurements were made at three sites (NH1.6SC, NH4.4SCTD, and NH4.7SC). Five measurements were
collected at NH1.8SCTA, with the additional measurement occurring in January 2019, due to concerns over the
data quality from the first four sampling events (see Methods for details). Results are provided in Table 8.
Table 8 Stream discharge measurements, means, and standard deviations (SD) at core study sites.
All results were calculated by SonTek FlowTracker software except Staff Gage height, means, and standard deviations.
p
Staff Gage
Total
Average
Total
Discharge
Average
Wetted
Date
Height
Wetted
Velocity
Discharge
Uncertainty
Depth (ft)
Area (ft')
'^
(ft)
Width (ft)
(ft/s)
(cfs)
N
2/23/2018
0.91
24.0
0.9
21.78
0.0846
1.842
5.1
6/6/2018
-0.30
11.3
0.3
3.52
0.3720
1.309
6.4
u
8/16/2018
0.00
8.8
0.2
2.12
0.6113
1.293
10.1
Z
11/29/2018
-0.53
19.0
0.4
6.90
0.3645
2.514
11.7
Mean
0.02
15.8
0.5
8.6
0.3581
1.740
8.3
t SD
+ 0.55
t 6.1
+ 0.3
t 7.8
t 0.1865
t 0.499
+ 2.7
2/23/2018
0.81
17.3
1.4
24.36
0.0049
0.118
70.6
6/6/2018
0.80
14.2
0.7
9.81
0.0183
0.179
21.9
8/16/2018
0.78
14.0
0.8
11.21
-0.0019
-0.022
290.4
OR
11/29/2018
0.88
14.1
1.0
13.53
-0.0230
-0.311
34.7
Iq
Z
1/9/2019 a
0.86
8.8
0.3
2.88
0.2226
0.641
66.5
Mean
0.82
14.9
1.0
14.7
-0.0004
-0.009
104.4
+SDb
t0.04
+1.4
t0.3
+5.7
+0.0149
+0.189
t108.9
2/23/2018
0.96
13.0
1.0
12.60
0.0245
0.308
7.5
0
6/6/2018
0.95
14.6
0.9
13.50
0.0250
0.338
7.5
u
8/16/2018
1.04
13.2
1.1
13.95
0.0195
0.272
5.3
=
11/29/2018
1.02
13.0
1.1
14.50
0.0450
0.653
5.5
Z
Mean
0.99
13.5
1.0
13.6
0.0285
0.393
6.5
±SD
t0.04
t0.7
t0.1
t0.7
t0.0098
t0.152
t1.1
2/23/2018
0.43
13.9
1.1
14.87
0.0120
0.178
29.9
6/6/2018
0.42
15.5
1.2
17.97
0.0157
0.282
36.7
u
8/16/2018
0.35
15.0
1.0
15.67
0.0084
0.132
56.2
Z
11/29/2018
0.39
15.2
1.2
18.04
0.0406
0.733
27.0
Mean
0.40
14.9
1.1
16.6
0.0192
0.331
37.5
+SD
t0.03
t0.6
t0.1
t1.4
t0.0126
t0.238
t11.4
a January 2019 measurement at NH1.8SCTA was taken at a different location in the channel, so channel dimensions not
comparable to earlier results. This discharge measurement was of much higher quality and should be used as a
representative discharge for this site under baseflow conditions.
b Means and SDs shown for NH1.8SCTA are for 2018 results only.
27

Results from all four sampling events were very similar at NH4.4SCTD and NH4.7SC, based on the standard
deviations for each of the measurements. The standard deviations for stream discharge were relatively large
compared to the mean discharge measurements, though measured velocities and discharges were extremely
low. Results from the first four discharge measurements at NH1.8SCTA are highly suspect due to the challenging
conditions at this site for discharge measurements and the presence of negative discharge readings due to a
preponderance of eddy currents at this location. The final discharge measurement taken in January 2019 at a
reach downstream of the established cross section is believed to me more reliable and more representative of
baseflow discharge at this location. At NH1.6SC, the wide variability in channel morphometry over the study
period, previously discussed in the section on cross section results, is reflected in high standard deviations for
total wetted width, average depth, and wetted area.
Based on these results, the values below can be used as estimates of baseflow discharge for this study period.
Note that these values may not be appropriate for use during other years due to the extremely wet weather
prior to and during this study.
Table 9 Estimates of baseflow discharge (cfs) for each study site
Site ID
Estimated baseflow discharge (cfs)
NH1.6SC
1.740
NH1.8SCTA
0.641
NH4.4SCTD
0.393
N H4.7SC
0.331
Stream stage
Time series of stage data for the time period of 2/16/2018 — 1/9/2019 from In Situ Level Troll loggers and the
manual measurements from the staff gages are provided in Appendix 3. As noted previously in the Methods
section, the stage logger at NH1.6SC had to be lowered by 0.79 ft. in May due to shifts in the channel
morphometry and lower -than -anticipated baseflow levels. Logged stream stage levels were corrected prior to
analysis by subtracting 0.79 ft. to data collected after the adjustment to the logger height. At site NH1.8SCTA,
there was a significant data gap in May due to a programming error of the logger. In late September, logger
data at NH4.4SCTD began to become erratic, showing diurnal swings in stage. It was later determined that the
vented cable installed at this site had been flooded during Hurricane Florence, which caused corrosion or
blockage within the cable. This resulted in the inability of the logger to compensate for atmospheric pressure.
Therefore, all stage results collected at this site after 9/20/2018 were flagged as "rejected". Several instances of
possible outliers were also identified. These were times when there was an increase in stage recorded during dry
weather, which may indicate potential unknown discharges. Affected sites and dates include:
• NH1.6SC: 8/28/2018, 10/29—10/30/2018, 11/17—11/18/2018, and 1/1/2019.
• NH1.8SCTA: 9/25 — 9/26/2018, 10/31/2018, and 12/7 — 12/8/2018
• NH4.4SCTD:8/27/2018
a-]

Loggers at certain sites, particularly NH1.6SC, were often buried by streambed sediments to varying degrees
(from 1 cm deep to almost completely buried) due to shifting of the channel and depositional bars. The
manufacturer of the pressure transducers used asserts that burial in shallow sediments will not affect the
accuracy of the readings. This was confirmed through examinations of scatterplots and linear regressions (see
COD SW 2018a for results), which demonstrated that a relationship between staff gage height and
corresponding stream stage measured by the loggers remained constant at each site throughout the study
period.
Water quality
This section summarizes results from ambient, baseflow, stormflow, pesticide, and continuous specific
conductance water quality monitoring. Additional analyses were performed to compare metals concentrations
to site -specific criteria based on hardness.
QC samples and potential contamination issues
During 2018, special studies (including the Sandy Creek study) had reportable concentrations of analytes of
interest ("hits") in a relatively large number of QC blank samples. These hits primarily occurred in samples
analyzed by the private contract lab used by the Water Quality Unit, and rarely occurred in samples analyzed by
the City lab at the South Durham Water Reclamation Facility (SDWRF). This was particularly problematic for total
Kjeldahl nitrogen (TKN) and copper (Cu). A number of corrective actions were taken by WQU staff to try to
eliminate potential sources of contamination in the COD SW lab, during sampling activities, and for sample
handling, but hits in blanks continued to occur.
As part of the Sandy Creek study, inter -lab duplicate results for ammonia, TKN, and total phosphorus (TP) for
5/15/2018 samples were received from the private contract lab and SDWRF. While results from the two labs
were fairly similar for ammonia (0.15 and 0.16 mg/L) and TP (0.08 and 0.092 mg/L), results for TKN were very
different: SDWRF reported 0.49 mg/L and the contract lab reported 1.64 mg/L. This suggests that there may
potentially be significant variability due to lab -related factors, such as analytical methods, sample handling, or
cleanliness of sample containers. An analysis done on weekly TKN samples from another 2018 special study
(Falls Lake monitoring) that were also analyzed by the contract lab showed that results from TKN samples taken
on days with "dirty" blanks were statistically significantly higher than results from TKN samples taken on days
with "clean" blanks. The combination of evidence strongly suggests that the source of contamination was
related to the contract laboratory and whatever was contaminating the blanks also contaminating the
environmental samples.
When blanks are reported with measureable concentrations, COD SW SOP for data management (COD SW,
2018d) dictates that results from environmental samples from that day be assigned a qualifier code. Depending
on the concentrations reported in the blank, the results from environmental samples will either be given a code
of J7 ("Blank contamination evident, value may not accurate") or R ("Sample rejected due to blank
contamination exceeding value reported"). One or more blanks collected during the Sandy Creek Watershed
study were reported with concentrations above the reporting limit (RL) for DOC, TSS, Cu (total and dissolved), Fe
(dissolved), Mn (total), Na, NH3, NOx, and TKN (Table 10). In some cases, these were from equipment blanks
collected with automated storm sampling equipment, and low-level contamination can be expected when using
this type of equipment. Best professional judgment was used during data reviews to determine if environmental
29

samples associated with equipment blanks should be excluded. For all others, samples associated with blanks
with hits were usually rejected if the value in the samples was less than five times the amount in the blank.
These rejected results are shown in plots and summary tables in this report but are symbolized to indicate their
rejected status. All rejected results were excluded from statistical analyses. In some cases, such as TKN and Cu,
this resulted in the inability to make statistical comparisons due to small sample sizes of unrejected data.
Table 10 Summary of blank QC samples with concentrations above reporting limit.
Parameter
group
Parameter
Reporting
limit
Max value
for all blanks
# blanks
# blanks
with hits
% blanks
with hits
DOC (mg/L)
1
1.24
10
2
20%
Conventional
TSS (mg/L)
2.5
5.2
10
1
10%
Cu, diss (ug/L)
2
8.7
10
4
40%
Cu, total (ug/L)
2
9
10
6
60%
Metals
Fe, diss (ug/L)
25
68.9
10
1 1
10%
Mn, total (ug/L)
5
5.5
10
1
10%
Na, total (ug/L)
200
279
10
3
30%
NH3 (mg/L)
0.02
0.07
12
4
33%
Nutrients
NOx (mg/L)
0.1
7.4
12
2
17%
TKN (mg/L)
0.2
0.72
12
9
75%
Duplicate samples were also collected during Ambient, Baseflow, and Stormflow sampling. Results from
duplicate analyses (excluding rejected data) were used to calculate the relative percent difference (RPD), a
standard method for evaluating differences in duplicates. RPD is equal to the absolute value of the difference of
the two results divided by their mean. COD SW does not have defined acceptance or rejection criteria for
differences in duplicates, but these types of results do provide some rough guidelines when reviewing data. For
example, the grand mean RPD across all parameters was 7%, suggesting that variability from all sources was
relatively low overall. However, mean RPD by parameter ranged from <1% for certain metals to 60% for TKN.
Other parameters with mean RPDs above the grand mean for all parameters included Al (total), Cd (total), TSS,
NH3, TP, Cl, and Cu (total and dissolved).
30

1tSU`Yo
"
160%
•
VisitType
0 RPD distribution
• Ambient
Baseflow
140%
1 Storm
■ Synoptic Nutrients
120%
_ 100%
❑
� 80%
640% LLTT_M�'
i � � 1-1
4o°rb
A
r r+ i7 lip
R 7J V Q VI N Y N a N A N A V N N ice-. R ice-. N N y R N R N iR-. m x z
O o o a
a O r -o o 'a o o -o o -o v -o o v
❑ c _ _ z z r
E Q Q Q Q U U U U U U �? �i LL Y rn Z Z Z a N N
a =
C_
m
Y_
Q
Conventional Metals Nutrients
Parameter group / Parameter
Where(10 rows excluded)
Figure 12 Distributions of Relative Percent Differences (%) for duplicate samples by parameter. Results rejected due to blank
contamination were excluded from analysis.
Ambient samples
Ambient sampling was conducted in February, May, August, and November at NH3.3SC, which is part of the COD
SW's monthly ambient monitoring program. During these four months, regular ambient monitoring was
supplemented to include all parameters being sampled as part of this study's baseflow and storm monitoring.
With the exception of the May sampling, all sampling events occurred after significant rain in the previous 48
hours: 0.30" in February, 1.64" in August, and 3.28" in November (and raining the day of sampling).
Individual results for quarterly ambient monitoring at NH3.3SC are provided in Appendix 4. Results were
reported as non -detects for many total and dissolved metals (As, Cd, Cr, Ni, Pb) and for dissolved Zn. When
compared to historic ambient monitoring data, results were fairly comparable, with a few exceptions: TKN was
slightly elevated in August when compared to the historic median, and an historic low pH (5.7 SU) was recorded
during November.
Baseflow samples
Sampling was completed at all four core sites (NH1.6SC, NH1.8SCTA, NH4.4SCTD, and NH4.7SC) on 2/23/2018,
6/6/2018, 8/16/2018, and 11/29/2018. All results were reported as non -detects for As (total and dissolved), Cd
(total and dissolved), Cr (total and dissolved), Ni (total and dissolved), and Pb (dissolved). The majority of
samples were reported as non -detects for total Pb (81%) and dissolved Al (69%). Other parameters with at least
one result reported as a non -detect included total Zn (25%), dissolved Zn (38%), TP (25%), NH3 (13%), TSS (6%),
and NOx (6%). In addition, a relatively large number of results were rejected due to blank contamination for
total Cu (50%), TKN (50%), and NH3 (25%). Based on the non -detect and rejected results, the following
parameters were excluded from statistical analyses: total and dissolved As, total and dissolved Cd, total and
31

dissolved Cr, total and dissolved Ni, total and dissolved Pb, dissolved Al, total Cu, TKN, and NH3. For sites where
duplicate samples were collected, results were averaged prior to analysis.
Storm samples
A total of four storm events were sampled between May and December (Table 11, Table 12). The storm events
ranged widely in terms of duration (1.75 — 23.25 hrs.), total precipitation (1.21— 2.64 in.), and intensity (0.05 —
1.51 in./hr.). All three relative AMC conditions (Dry, Moderate, and Wet) were captured for NH1.6SC, NH4.7SC,
and NH1.8SCTA. An equipment failure occurred at NH4.4SCTD during the Moderate AMC storm event in June, so
the three storm events collected at this site represented only Dry and Wet conditions. The third storm event was
unusual, in that it included a period of rain, followed by a period of heavy snow, and then another period of rain.
Precipitation data were originally downloaded from the USGS website in December and are being used for
interpretation of this storm event. However, USGS has since removed these results from their website,
presumably for data quality issues. Because of the unusual nature of this storm event, the fourth sampling event
was collected, though this was also influenced by snowmelt and potentially any road treatments applied during
the prior snow event. This was also the only storm event where samples were successfully collected at all sites.
Subsamples to be used for compositing for each storm and site were selected based on stream stage data
collected by In Situ Level Loggers at each site. The first subsample was the first collected on the rising limb, and
the final sample was generally approximately halfway down the falling limb of the hydrograph, so the entire
sampling period was longer than the time of active precipitation. Details of the sampling regime for each site are
provided in Appendix 6.
Table 11 Summary of storm events sampled.
Storm
Event
Rain Start Time
Rain End Time
Event
duration
(hrs.)
Event
Cumulative
Precipitation
(in.)
Average
intensity
(in./hr.)
Soil moisture
prior to
storm
(m3/m3)
Relative soil
AMC
1A
5/16/2018 19:30
5/16/2018 22:45
3.25
1.33
0.41
0.30
Dry
16
5/17/2018 22:30
5/18/2018 12:15
1.75
2.64
1.51
0.30
Dry
2
6/26/201811:45
6/26/201816:30
4.75
1.66
0.35
0.38
Moderate
3
12/9/2018 15:15
12/10/2018 7:30
16.25
1.21
0.07
0.42
Wet
4
12/14/2018 09:15
12/15/2018 10:00
23,251
1.22
0.05
0.421
Wet
Table 12 Completed storm events sampled by study site.
Storm Event
NH1.6SC
Sandy Cr. at
Larchmont Rd.
NH4.7SC
Sandy Cr. at
Morreene Rd.
NH1.8SCTA
Tributary A at
MILK Jr. Pkwy.
NH4.4SCTD
Tributary D at
Academy Rd.
QC samples
1A
✓
✓
✓
✓
1B
✓
2
✓
✓
✓
✓
3
✓
✓
✓
✓
4
✓
✓
✓
✓
32

Samples were collected at all sites for the Dry AMC (Storm 1), though it was split over two consecutive days
(Storms 1A & 1B) due to equipment failure at one site (NH1.6SC) during Storm 1A. During the Moderate AMC
storm event (Storm 2), there was an equipment failure at NH4.4SCTD and so samples were only collected at
three of the four sites. Total rainfall and average intensity were fairly similar between storm 1A and storm 2, but
Storm 1B had a much higher total rainfall amount and much higher intensity. Storm hydrographs were fairly
typical of urban streams during an intense rain event (Figure 13, Figure 14), with a sharp increase on the rising
limb and longer, more gradual drops on the falling limb. Of interest were the peak stages at NH4.4SCTD, most
pronounced during Storm 113 (Figure 13). The hydrograph peak remained at a relatively consistent stage for
several hours before it declined sharply. This site is located near the downstream end of a stream and floodplain
restoration project, and it seems to access its relatively large floodplain during most rain events. As the stream
stage rises above its banks, it will dampen additional rise in stage, and sharp drops in stage would not occur until
the stream drops below the level of the top of bank.
Storms 3 and 4 occurred during Wet AMC conditions. These storms were much lower intensity in terms of
precipitation rates and occurred over a relatively long period of time. Storm 3 was certainly an atypical event for
storm water quality monitoring. Storm 3 (Figure 15) was actually a mixed precipitation event, beginning as rain
on 12/09/2018, changing to snow, and then back to rain on 12/10/2018. Some sleet and freezing rain also
occurred during this period. This was an historic snow event for the area, with 8.9 in. of snow recorded at
Raleigh -Durham International Airport over this two day period by the National Weather Service (NWS), and even
higher totals reported in Durham. An additional 1.75 in. of rain was also recorded by the NWS on 12/9/2018.
The mix of precipitation is likely why the USGS later removed these precipitation records from their website, and
the absolute precipitation values shown in the graphs should be interpreted with caution. Snow melt occurred
over the next few days, which caused slight increases in stream stage during the afternoons of 12/11-13/2018.
Stage returned to consistent levels each night. Precipitation for Storm 4 (Figure 16) began the morning of
12/14/2018 and continued into the following morning. This was another low intensity event, and was primarily
rain. The hydrograph response was quite muted as compared to Storms 1-2.
33

Storm 1 sam piing su m m a ry
a a o a a a a ao a a a o o a a a
o R 9 R �4Ny4 4 R
Date and time
—Stream stage (ft.) MPrecipitation (in.] Sample
Figure 13 Summary of stream stage (ft.), precipitation (in.), and sample times for Storm Events 1A & 1B (5/16/2018 - 5/18/2018).
z
x
z
x
z
2
A
J
Q a a a a a a
o a a a o a o
O O N N N N O
b b b b
o P P
Date and time
—Stream stage (ft.) Precipitation (in.) Sample
Figure 14 Summary of stream stage (ft.), precipitation (in.), and sample times for Storm Event 2 (6/26/2018).
No samples were collected at site NH4.4SCTD.
34

Storm 3 sampling summary
z
a o o a ao as o a a a o o a a o o a a ao 0 o a ao
N O O N N !V ry O O ry ry ry N O O ry N N ry O O N N
~ ~ ~ Date and time ~ ~ ~
—Streamstage[fi,] MPrecipitatian(in.) 91Sample
Figure 15 Summary of stream stage (ft.), precipitation (in.), and sample times for Storm Event 3 (12/9/2018—12/11/2018).
No samples were collected at NH1.6SC.
Storm 4 sam olina su m mars
-----�
'TT. YIT7TI1"rTT�'.
i i ITTT'
■-------
MEE
II
11
�
������
ruluu4uJuu4u�luuLutl.�uuui
11
��
03 7
tx+
O fi o,
12
03
.x-
0,6
� x
09�-
12 a n
03 y
Ofi N
n
0,9 v
12
03 z
x
0e
09
12
a a a a o o a o 0 o a a a a a a
��v ��y ry� day �Vv NV NV ry� ury1 .+ry1 .+ry1 �ry1 �ryv �Nv �Nv ��yy .ryL .eLy .eLy
Date and time
—Stream stage ft) Precipitation [in.] Sample
Figure 16 Summary of stream stage (ft.), precipitation (in.), and sample times for Storm Event 4 (12/14/2018—12/15/2018).
35

Box plots of results for each parameter grouped by AMC are shown in Figure 17, Figure 18, and Figure 19.
Results that were rejected due to contamination in associated blanks are shown as black markers but were
excluded from statistical analyses. Total and dissolved As, Cd, Cr, Ni, and Pb were also excluded due to large
percentages (>70%) of non -detects reported for each of these parameters. Because there were two sample sets
for most sites for Wet AMC, results were averaged for each site and parameter prior to analysis.
Wilcoxon tests were used to compare concentrations under different AMC conditions for each parameter. Very
few significant (p<0.05) differences were found, though sample sizes were quite small. In general, the Wet AMC
results showed the most differences as compared to other AMC conditions. Wet AMC had significantly higher Cl
and Na than Dry AMC, though the p-value from the comparison of Wet and Moderate was borderline
(p=0.0518). SO4 concentrations were also significantly higher for the Wet AMC than for Dry and Moderate. For
metals, the only significant difference was lower total Cu during Wet AMC than for Moderate, though there
were no data for Dry AMC. For nutrients, NH3 was significantly higher during Moderate AMC as compared to
Wet AMC, and again with no data available for Dry AMC. Wet AMC also had significantly lower concentrations
than Dry AMC, with borderline results (p=0.0518) from the comparison of Wet and Moderate.
Wilcoxon tests were also used to compare concentrations of each parameter across sites. No significant (p<0.05)
differences were found between sites for any parameter.
Many of the differences identified when comparing events by AMC, particularly Cl, Na, and SO4, are likely
attributable to road treatments used prior to the Wet AMC storm event. Therefore, with the exception of lower
total Cu, NH3, and TP concentrations during Wet AMC events, the concentrations from the whole -storm
composites were extremely similar when comparing across relative AMC and as well as when comparing across
sites. However, the use of a whole storm composite with fixed volume and fixed time interval sampling may not
provide the best estimate of overall storm conditions. Flow -weighted sampling, where sample volume and/or
frequency is tied to the stream discharge, would likely have provided a more representative sample, but these
types of methods are extremely difficult to apply in open channels with unstable cross sections. Weirs would
provide a suitable method for estimating discharge, but would have been difficult to install at these sites.
36

Dry AMA.
3.0
I
2,5 •
1.0
Offil
OA
AMC
Moderate AMA.
Wet AMC
J •
E
Station Name
■ N H 1, S SCTA
■ NH4.4SCTD
■ NH4.7SC
Figure 17 Nutrient concentrations for all sites by relative AMC.
Black symbols indicate results were rejected due to associated blank contamination and were excluded from statistical analyses.
4M
2M
100
70
sf5 ]Q
10
3
2
1
AMC
rl_. AAAY— AAa J--- Ar.Ar-
}},`
Station Name
i NH1,6SC
■ N H 1. S SCTA
■ NH4.4SCTD
■ NH4,7SC
Figure 18 Alkalinity, cation, anion, and TSS concentrations for all sites by relative AMC.
37

100000
e0000
200fl0
10000
1000
2x
100
10
i
100000
1L�1
3
ifs
100
60
20
10
6
2
100000
�6�Up7p0p0
30000
10000
6600
3007
u 1000
:3 App
� 205
100
60
i0
10
6
q
3
2
1
Cn
RS
Z
d
Wet AMC
H
k
k
C
QS Q7
QS
R1
¢S
Q7
Rf
Rf
¢S
Q7
45
45
¢S
m a
LAm
LA
m
m
L
a
`m
Figure 19 Metal concentrations for all sites by relative AMC.
As, Cd, Cr, Ni, and Pb not shown due to high percentages of non -detects.
Station Name
i NH1.6SC
A NHI.SSCTA
■ NH4.4SCTD
■ NH4.7SC
38

Comparisons to screening criteria
NC freshwater Class C quality standards and EPA Nationally Recommended Water Quality Criteria (NRWQC) for
protection of Aquatic Life applicable to parameters included in this study are shown in Table 13. These were
used for comparison to results from Ambient, Baseflow, and Storm sampling, with the exception of As, Cd, Cr,
Pb, and Ni, due to no samples being reported above the laboratory detection limit. The EPA NRWQC for Fe was
also not assessed, since EPA previously approved removal of this standard for NC.
Table 13 Applicable NC water quality standards and EPA Nationally Recommended Water Quality Criteria.
Parameter
Criteria
Criteria type
Comment
Alkalinity
20 mg/L
EPA NRWQC
Dissolved oxygen
4.0 mg/L (instantaneous)
5.0 mg/L (daily average)
NC Standard
pH
6.0 — 9.0 SU
NC Standard
Turbidity
50 NTU
NC Standard
Chloride
230 mg/L
NC Standard
As, dissolved
340 ug/L (acute)
150 ug/L (chronic)
NC Standard
Not assessed, all results non -detects
Cd, dissolved
Hardness -dependent
NC Standard
Not assessed, all results non -detects
Cr III, dissolved
Hardness -dependent
NC Standard
Not assessed, all results non -detects
Cr VI, dissolved
16 ug/L (acute)
11 ug/L (chronic)
NC Standard
Not assessed, all results non -detects
Cu, dissolved
Hardness -dependent
NC Standard
Fe
1000 ug/L
EPA NRWQC
Not assessed. EPA approved removal of NC
standard during 2007-2016 triennial review.
Pb, dissolved
Hardness -dependent
NC Standard
Not assessed, all results non -detects
Ni, dissolved
Hardness -dependent
NC Standard
Not assessed, all results non -detects
Zn, dissolved
Hardness -dependent
NC Standard
Acute and chronic hardness -dependent standards were only assessed for dissolved Cu and Zn, since all results
were reported as non -detects for As, Cd, Cr, Pb, and Ni. Median hardness values for each site were calculated
separately for baseflow and stormflow conditions, and these were used to calculate an acute and chronic
criterion for each site. These were then compared to individual results under stormflow (acute criterion) and
baseflow (chronic criterion) conditions to determine the percent of samples that exceeded each threshold
(Table 14 Summary of comparisons of dissolved Cu and Zn to hardness -based water quality standards). Acute
standards were much lower than those for chronic due to the lower hardness values seen under high flow
conditions. Even so, exceedance of the standard for Zn, either acute or chronic, was extremely rare. Cu also
rarely exceeded the chronic standard under baseflow conditions, but almost all samples exceeded the acute
standard under stormflow conditions.
39

Table 14 Summary of comparisons of dissolved Cu and Zn to hardness -based water quality standards
Baseflow (chronic criterion)
Stormflow (acute criterion)
Site
0
0
_
N M
_
N M
C O
J
C O-0
U
J
On
J
J
M U
t
._. -a
VI
41
v
VIbn
U1
M U
t V)
VI
VI
ro
f6 J
L
Q L
L
Q L
_
�
L
Q L
�-a
L
Q L
_ m
\
C
\
p
Cpp
C
C
C
.�
CJ
o
N
o
.�
U
o
N
(6
o
NH1.6SC
79.5
7.36
25%
97.3
0%
41.0
5.80
100%
55.1
0%
N H 1.8SCTA
143.5
12.19
0%
160.4
00o
25.5
3.71
100%
36.8
25%
NH4.4SCTD1
07.0
9.49
0%
125.1
0%
46.0
6.71
67%
60.7
0%
NH4.7SC
L 106.0
9.41
0%
124.1
0%
37.8
5.37
100%
51.3
0%
These results were also compared to the "default" values used by NC DEQ for attainment of water quality
standards when hardness data are unavailable. NC DEQ calculated summary statistics for their monitoring data
for each 8-digit Hydrologic Unit Code (HUC) in the state. The Sandy Creek watershed is part of the New Hope
Creek HUC (03030002), which is almost 1.1 million acres in size. The median hardness value is 43 mg/L, which is
quite low in comparison to the hardness results found at the Sandy Creek sites throughout the duration of this
study. The median hardness value (43 mg/L) may be representative of Sandy Creek during storm events, when
hardness values tend to be in a range from 20-40 mg/L. Based on the NCDWR hardness statistics, the maximum
hardness value was 180, which is closer in value to the median hardness values during baseflow. Although 180
mg/L is slightly higher than the hardness values that were found at the Sandy Creek core sites, it may be a closer
estimate when hardness values are unknown. However, using the maximum hardness value would lead to
higher calculated water quality standards. Either way, using the NC DEQ-recommended hardness statistics for
the Sandy Creek sites may not yield accurate values for hardness -based water quality standards. Given the wide
range of hardness values under different flow regimes, a "one size fits all" estimate is likely inappropriate as
well. These results unscore the importance of collecting hardness data and using site -specific data to calculate
associated water quality standards.
Synoptic nutrients
Field measurements and samples for nutrient analyses were collected at all sites on 7/18/2018. Results are
summarized in Table 14 and Figure 21. TP was relatively uniform across all sites. Reviewing nitrogen species for
each site (Figure 20) showed a very high NH3 concentration at SCSN17, located in the eastern portion of the
watershed, but this suggests an illicit discharge, such as sewage, rather than a watershed -based issue. High total
inorganic nitrogen (TIN) due predominantly to high NOx concentrations was seen at several locations, including
SCSN04, SCSN07, and SCSN18. These represent a range of land uses: R20 (high density housing) at SCSN04,
Freeway (draining NC 147) at SCSN07, and University (Duke campus) at SCSN18. The last catchment had very
high NOx (2.74 mg/L), and contains Duke's Wallace Wade Stadium and the intensively managed football field.
Turf management often requires application of nitrogen fertilizers (nitrate- and ammonium -based), so these
practices may be a potential NOx source in catchment SCSN18. SCSN10 and SCSN11 adjoin SCSN18 and also
contain athletic fields associated with Duke University, but NOx levels were much lower. Additional work would
be required to determine if these differences suggest that there is another source of NOx within catchment
40

SCSN18, or differences in NOx levels may be due to differences in turf management practices or the design or
extent of athletic fields in each.
5.0
4.5 `TON (mg/L)
4.0 ■ NOx (mg/L)
3.5 NH3 (mg/L)
3.0
2.5
2.0
1.5
1.0 -
0.5
0.0
00 01 y0
'R\
Figure 20 Inorganic and organic nitrogen at synoptic nutrient sites.
Table 15 Results from synoptic nutrient sampling.
Site
DO
(mg/L)
DO
sat
(%)
pH
(SU)
SC
(US/Cm
at
25°C)
Turbidity
(NTU)
Water
temp
(°C)
NH3
(mg/L)
NOx
(mg/L)
TKN
(mg/L)
TIN
(mg/L)
TON
(mg/L)
TN
(mg/L)
TP
(mg/L)
SCSN01
2.8
34
5.8
128
22.3
24.2
0.02
0.61
1.00
0.63
0.98
1.61
0.07
SCSN03
2.7
32
6.6
239
49.0
24.9
0.02
0.10
1.30
0.12
1.28
1.40
0.21
SCSN04
2.5
27
5.3
207
14.1
20.7
0.08
1.21
0.56
1.29
0.48
1.77
0.02
SCSN06
2.2
26
7.2
625
23.5
23.9
0.04
0.10
0.56
0.14
0.52
0.66
0.22
SCSN07
6.1
72
7.6
496
12.4
23.3
0.02
1.46
0.75
1.48
0.73
2.21
0.10
SCSN10
7.8
86
7.6
222
1.1
20.4
0.05
0.51
0.55
0.56
0.50
1.06
0.05
SCSN11
6.8
79
7.7
388
3.4
22.9
0.02
0.21
0.69
0.23
0.67
0.90
0.04
SCSN13
4.6
53
6.7
267
5.9
22.3
0.04
0.32
0.92
0.36
0.88
1.24
0.06
SCSN14
7.8
90
7.4
381
4.6
21.5
0.03
0.92
0.75
0.95
0.72
1.67
0.36
SCSN15
5.5
65
7.5
613
3.0
23.0
0.02
0.59
0.63
0.61
0.61
1.22
0.07
SCSN17
1.7
19
7.1
635
6.8
23.1
3.40
0.18
4.26
3.58
0.86
4.44
0.33
SCSN18
7.0
81
7.0
458
23.0
22.6
0.02
2.74
1.48
2.76
1.46
4.22
0.17
SCSN21
3.1
37
6.6
246
5.1
23.2
0.09
0.10
1.24
0.19
1.15
1.34
0.16
41

TKN • SCSNS Sites 'k Synoptic Study
Sandy Cree
• 0.55-0.63 Catchment Larl �j .
0.63 - 2.50 1 Freeway •a La b Res u it
-
2.50 - 4.26 Low Density Residential a+�
NOx Open Space
0.10 - 0.61 b R20 r O 7
0.61- 1.46
` University •Q -
NH3 Urban-Commerclal
0 0.02 - 0.05 LVW Sandy Creek
O 0.05 - 1.5 Sandy Creek Trib D
O 1.50 - 3.40
TP
0 0.024 - 0.047
O 0.047-0.215 'I
O 14 \
0.215 - 0.357 0 •
I
13
•
147
1
o �
�
r. O
O o
0
O 4 O
3 • o
• O s.
21 O
I
1f
I
O
I
I O �rll
• 161501
1
}
2,500 1:250 0 2,500 Feet
Site
Number NH3
1 0.02
NH3qual TKN
1
NOx
Nxqual O
TP
0.61
0.065
3
0.02
<
1.3
0.1
<
0.213
4
0.08
0.56
1.21
0.024
6
0.04
O.Sfi
0.1
<
0.215
7
0.02
<
0.75
1.46
0.096
10
0.05
0.55
0.51
0.047
11
0.02
<
0.69
0.21
0.04
13
0.04
0.99
0.31
0.058
13
0.04
0.85
0.32
0.063
14
0.03
0.75
0.92
0.357
15
0.02
<
0.63
0.59
0.071
17
3.4
4.26
0.18
0.334
18
0.02
1.48
2.74
0.169
21
0.09
1.24
0.1
<
0.162
Figure 21 Nutrient concentrations and land use at synoptic nutrient sites.
42

Specific conductance
Specific conductance (µS/cm at 25°C) time series for the first half of this study are provided in Appendix 8. Some
readings were flagged as "rejected" due to documented logger issues. For example, the logger at NH1.6SC was
not submerged for extended periods earlier in the study due to shifting of the channel. The installation of this
logger was modified in late May to better ensure that the sensor would remain submerged even at low stream
stages. These rejected results were excluded from analyses. Other individual readings were flagged as "outliers"
For example, NH4.7SC had two significant spikes in specific conductance that did not appear to be related to
other factors, such as precipitation (or lack thereof), and may be related to illicit discharges. The outliers were
included in analyses.
Distributions are described below by histograms (Figure 18) and percentiles (Table 13). Student t- and Tukey
HSD tests indicated that each site is significantly different (p<0.05) from all others, i.e., each site has a unique
distribution for specific conductance. The highest medians and maximum values were seen at NH1.8SCTA and
NH4.7SC. A large number of observations were identified as potential outliers at these locations, suggesting that
these sites are possibly affected by illicit discharges.
■■■■■■■I
■10■1■I■1
■
■■■■■■■I
■10■1■I■11■
w
111E1lMlE11M
■■■■■■■I
0■1■I■11■
or-7
■■■■■■■I
0■1■I■11■
M ■
■■■■■■■I
■ram
■■■■■■■I
"■■I■I■■l■
1I■I■11■
MIMEEM
■■
!1111■
1■■■■■■■I
■■
NNE
off
43

;
'
• "
"
; ;
1■■I■I■■1■
NINE■1■■I■I■■m
Wa
1■■1■1■■1■
�
■1■■I■I■■O
■1■■I■I■■�
W
E a
1■■I■I■■1■
NINE`
1■■1■1■■1■
�
■1■■I■I■■N
= W
1101■I■■I■
�
■1■■I■1001
WM
1■■1■1■■1■
�
■1■■I■I■■N
ELAN
1■■1■1■■1■
�
■1■■I■I■■N
W.71
1■■I■I■■1■
NINE;
■1■■I■I■■�
i 71
1■■1■1■■1■
�
■1■■I■I■■N
Wm
1■■I■I■■1■
NINE
■1■■I■1001
W =
1■■I■I■■1■
MEN!■1■■I■I■■�
i W1■■I■I■■1■
■1■■I■I■■�
!J
1■■1■1■■1■
NINE
■1■■1■1■■m
■■
JI��I�I�no
1JI■I■■1
.a■1■I■■1
��'■■�
'■I■■�
HE
'�1
■■
■■1■1■■1■
.MI■I■■1■
"11■■1■
'M"1
.=1■
■■
■■
XWE
1153■
L'L��
■■
llrM-7
WMEM
■■
■■
i�UEEIEIEE1E
��
MEIN
Figure 22 Histograms of specific conductance results.
Table 16 Percentiles for specific conductance results for core monitoring sites.
site
N Rows
Percentiles (u5/cm at 25*Q
Min
loth
25th
50th
75th
goth
Max
N H 1.6SC
9808
33
141
193
256
421
506
698
N H 1.8SCTA
14442
16
217
362
492
578
681
2040
NH4.4SCTD
14205
34
186
255
334
450
548
725
NH4.7SC
14382
27
217
332
422
570
649
2173
Pesticides
Samples were collected at sites NH3.4SC, NH3.4SCTD, NH4.7SC, and NH4.4SCTD on 6/6/2018. There was an
expectation that this first round of sampling would coincide with the widest and most intense use of these
pesticides. All results, however, were reported as below laboratory detection limits.
A significant issue with monitoring these types of compounds is that effect levels (such as concentrations which
may cause toxicity to instream biota) are often measured at the ng/L range. For example, Imidacloprid, one of
the most widely used neonicotinoid in urban areas, has a reported chronic toxicity threshold of 10ng/L (COD SW,
2018e). The USGS laboratories have developed specialized analytical methods that are able to attain these levels
of detection with reasonable certainty. However, very few (if any) commercial laboratories can realistically
obtain such low detection levels and will generally report results in the ug/L range. Results for some compounds
in the USGS study were high enough, though, that similar levels could potentially be detected even when using
the higher detection levels of commercial laboratories. The commercial laboratory used for this study had
reporting limits below the USGS maximum values for three compounds (2,4-D; Prometon; and Triclopyr). For the
remaining 19 compounds, the commercial lab reporting limit is 2 — 300 times higher than the maximum value
found by USGS. This includes Imidacloprid. The contract lab was able to provide a reporting level of 1,OOOng/L,
44

but this compound has a chronic toxicity threshold of 10ng/L and the highest concentration found by the USGS
was 97.2ng/L.
Since we were unable to identify detectable levels of any of the pesticides, further refinement of potential
source areas for these compounds was not possible. Our results were also unable to confirm the presence of
relatively high concentrations of 2,4-D, Prometon, and Triclopyr that were seen in the USGS study. This may be
due to those compounds not being present or due to different sampling, handling, and analysis methods used in
this study.
Table 17 Pesticides and associated reporting limits for contract laboratory.
All WQU samples were reported as non -detect (below reporting limit) for all parameters.
Compound
Contract laboratory
reporting limit (ng/L)
USGS maximum reported
value (ng/L)
2,4-D
2,500
3,290
Acephate
10,000
52.6
Aminomethylphosphonic acid (AMPA)
10,000
610
Atrazine
250
116
Azoxystrobin
250
10
Carbaryl
2,000
17.4
Carbendazim
5,000
135
Dimethenamid-P
250
53.5
Diuron
500
37.5
Fipronil
2,000
6.54
Glyphosate
10,000
2,800
Imidacloprid
1,000
491
MCPA
2,500
147
Metolachlor
2,000
55.2
Myclobutanil
500
9.21
Prometon
250
1640
Propiconazole
750
13.8
Simazine
500
52.2
Sulfometuron-methyl
500
115
Tebuconazole
1,000
25.1
Tebuthiuron
250
14.1
Triclopyr
250
440
Sediment quality and toxicity
Physical analyses
[Bulk density, porosity, particle size distribution]
Sediment quality and potential toxicity
Sediment samples were collected at the four core sites (NH1.6SC, NH1.8SCTA, NH4.4SCTD, and NH4.7SC) on
12/7/2018. All individual results are provided in Appendix 9, along with the corresponding TEC and PEC values
for each parameter. Table 15 summarizes the calculated PEC quotient (PEC-Q) and the Incidence of Toxicity (%)
for each site.
45

All results for metals and PAHs from all sites were below the associated TEC and PEC, suggesting little risk of
toxicity due to the individual constituents of stream sediments. The Incidence of Toxicity describes the
percentage of samples that would be likely to cause toxicity in one or more aquatic test species. The calculated
Incidences of Toxicity for each site ranged from 1.4 - 3.5%, which suggests that potential for toxicity from the
combination of metals and PAHs is also very low. All sites were also below the 4.9% Incidence of Toxicity
calculated from the Citywide mean values for metals and PAHs from prior studies, which suggests that the Sandy
Creek watershed should be a low -priority watershed for further Citywide studies of toxicity in stream sediments,
and toxicity due to sediment chemistry is less likely to be a significant cause of impacts to benthic communities.
Comparisons of results from Sandy Creek sites to historic sediment quality data for PAHs was not possible due to
the higher reporting limits used by the contract lab in this current study. All mean values for individual PAHs for
historic data were between 0.024 - 0.072 mg/kg, but the reporting limit for individual PAHs for the current
study was 0.334 mg/kg. So, even though non -detects were reported for all PAHs for sites NH1.6SC and
NH4.4SCTD, and for almost all PAHs for NH1.8SCTA, it is difficult to determine how they actually compare to
data previously collected by COD SW in prior studies. However, there were three compounds found at
NH1.8SCTA that were orders of magnitude higher than the historic mean: Fluoranthene, Pyrene, and
Phenanthrene. At NH4.7SC, almost all PAHs (excluding Naphthalene and Dibenzo(a,h)anthracene) were well
above the historic mean values, suggesting that this area is a "hotspot" relative to other stream sampling done
elsewhere in the City, though as noted above, the levels represent a low risk for toxicity.
For metals, comparisons to historic means could not be made for Cd due to the higher reporting limits provided
by the contract lab in this study. The majority of other metals were below historic means, with the exception of
Zn at NH1.6SC, NH4.4SCTD, and NH4.7SC, and Pb at NH4.4SCTD and NH4.7SC.
Results from physical analyses of sediments are shown in Table 16 and Figure 19. The two headwater sites
(NH4.4SCTD and NH4.7SC) had a more even mix of sediment sizes with slightly D50's corresponding to fine -
medium sand. NH1.8SCTA and NH1.6SC were much more predominantly medium sand and coarser, with D50's
equivalent to coarse sand.
Table 18 Mean PEC-Q and Incidence of Toxicity for sediment samples.
Citywide means
Parameter
NH1.8SCTA
NH1.6SC
NH4.4SCTD
NH4.7SC
(COD SW data)
Mean PEC-Q
0.013
0.018
0.026
0.034
0.048
Incidence of
1.4
1.9
2.6
3.5
4.9
Toxicity (%)
Table 19 Results from physical analyses of stream sediment
Parameter T NH1.8SCTA I NH1.6SC
NH4.4SCTD
I NH4.7SC
Particle size analysis
D50 (mm)
0.6
0.7
0.2
0.4
Silt/clay (%)
1.1
0.4
19.8
7.9
Very fine sand (%)
2.1
0.4
16.4
3.8
Fine sand (%)
11.8
2.6
21.5
13.0
46

Medium sand (%)
21.5
12.5
23.5
30.7
Coarse sand (%)
40.8
48.1
14.1
29.5
Very coarse sand (%)
20.2
33.9
3.6
10.5
Very fine gravel (%)
2.5
2.1
1.1
4.6
Bulk Density (g/cm3)
1.81
2.32
1.37
1.14
Porosity
0.32
0.12
0.48
0.57
Sample volume (cm3)
2,221
2,995
2,022
2,638
Dry weight of sample (g)
4,019
6,936
2,763
3,020
Total Solids (%)
74.6
64.6
57.3
68.3
Moisture % of Sample (%)
25.4
35.4
42.7
31.7
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1.0
0.9
0.8
■ Silt/clay (%)
0.7
■ Very fine sand (%)
0.6
■ Fire sand (%)
0.5
❑ Medium sand (%)
0.4
❑ Coarse sand (%)
0.3
❑ Very coarse sand (%)
0.2
■ Very fine gravel (%)
0.1
♦ D50 (mm)
0.0
NH1.8SCTA NH1.6SC NH4.4SCTD NH4.7SC
Figure 23 Results from sediment particle size analyses
Midge Deformity Analysis
Midge samples were collected on 10/2/2018. Out of the five proposed sampling locations midges were only
found at 3 locations: NH4.4SCTD, NH4.7SC, and NH3.3SC. None were found at NH1.6SC or NH1.8SCTA. At both
NH4.4SCTD and NH4.7SC, only one midge was found at each site and only 12 midges were found at NH3.3SC.
The lack of midges was likely due to the absence of silty habitat that was scoured due to the historic high flows
that occurred due to Hurricane Florence that occurred on 9/17/2018. The contract biologist indicated that there
may have been seasonal issues as well. The sampling was scheduled after many larvae would have emerged
from the streams as adults, and he recommended that future data collections occur in the spring or early
summer.
Midges from sites NH4.4SCTD and NH4.7SC were examined and not found to have any deformities. Additional
interpretation, calculation of deformity rate, and calculation of Toxicity Score were not performed since only
one specimen was available at each site. The number of midges (12) found at NH3.3SC was below the ideal
amount (minimum of 20) to complete a statistically significant toxicity test; however, the specimens collected
were analyzed by the contract biologist to give an estimated Toxicity Score. Two specimens found at NH3.3SC
had abnormalities. One had a worn trifid mental tooth, which the contract biologist did not feel was a deformity
attributable to toxicity. The second abnormal specimen had a missing lateral tooth, which was determined to be
a Class II deformity. Based on these findings, the sample was calculated to have an 8% deformity rate and a toxic
score of 1.7, which is well below the toxic score range of 20-40 needed to identify a stream as toxic.
47

References
City of Durham, Stormwater & GIS Services (COD SW). 2008. Procedure for Determining Wadeable Stream
Discharge with Hand -Held Current Meters Standard Operating Procedures.
COD SW. 2014. Standard Operating Procedures for Sediment Quality Monitoring.
COD SW. 2017. Sandy Cr. Summary of Existing Data — DRAFT. Available internally at F:\SW\Division Files\Water
Quality\Projects\16-001 Sandy Creek Watershed Stu dy\Documents\QAPP.
COD SW. 2018a. Sandy Creek Watershed Study Interim Report. Durham, NC. Water Quality Unit.
COD SW. 2018b. Sandy Creek Watershed Study (Project #16-001) Quality Assurance Project Plan (QAPP). Water
Quality Unit. Durham, NC. Water Quality Unit.
COD SW. 2018c. Sediment Data Study: Summary of Existing Sediment Chemistry Data in Durham County, NC and
Surrounding Counties. Durham, NC. Water Quality Unit.
COD SW. 2018d. Standard Operating Procedures for Use of the Water Quality Web Portal. Durham, NC. Water
Quality Unit.
COD SW. 2018e. Study Plan for Assessing Pesticides in Sandy Creek. Durham, NC. Water Quality Unit.
COD SW. 2018f. Synoptic Nutrient Monitoring Study Plan. Durham, NC. Water Quality Unit.
Eaton, L. 2017. Chironomus Mentum Deformities (memorandum to COD SW).
Gilliom, RJ, Barbash, JE, Crawford, CG, et al. 2006. The Quality of Our Nation's Waters - Pesticides in the Nation's
Streams and Ground Water, 1992-2001 (Revised February 15, 2007). US Geological Service (USGS) Circular 1291.
USGS. Reston, VA.
Griffith, G., Omernik, J., Comstock, J. 2002. Ecoregions of North Carolina: Regional Descriptions. Accessed
8/8/2017 at https://www.epa.gov/eco-research/ecoregion-download-files-state-region-4#pane-31
Harrelson, C.C, Rawlins, C.L., Potyondy, J.P. 1994. Stream Channel Reference Sites: An Illustrated Guide to Field
Technique. General Technical Report RM-245. Fort Collins, CO. US Dept. of Agriculture, Forest Service, Rocky
Mountain Forest and Range Experiment Station. Available online at
https://www.fs.usda.gov/treesearch/pubs/20753.
SonTek/YSI. 2009. FlowTracker Handheld ADV User's Manual, Firmware Version 3.7. San Diego, CA.
48

Appendix 1. Soil moisture historic data summaries [NO UPDATE NEEDED]
A time series of all available soil moisture data available from ECONET site DURH shows a step -type shift in annual range of soil moisture in approximately 2011.
Quantiles from results after this shift (i.e., 2012-2017) were used to determine relative values for "wet" (>75t"), "moderate" (25t" — 75t") and "dry" (<25tn)
antecedent soil moisture conditions for the purposes of storm water quality monitoring.
0.5
0.48
046 • • • • •� � ' Ai
m •
< 0.44 % • •• •• •
c s s• t •
m 042-
< • • • • • • • •. • M• • {i • •
• • • • , • �` I • i • • •
038 • i • • • • • • • i ; i ,•• •• • • • •• •
0.36
O 0.34 • �• • • i�• • • M, • • •� ' • • • i
Z • ••• •ti�i• R• `• f • • •• • • •M.i • •
0.32 _ • • • • i •? • • • • •
vOi 0.3 j • • • • H •i 1 •• • ~ • f • •• • • •
of 0.280.26
•1 • • • •�, •■ �•• • • • • • •N • •
• fi •" 1 • • I • • •
do
0.2• ifff iiii t • •
0.22 •
0.2
CO O) CY) a, O O O .--i .ti .--i N N N m m m Y Y Y Vt Vt V1 10 M
O O O O O O O O O O O O O O O O O O O O O O O O O O O O O
N N N N N N N N N N N N N N N N N N N N N N N N N N N N N
.-I
O O O O O O O O O O O O O O O O O O O O O O O O O O O O O
'� y �
ll1 6l l./1 6l 4.f1 Ol Ln Ol vt Ol Vt 6l l./1 6l Lr Ol Lr C
O O O O O O O O O O O O O O O O O O O O O O O O O O O O O
Date
Figure 24 Time series of soil moisture from NC ECONET site DURH, 2008-2017 (period of record).
Distributions Average Soil Moisture (m3/m3), 2012 - 2017
V N V co 10 V N M W l0 V N N
V V p CD
O O O O O O O O O O
=igure 25 Distributions of soil moisture (m3/m3) at ECONET site DURH, 2012 — 2017.
Quantiles
100.0%
maximum
0.433
99.5%
0.432
97.5%
0.428
90.0%
0.424
75.0%
quartile
0.417
50.0%
median
0.404
25.0%
quartile
0.343
10.0%
0.267
2.5%
0.238
0.5%
0.225
0.0%
minimum
0.206
49

Appendix 2. Summary of rain events [UPDATED 4/24]
The table below provides a list of days during the study period (2/1/2018-1/9/2019) with measurable precipitation at
the USGS rain gage at Maureen Joy Charter School (USGS site 355852078572045) and the total precipitation measured
on each day. Data were downloaded from https://waterdata.usgs.gov/nwis on 12/11/2018 and 2/18/2019. "Qualifier"
indicates USGS status of the data, where A = Approved, P = Provisional, R = Rejected. There are known data gaps, and so
results may not be accurate, for the following days in 2018: 3/12, 3/13, 3/14, 5/30, 6/4, 12/11, and 12/12.
Date
Precipitation
(in.)
Qualifier
02/02/2018
0.17
P
02/04/2018
0.81
P
02/07/2018
0.21
P
02/10/2018
0.29
P
02/11/2018
0.06
P
02/12/2018
0.08
P
02/16/2018
0.03
P
02/17/2018
0.02
P
02/19/2018
0.30
P
02/26/2018
0.14
P
02/27/2018
0.01
P
02/28/2018
0.03
P
03/01/2018
0.56
A
03/06/2018
0.51
A
03/07/2018
0.01
A
03/10/2018
0.01
A
03/11/2018
0.21
A
03/17/2018
0.01
A
03/18/2018
0.05
A
03/19/2018
0.01
A
03/20/2018
1.04
A
03/21/2018
0.03
A
03/24/2018
0.62
P
03/25/2018
0.61
P
03/30/2018
0.02
P
04/07/2018
1.06
P
04/08/2018
0.01
P
04/15/2018
2.88
P
04/16/2018
0.17
P
04/23/2018
0.04
P
04/24/2018
1.14
P
04/26/2018
0.54
P
04/27/2018
0.06
P
05/06/2018
0.05
P
05/15/2018
0.01
P
05/16/2018
1.59
P
05/17/2018
2.64
P
05/18/2018
0.34
P
05/19/2018
0.47
P
05/21/2018
1.43
P
05/22/2018
0.01
P
05/26/2018
0.29
P
Date
Precipitation
(in.)
Qualifier
05/27/2018
0.01
P
05/28/2018
0.57
P
05/29/2018
0.28
P
06/01/2018
0.02
P
06/02/2018
0.15
P
06/06/2018
0.01
P
06/08/2018
0.06
P
06/10/2018
0.14
P
06/11/2018
1.09
P
06/12/2018
0.03
P
06/21/2018
0.01
P
06/26/2018
1.74
P
07/04/2018
0.68
P
07/05/2018
0.81
P
07/06/2018
0.26
P
07/07/2018
0.09
P
07/12/2018
0.01
P
07/16/2018
0.01
P
07/17/2018
0.10
P
07/21/2018
0.02
P
07/22/2018
0.91
P
07/23/2018
0.17
P
07/25/2018
0.06
P
07/26/2018
0.01
P
07/29/2018
2.39
P
07/30/2018
0.32
P
07/31/2018
0.35
P
08/01/2018
0.01
P
08/02/2018
2.13
P
08/03/2018
1.01
P
08/07/2018
0.01
P
08/08/2018
1.55
P
08/11/2018
0.03
P
08/12/2018
0.43
P
08/13/2018
0.11
P
08/14/2018
0.07
P
08/18/2018
0.11
P
08/19/2018
0.76
P
08/20/2018
0.94
P
08/21/2018
0.05
P
08/22/2018
0.02
P
08/31/2018
0.40
P
50

Date
Precipitation
(in.)
Qualifier
09/01/2018
0.40
P
09/04/2018
0.01
P
09/07/2018
0.09
P
09/09/2018
0.33
P
09/10/2018
0.39
P
09/11/2018
0.01
P
09/12/2018
0.01
P
09/13/2018
0.05
P
09/14/2018
1.69
P
09/15/2018
1.14
P
09/16/2018
1.52
P
09/17/2018
5.14
P
09/26/2018
0.21
P
09/27/2018
0.25
P
09/28/2018
0.20
P
10/09/2018
0.20
P
10/10/2018
0.07
P
10/11/2018
2.71
P
10/14/2018
0.01
P
10/15/2018
0.01
P
10/17/2018
0.01
P
10/20/2018
0.14
P
10/21/2018
0.01
P
10/26/2018
2.17
P
10/27/2018
0.01
P
11/01/2018
0.16
P
11/02/2018
0.10
P
11/05/2018
1.42
P
11/06/2018
0.09
P
11/09/2018
0.28
P
11/12/2018
3.14
P
11/13/2018
0.86
P
11/14/2018
0.05
P
11/15/2018
0.88
P
11/24/2018
0.85
P
11/26/2018
0.04
P
11/30/2018
0.03
P
12/01/2018
0.39
P
12/02/2018
0.02
P
12/09/2018
0.85
R
12/10/2018
0.45
R
12/11/2018
0.24
R
12/14/2018
0.70
P
12/15/2018
0.54
P
12/16/2018
0.01
P
12/20/2018
0.73
P
12/21/2018
0.15
P
12/27/2018
0.01
P
Date
Precipitation
(in.)
Qualifier
12/28/2018
1.13
P
12/30/2018
0.01
P
12/31/2018
0.16
P
01/02/2019
0.02
P
01/03/2019
0.28
P
01/04/2019
0.42
P
01/05/2019
0.01
P
TOTAL FOR
STUDY PERIOD:
67.3
51

Appendix 3. Stream stage logger results [UPDATED 4/15]
Time series show stream stage (ft.) (black circles) and manual staff gage readings (ft.) (green rectangles). Individual stage readings have been flagged as possible outliers (orange) and as rejected (red) as
appropriate.
Stream stage (ft.) by dateltime
j • Logger depth (ft.)
8 € 1 1 Staff gage [ft.]
•
6 t ' [
V L kk
_ i €
Z
8
7jj
i 5
z 3]
q 0
E
v8
0 7
� 5 1
t l
z 3 ` F
1
0
8
7
r
_ 4
2
i
0
❑,�4 �'�� �'�� �'�� 0,�4 �'�� �'�� �'�$ �'�$ 0,�4 �'�$ �'�� �'�� �'�� 0,�4 ❑,�4 �'�� �'�� �'�� d'v� �'�� �'�� �'�$ �'�$ d'v`b ❑,y4 �'�� �'�� �'�� d'v� �'�� �'�� �'�� �'�� ❑,�4 �'�� �'�� �'�� �'�$ ❑,�4 �'�� �'�� �'�� �'�� ❑,�4 �'�°
a �,�. '❑ a `�e a � a �•y�' �� a�'�, o�� ��� o��� a'❑3 0 �o a �,� a �a a � �❑,��o�a�'o�,�'o��y�'a'�,ltia �•y�'a �y�'a �� o�❑•yy�'o�❑yltio�,��'o�3�'a�'p o°,'d, 0°,�3 ��p� o��,��tio'��tio�,�tio���ti1�y�tiy�,�tiy��.y�y�tiy�,�tiy���ti�4��ti����ti��a�ti��,��ay'❑3�ti
h
Date(Time
52

Appendix 4. Ambient water quality results [UPDATED 4/221
The table below contains results for NH3.3SC.
2/20/2018
5/15/2018
8/21/2018
11/13/2018
v
v
E o
2 ao
Parameter
Result
QA
code i
Result
QA code
1
Result
QA code 1
Result
QA
code i
Historic median
(Ambient
monitoring
program)
Alkalinity
(mgCaCO3/L)
62
92
48
J7, DUPMN
25
73
c
Cl (mg/L)
44
32
10.5
J7, DUPMN
5
11
DOC (mg/L)
5.51
6.94
8.41
DUPMN
13.2
N/A
>
Hardness (mg/L)
97
109
52
DUPMN
35
75
u
SO4 (mg/L)
17
22
25
DUPMN
5
U
22
TSS (mg/L)
6.0
DUPMN
5.0
U
11
DUPMN
73
6
Turbidity (NTU)
10.1
6.5
15.3
93.3
9.1
DO (mg/L)
9.6
5.0
5.9
11
7.6
DO sat (%)
87
59
72
99
78
a,
pH (SU)
7.3
7.2
6.8
5.7
7.3
SC (uS/cm at 25°C)
313
347
152
87
271
Water temp (°C)
10.7
23.5
24.9
10.7
16.0
Al, diss (ug/L)
26
87.8
317.5
DUPMN
220
N/A
Al, total (ug/L)
206
144
336
DUPMN
2,040
N/A
As, diss (ug/L)
2
U
10
U
10
U, DUPMN
10
U
N/A
As, total (ug/L)
10
U
10
U
10
U, DUPMN
10
U
N/A
Ca, total (ug/L)
27,200
29,400
14,950
DUPMN
9,670
18,906
Cd, diss (ug/L)
2
U
2
U
2
U, DUPMN
2
U
N/A
Cd, total (ug/L)
2
U
2
U
2
U, DUPMN
2
U
N/A
Cr, diss (ug/L)
5
U
5
U
5
U, DUPMN
5
U
N/A
Cr, total (ug/L)
5
U
5
U
5
U, DUPMN
5
U
N/A
Cu, diss (ug/L)
3.46
5.2
5.85
J7, DUPMN
7.5
J7
5
Cu, total (ug/L)
4.86
5.0
6.7
DUPMN
10.9
5
°
Fe, diss (ug/L)
303
444
741
DUPMN
306
N/A
Fe, total (ug/L)
940
1,170
778
DUPMN
2,090
N/A
K, total (ug/L)
3,510
5,050
3,090
DUPMN
2,720
N/A
Mg, total (ug/L)
6,960
8,660
3,520
DUPMN
2,600
5,433
Mn, diss (ug/L)
132
212
63.5
DUPMN
69
N/A
Mn, total (ug/L)
139
255
74.8
DUPMN
115
N/A
Na, total (ug/L)
3,000
23,700
9,505
J7, DUPMN
4,130
N/A
Ni, diss (ug/L)
10
U
10
U
10
U, DUPMN
10
U
N/A
Ni, total (ug/L)
10
U
10
U
10
U, DUPMN
10
U
N/A
Pb, diss (ug/L)
10
U
10
U
10
U, DUPMN
10
U
N/A
Pb, total (ug/L)
10
U
10
U
10
U, DUPMN
10
U
N/A
Zn, diss (ug/L)
10
U
10
U
10
U, DUPMN
10
U
10
Zn, total (ug/L)
18
5.6
10
U, DUPMN
16.1
J7
9.9
NH3 (mg/L)
0.04
0.16
DUPMN
0.12
DUPMN
0.02
J7
0.06
v
NOx (mg/L)
0.34
0.89
0.31
DUPMN
0.61
0.23
TO (mg/L)
0.77
1.07
DUPMN
1.77
J7, DUPMN
1.05
0.5
z
TP (mg/L)
1 0.04
1 0.09
1 DUPMN
0.37
DUPMN
0.11
0.07
1 QA codes: DUPMN: Result shown is the average of two duplicate samples.
17: Estimated. Blank contamination evident, value may not be accurate.
U: Non -detect. The contaminant was not detected at a concentration greater than the reporting limit. Reporting
limit is shown in the Result column.
53

Appendix 5. Baseflow water quality results
The following tables contain all results from quarterly baseflow sampling at the four core sites (NH1.8SCTA, NH1.6SC,
NH4.4SCTD, and NH4.7SC) that occurred on 2/23/2018 (Table 17) and 6/6/2018 (Table 18). Duplicate samples (both
analyzed by the contract lab) were collected during each sampling event; the site where they were collected is indicated
by an asterisk (*), and the table contains the average of the two analyses. For certain parameters, reportable
concentrations were reported in the associated field blanks; results for the affected parameters are shown in bold.
Table 20 Results from baseflow sampling, 2/23/2018
Parameter
NH1.8SCTA
NH1.6SC
NH4.4SCTD
*NH4.7SC
Alkalinity (mgCaCO3/L)
118
67
92
86
Total Aluminum (µg/L)
113
275
299
570
Dissolved Aluminum (µg/L)
<50
<50
50.7
<50
Ammonia Nitrogen (mg/L)
0.18
0.04
0.05
0.06
Total Arsenic (µg/L)
<2
<2
<2
<2
Dissolved Arsenic (µg/L)
<2
<2
<2
<2
Total Cadmium (µg/L)
<2
<2
<2
<2
Dissolved Cadmium (µg/L)
<2
<2
<2
<2
Total Calcium (µg/L)
38,600
23,200
30,200
30,550
Chloride (mg/L)
54
51
54
53
Total Chromium (µg/L)
<5
<5
<5
<5
Dissolved Chromium (µg/L)
<5
<5
<5
<5
Specific Conductivity (µS/cm at 25 IC)
462
330
383
409
Total Copper (µg/L)
3.18
7.11
4.83
5.50
Dissolved Copper (µg/L)
2.09
5.04
3.61
3.13
Dissolved Oxygen Saturation (%)
92
82
116
83
Dissolved Oxygen Concentration (mg/L)
9.5
8.4
11.5
8.4
Hardness (mg/L)
137
83
106
113
Total Iron (µg/L)
2,340
1,360
2,360
1,545
Dissolved Iron(µg/L)
274
453
1220
153
Total Lead (µg/L)
<0.5
1.92
1.34
1.38
Dissolved Lead (µg/L)
<0.5
<0.5
<0.5
<0.5
Total Magnesium (µg/L)
9,930
6,080
7,410
8,895
Total Manganese (µg/L)
329
262
313
196
Dissolved Manganese (µg/L)
303
216
301
131
Total Nickel (µg/L)
<10
<10
<10
<10
Dissolved Nickel (µg/L)
<10
<10
<10
<10
Nitrate + Nitrite as Nitrogen (mg/L)
0.39
<0.1
0.11
0.33
Dissolved Organic Carbon (mg/L)
4.66
5.92
5.46
4.02
pH
7.6
7.3
7.3
7.5
Total Potassium (µg/L)
2,460
3,140
2,850
3,155
Total Sodium (µg/L)
37,100
35,200
36,600
41,100
Sulfate (mg/L)
27
19
21
29
Temperature (IC)
14.1
14.1
16.0
14.7
Total Kjeldahl Nitrogen (mg/L)
0.54
0.44
0.69
0.57
Total Phosphorus (mg/L)
<0.02
<0.02
<0.02
<0.02
Total Suspended Solids (mg/L)
7.0
<2.5
16.0
28.5
Turbidity (NTU)
18.7
5.6
13.5
16.5
Total Zinc (µg/L)
53.8
22.4
28.5
66.7
Dissolved Zinc (µg/L)
1 36.4
1 14
1 22.3
1 43.2
54

Table 21 Results from baseflow sampling, 6/6/2018
Parameter
NH1.8SCTA
NH1.6SC
NH4.4SCTD
*NH4.7SC
Alkalinity (mgCaCO3/L)
140
68.5
101
86
Total Aluminum (µg/L)
123
116
229
427
Dissolved Aluminum (µg/L)
63.8
55.8
58.3
53.7
Ammonia Nitrogen (mg/L)
0.13
0.11
0.14
0.05
Total Arsenic (µg/L)
<10
<10
<10
<10
Dissolved Arsenic (µg/L)
<10
<10
<10
<10
Total Cadmium (µg/L)
<2
<2
<2
<2
Dissolved Cadmium (µg/L)
<2
<2
<2
<2
Total Calcium (µg/L)
42,700
22,500
30,500
26,600
Chloride (mg/L)
34
20
28
32
Total Chromium (µg/L)
<5
<5
<5
<5
Dissolved Chromium (µg/L)
<5
<5
<5
<5
Specific Conductivity (µS/cm at 25 °C)
444
228
318
333
Total Copper (µg/L)
6
12.35
5.4
6.5
Dissolved Copper (µg/L)
4.8
7.9
5.1
4.6
Dissolved Oxygen Saturation (%)
79
75
54
64
Dissolved Oxygen Concentration (mg/L)
7.2
6.4
4.8
5.7
Hardness (mg/L)
153
81
108
99
Total Iron (µg/L)
3,050
1,230
2,050
1,470
Dissolved Iron(µg/L)
186
767
1,000
198
Total Lead (µg/L)
<10
<10
<10
<10
Dissolved Lead (µg/L)
<10
<10
<10
<10
Total Magnesium (µg/L)
11,200
6,095
7,740
8,010
Total Manganese (µg/L)
367
280
254
249
Dissolved Manganese (µg/L)
353
263
236
93.5
Total Nickel (µg/L)
<10
<10
<10
<10
Dissolved Nickel (µg/L)
<10
<10
<10
<10
Nitrate + Nitrite as Nitrogen (mg/L)
0.47
0.18
0.24
0.36
Dissolved Organic Carbon (mg/L)
5.48
8.44
7.64
4.58
pH
7.4
7.3
6.6
7.1
Total Potassium (µg/L)
2,630
3,565
3,420
2,890
Total Sodium (µg/L)
40,100
15,350
21,600
31,200
Sulfate (mg/L)
23
14
208
28
Temperature (°C)
19.9
22.9
21.2
20.8
Total Kjeldahl Nitrogen (mg/L)
0.75
0.81
0.80
0.58
Total Phosphorus (mg/L)
0.041
0.063
0.067
0.099
Total Suspended Solids (mg/L)
8.0
10.6
10.6
37.6
Turbidity (NTU)
24.2
6.8
16.5
20.9
Total Zinc (µg/L)
42.8
<10
<10
29
Dissolved Zinc (µg/L)
26.7
<10
<10
10.6
55

Appendix 6. Storm sampling summaries [UPDATED 4/241
NH1.6SC: Sandy Cr. at Larchmont Rd.
Sampling summary
Sample start
Sample end
Sample enable (ft.)
Sample volume (mQ
Sample interval
(min.)
# samples collected
Total # bottles
collected
Bottle volume, type
ISCO unit
Battery #
Sample processing
Bottles composited
Composite start
time
Composite end time
Comment
Storm 1A
N/A
N/A
0.066 + 0.25ft. = 0.30
250
15
0
0
1L, ProPak
2070
Unknown
N/A
N/A
Storm 1B
5/17/2018 15:37
5/18/2018 09:22
0.565 + 0.25 = 0.75
250
15
59
18
1L, ProPak
2070
Unknown
#5-15
5/17/2018 19:37
Storm 2
6/26/2018 12:32
6/26/2018 22:17
1.117 + 0.25 = 1.37
250
15
40
10
ProPak
2070
C
#1-8
6/26/2018 12:32
Storm 3
N/A
N/A
0.54 + 0.25 = 0.80
100
15
0
0
1L, ProPak
2070
E
N/A
N/A
Storm 4
12/13/2018 18:23
12/15/2018 15:07
1.096 + 0.35 = 1.45
100
15
180
18
1L, ProPak
2070
Boat batter,
#8-18
12/14/2018 11:53
N/A
5/18/2018 06:22
6/26/2018 20:17
N/A
12/15/2018 15:08
Equipment failure, no
Expected stream stage
Bottles 2 and 8 only
Power failure at
Sampling initiated early;
samples collected. ISCO
to drop slightly before
partially filled (about %2--
initiation of sampling, no
there was a rise in stage
was not communicating
rain started so used
%full).
samples collected.
due to snow melt. Stage
with Level Troll.
lower stage for enable.
decreased again before
Sampler triggered early
rain began.
(before start of rain), but
stage logger data shows
two small "bumps" in
stage during this
afternoon. Sampling still
active when retrieved on
5/19, program manually
stopped at
approximately 10:15.
56

NH1.8SCTA: Tributary A at Martin Luther King, Jr. Blvd.
Storm 1A Storm 2 Storm 3 Storm 4
Sampling summary
Sample start
5/16/2018 19:39
6/26/2018 12:00
12/9/2018 13:21
12/14/2018 11:03
Sample end
5/17/2018 02:09
6/26/2018 16:15
12/11/2018 11:06
12/16/2018 7:47
Sample enable (ft.)
-0.200 + 0.25 = 0.10
-0.14 + 0.25 = 0.11
-0.11 + 0.25 = 0.14
0.213 + 0.35 = 0.56
Sample volume
250
500
100
100
(mQ
Sample interval
15
15
15
15
(min.)
# samples collected
26
18
180
180
Total # bottles
6.5
18
18
18
collected
Bottle volume,
1L, ProPak
1L, ProPak
1L, ProPak
1L, ProPak
type
ISCO unit
4475 (model 6700)
4475 (model 6700)
4475 (model 6700)
4475 (model 6700)
Battery #
Unknown
D
Unknown
D
Sample processing
Bottles composited
#1 - 7
#1— 18
#1 - 14
#1— 11
Composite start
5/16/2018 19:39
6/26/2018 12:18
12/9/2018 13:21
12/14/2018 11:03
time
Composite end
5/17/2018 01:30
6/26/2018 17:03
12/11/2018 00:05
12/15/2018 14:17
time
Comment
Used ISCO Flow Module and
Equipment blank, field blank,
Used ISCO Flow Module and
Used ISCO Flow Module and
Flow Probe to monitor
and duplicate collected here.
Flow Probe to monitor stream
Flow Probe to monitor stream
stream stage (did not have
Used ISCO Flow Module and
stage. ISCO probe stage
stage. ISCO probe stage
proper cord for hooking to
Flow Probe to monitor stream
readings are biased (lower than
readings are biased (lower than
Level Troll). ISCO probe stage
stage. ISCO probe stage
actual). Equipment blank and
actual).
readings are biased (lower
readings are biased (lower than
field blank collected at this site.
than actual). Program error
actual). Error in programming —
(no liquid detected) during
only collected one 500ml-
filling of bottle 7 so only
sample/bottle.
partially filled.
57

NH4.4SCTD: Tributary D at Academy Rd.
Storm 1A Storm 2 Storm 3 Storm 4
Sampling summary
Sample start
5/16/2018 20:37
N/A
12/9/2018 15:05
12/14/2018 12:02
Sample end
5/17/2018 03:52
N/A
12/11/2018 11:50
12/16/2018 8:47
Sample enable (ft.)
1.267 + 0.25 = 1.50
1.117 + 0.25 = 1.37
Liquid level actuator installed at
0.25ft. above current stage
Liquid level actuator installed at
0.34ft. above current stage
Sample volume
(mL)
250
250
100
100
Sample interval
(min.)
15
15
15
15
# samples collected
30
0
180
180
Total # bottles
collected
7.5
0
18
18
Bottle volume,
type
1L, ProPak
1L, ProPak
1L, ProPak
1L, ProPak
ISCO unit
0099
0099
0099
0099
Battery #
Unknown
B
Unknown
B
Sample processing
Bottles composited
#1 - 7
N/A
#1 - 13
#1— 11
Composite start
time
5/16/2018 20:37
N/A
12/9/2018 15:05
12/14/2018 12:02
Composite end
time
5/17/2018 03:52
N/A
12/10/2018 23:20
12/15/2018 15:17
Comment
Power failure during
sampling, bottle 8 only
partially filled.
Equipment failure.
Stream stage during installation
was 0.31 ft. at the staff gage.
Samples composited and
processed in field.
W

NH4.7SC: Sandy Cr. at Morreene Rd.
Storm 1A Storm 2 Storm 3 Storm 4
Sampling summary
Sample start
5/16/2018 19:54
6/26/2018 12:18
12/9/2018 14:40
12/14/2018 11:30
Sample end
5/17/2018 10:24 (stopped
manually)
6/26/2018 17:03
12/11/2018 7:56
12/16/2018 8:15
Sample enable (ft.)
0.554 + 0.25 = 0.80
0.400 + 0.25 = 0.65
0.416 + 0.25 = 0.67
0.844 + 0.35 = 1.19
Sample volume
(mL)
250
250
100
100
Sample interval
(min.)
15
15
15
15
# samples collected
59
20
161
180
Total # bottles
collected
14.75
5
16.1
18
Bottle volume,
type
1L, ProPak
1L, ProPak
1L, ProPak
1L, ProPak
ISCO unit
0336
0336
0336
0336
Battery #
Unknown
A
C
C
Sample processing
Bottles composited
#1 - 7
#1 - 5
#1 - 12
#1 - 11
Composite start
time
5/16/2018 19:54
6/26/2018 12:18
12/9/2018 14:40
12/14/2018 11:30
Composite end
time
5/17/2018 02:39
6/26/2018 17:03
12/10/2018 20:25
12/15/2018 14:45
Comment
Duplicate samples
Power failure at 17:18 (first
sample of bottle 6).
Power failure 12/11/2018 7:11
Samples composited and
processed in field.
59

Appendix 7. Storm water chemistry results
Table 22 Water chemistry results for Storm Events 1 and 2.
Results from duplicate samples are shown for NH4.7SC (Storm 1) and NH1.8SCTA (Storm 2).
Site
NH1.6SC
NH1.8SCTA
NH4.4SCTD
NH4.7SC
Mean, all sites
Storm #
1
2
1
2
1
1
2
1
2
Alkalinity (mgCaCO3/L)
22
28
13.9
12
10.5
29
22
21
30
21.58
20.125
Aluminum (ug/L)
6,200
4,160
1,380
1,960
1,810
2,030
2,540
2,630
1,220
2,956
2,288
Aluminum, dissolved (ug/L)
54.1
83.7
140
75.9
93.6
65.6
599
581
81.3
288
84
Ammonia Nitrogen (mg/L)
0.31
0.12
0.18
0.29
0.29
0.18
0.2
0.17
0.14
0.21
0.21
Arsenic(ug/L)
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
Arsenic, dissolved (ug/L)
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
Cadmium (ug/L)
<2
<2
<2
2.9
<2
<2
<2
2.5
<2
2.1
2.2
Cadmium, dissolved (ug/L)
<2
<2
8.5
<2
<2
3.2
<2
2.2
<2
3.6
<2
Calcium (ug/L)
9,890
10,200
5,620
4,870
4,900
10,000
8,000
8,380
7,400
8,378
6,843
Chloride (mg/L)
7
7
8
2
1
8
6
7
8
7
5
Chromium (ug/L)
13.4
9.1
<5
<5
<5
<5
7.8
7.3
<5
7.7
6.0
Chromium, dissolved (ug/L)
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
Copper (ug/L)
24.7
22.6
16.4
21.0
29.0
16.6
31.2
26.4
14.8
23.1
21.9
Copper, dissolved (ug/L)
9.5
9.2
38.2
6.2
11.3
26.9
20.7
18.9
6.2
22.8
8.2
Iron (ug/L)
7,970
5,840
2,070
4,200
4,430
2,540
4,020
3,960
1,820
4,112
4,073
Iron, dissolved (ug/L)
196
130
188
96.1
116
223
890
874
90.9
474
108
Lead (ug/L)
21.1
12.9
<10
<10
<10
<10
<10
10.7
<10
12.4
10.7
Lead, dissolved (ug/L)
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
Magnesium (ug/L)
4,070
3,510
1,480
1,390
1,410
2,770
2,570
2,680
1,940
2,714
2,063
Manganese(ug/L)
443
686
142
416
352
151
325
312
168
275
406
Manganese, dissolved (ug/L)
64.7
18.6
26.4
59.6
59.1
60.7
53.5
55.2
16.6
52.1
38.5
Nickel (ug/L)
15.1
10.8
<10
<10
<10
<10
10.1
<10
<10
11.0
10.2
Nickel, dissolved (ug/L)
<10
<10
<10
<10
<10
<10
<10
<10
<10
10
10
Nitrate + Nitrite as N (mg/L)
1.52
0.31
0.36
0.24
0.22
0.46
2.69
0.54
0.31
1.11
0.27
Organic Carbon, dissolved (mg/L)
8.64
8.56
8.33
9.46
9.25
8.5
9.44
9.41
7.97
8.86
8.81
Potassium (ug/L)
3,420
3,020
2,260
1,690
1,690
3,030
2,810
2,910
1,950
2,886
2,088
Sodium (ug/L)
6,780
6,900
4,300
2,910
2,920
7,620
6,830
7,100
7,660
6,526
5,098
Sulfate (mg/L)
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
Total Kjeldahl Nitrogen (mg/L)
2.99
1.20
1.76
<0.2
1.08
1.60
2.41
2.67
0.81
2.29
0.82
Total Phosphorus (mg/L)
0.227
0.260
0.184
0.225
0.219
0.163
0.281
0.289
0.153
0.229
0.214
Total Suspended Solids (mg/L)
357
279
81.3
145
153
82.5
157
152
115
166
173
Zinc (ug/L)
71.0
70.8
68.0
136
127
29.5
110
121
74.6
79.9
102
Zinc, dissolved (ug/L)
<10
10.2
27.1
28.7
31.8
11.7
30.4
30.2
18.4
21.9
22.3

Table 23 Results from storm sampling equipment and field blanks.
STORM 1
STORM 2
Parameter
Equipment
Blank
Field
Blank
Equipment
Blank
Field
Blank
Alkalinity (mgCaCO3/L)
<1
<1
<1
<1
Aluminum (ug/L)
<50
<50
<50
<50
Aluminum, dissolved (ug/L)
<50
<50
<50
<50
Ammonia Nitrogen (mg/L)
0.07
0.04
0.02
<0.02
Arsenic (ug/L)
<10
<10
<10
<10
Arsenic, dissolved (ug/L)
<10
<10
<10
<10
Cadmium (ug/L)
<2
<2
<2
<2
Cadmium, dissolved (ug/L)
<2
<2
<2
<2
Calcium (ug/L)
<100
<100
<100
<100
Chloride (mg/L)
<0.5
<0.5
<0.5
<0.5
Chromium (ug/L)
<5
<5
<5
<5
Chromium, dissolved (ug/L)
<5
<5
<5
<5
Copper (ug/L)
5.9
6.4
4.1
<2
Copper, dissolved (ug/L)
7.1
8.7
3.7
<2
Iron (ug/L)
<25
<25
<25
<25
Iron, dissolved (ug/L)
<25
68.9
<25
<25
Lead(ug/L)
<10
<10
<10
<10
Lead, dissolved (ug/L)
<10
<10
<10
<10
Magnesium (ug/L)
<100
<100
<100
<100
Manganese (ug/L)
<5
<5
<5
<5
Manganese, dissolved (ug/L)
<5
<5
<5
<5
Nickel (ug/L)
<10
<10
<10
<10
Nickel, dissolved (ug/L)
<10
<10
<10
<10
Nitrate + Nitrite as N (mg/L)
0.27
<0.1
7.4
<0.1
Organic Carbon, dissolved (mg/L)
<1
<1
1.24
<1
Potassium (ug/L)
<200
<200
<200
<200
Sodium (ug/L)
277
<200
<200
279
Sulfate (mg/L)
<5
<5
<5
<5
Total Kjeldahl Nitrogen (mg/L)
0.6
0.44
<0.2
<0.2
Total Phosphorus (mg/L)
<0.02
<0.02
<0.02
<0.02
Total Suspended Solids (mg/L)
<2.5
<2.5
<2.5
<2.5
Zinc (ug/L)
<10
<10
<10
<10
Zinc, dissolved (ug/L)
<10
<10
<10
<10
61

Appendix 8. Specific conductance time series
Graphs below show results from continuous monitoring of specific conductance (uS/cm at 2S°C) using instream loggers.
Red points have been flagged as rejected based on documented logger issues. Blue asterisks have been flagged as
possible outliers, possibly due to potential illicit discharges.
NH1.6SC
800
700 3
600
a 500: i •
c 300
s
00
100
a a a a a s s s s a a a o o a a o 0 0
N H 1.8SCTA
210:- .
_ zo
^+ 1
E 15.
14 ��
a a a o o a a a a a a a a a a a a a
O� F V .�i
O O O O O O O O O 4 O O O O O O O O
Date
NH4.4SCTD
750
z 700
650
600
550
50"
z a 450
400
c 350
300
0 250
u 200
150
100 —
50 —
0
Date Time. GMT-05:00
62

NH4.7SC
z
zz�
z� -
"' isc�
16K
15-,-.
i 14K
13"-,
1100
1000
900
0 70000
500
a 400
200
100
AU;
a a o o a o c c a
.Y � .Y .Y � .Y .YZ.
RO R R V
O O O O O O O O O
Date
lvr�
63

Appendix 9. Sediment chemistry results
Individual results from sediment chemistry analyses are provided in the table below.
Parameter
TEC
mg/kg
PEC
mg/kg
NH1.8SCTA
NH1.6SC
NH4.4SCTD
NH4.7SC
Citywide means
(COD SW data)
Result
(mg/kg)
PEC-
Q
Result
(mg/kg)
PEC-
Q
Result
(mg/kg)
PEC-
Q
Result
(mg/kg)
PEC-Q
Result
(mg/kg)
PEC-Q
Organic carbon
N/A
N/A
12,900
(1.29%)
N/A
22,800
(2.28%)
N/A
21,700
(2.17%)
N/A
22,700
(2.27%)
N/A
75,022
(7.50%)
N/A
Al
N/A
N/A
1,461
N/A
1,492
N/A
3,229
N/A
3,939
N/A
4,596
N/A
Fe
N/A
N/A
1,904
N/A
2,817
N/A
5,602
N/A
6,852
N/A
N/A
N/A
Pb
35.8
128
2.04
0.02
5.14
0.04
10.56
0.08
7.96
0.06
7.07
0.055
Mn
N/A
N/A
45.4
N/A
83.13
N/A
149.7
N/A
145.10
N/A
N/A
N/A
As
9.79
33
< 0.670
0.020
< 0.774
0.023
< 0.873
0.026
< 0.732
0.022
2.43
0.074
Cd
0.99
4.98
< 0.134
0.027
< 0.155
0.031
< 0.175
0.035
< 0.146
0.029
0.05
0.010
Cr
43.4
111
1.78
0.02
4.27
0.04
5.81
0.05
9.56
0.09
34.79
0.313
Ni
22.7
48.6
3.53
0.07
3.27
0.07
3.73
0.08
7.01
0.14
8.01
0.165
Zn
121
459
17.30
0.04
28.80
0.06
33.50
0.07
54.76
0.12
25.82
0.056
Cu
31.6
149
1.42
0.01
1.49
0.01
5.08
0.03
6.52
0.04
8.00
0.054
Anthracene
57.2
845
< 0.334
0.000
< 0.334
0.000
< 0.334
0.000
< 0.334
0.000
<0.063 a
0.000
Benzo(a)pyrene
150
1,450
< 0.334
0.000
< 0.334
0.000
< 0.334
0.000
1.09
0.001
0.024
0.000
Chrysene
166
1,290
< 0.334
0.000
< 0.334
0.000
< 0.334
0.000
1.75
0.001
0.040
0.000
Fluoranthene
423
2,230
0.506
0.000
< 0.334
0.000
< 0.334
0.000
3.28
0.001
0.072
0.000
Naphthalene
176
561
< 0.334
0.001
< 0.334
0.001
< 0.334
0.001
< 0.334
0.001
<0.046 a
0.000
Pyrene
195
1,520
0.453
0.000
< 0.334
0.000
< 0.334
0.000
2.64
0.002
0.050
0.000
Benzo(a)anthracene
108
1,050
< 0.334
0.000
< 0.334
0.000
< 0.334
0.000
1.26
0.001
0.027
0.000
Benzo(g,h,i)perylene
N/A
N/A
< 0.334
N/A
< 0.334
N/A
< 0.334
N/A
0.684
N/A
0.058
N/A
Dibenzo(a,h)anth racene
33
N/A
< 0.334
N/A
< 0.334
N/A
< 0.334
N/A
< 0.334
N/A
0.012
N/A
Indenol(1,2,3-cd)pyrene
N/A
N/A
< 0.334
N/A
< 0.334
N/A
< 0.334
N/A
0.627
N/A
0.058
N/A
Phenanthrene
204
1,170
0.463
0.000
< 0.334
0.000
< 0.334
0.000
1.28
0.001
0.058
0.000
Total PAH
1,610
22,800
1.422
N/A
<0.334
N/A
<0.334
N/A
11.3
N/A
0.508
0.000
Mean PEC-Q
N/A
N/A
0.013
0.018
0.026
0.034
0.048
Incidence of Toxicity (%)
N/A
N/A
1.4
1.9
2.6
3.5
4.9
64

Appendix 10. Existing Data Summary [PROVIDE AS A SEPARATE DOCUMENT]
[�