Ft_Jackson_Monitoring_Report_20160216.pdf

Woolpert Inc.

February 2016 Fort Jackson Environmental Division 1

Continuous Water Quality Monitoring

Report Fort Jackson Environmental Division

February 2016

1 Introduction The Fort Jackson Environmental Division (ENV) has implemented a progressive water quality monitoring program designed to assess the water quality conditions draining Fort Jackson’s cantonment area and to ensure compliance with ENV’s Phase II National Pollutant Discharge and Elimination System (NPDES) permit for its Municipal Separate Storm Sewer System (MS4). ENV’s continuous monitoring data is supplemented by a bacteria grab sample collection program. The report contained herein provides a description of the ENV monitoring program, including relevant background information and an explanation of program methods, and summarizes continuous water quality data collected during the 2015 calendar year.

1.1 Background

On January 1st, 2014, the new permit issued by the South Carolina Department of Health and Environmental Control (SCDHEC) (Permit No. SCR030000) went into effect. Fort Jackson, as a Phase II permittee, is required to comply with this permit, which includes provisions to create and implement a monitoring plan to assess water quality from the MS4 if located within a TMDL watershed. In 2009, fecal coliform and dissolved oxygen (DO) TMDL requirements were established for the Gills Creek Watershed thereby, requiring the Fort to monitor the water quality of representative MS4 discharges to the water body.

Following the publication of the Gills Creek Watershed DO and fecal coliform TMDLs, ENV focused its monitoring program on a continuous ambient water quality monitoring approach for tributaries draining to the Gills Creek Watershed.

In January 2015, the station was permanently installed in its current location, referred to as “ENV5,” an acronym assigned to this location during the initial sampling site inventory process. The ENV5 station is located on Wildcat Creek, within the boundary of Fort Jackson, providing ENV with access to the monitoring site and draining approximately 40% of the urbanized area within Fort Jackson that is located in the Gills Creek TMDL watershed. The ENV5 location has been recording data since January, 2015, and provides a good representation of the water quality of Wildcat Creek as it exits the Fort’s jurisdiction.

1.2 Continuous Water Quality Monitoring

For the past several stages of its monitoring program, ENV has elected to use high frequency, continuous data, coupled with grab sample data, rather than solely relying on periodic grab sample data to assess water quality. In a continuous monitoring approach, sensors are deployed in a waterbody for extended periods of time, generally 4 to 6 weeks. During these deployments, water quality parameters are sampled at high frequencies, in the case of ENV’s program, at 15-minute increments. Continuous data collection has enabled ENV to observe parameter concentration fluctuations in greater detail, after working hours, and during inclement weather—conditions that would be difficult to characterize through manual grab sample analysis alone. Additionally, ENV maintains a remote telemetry system which communicates the collected data in real-time to a monitoring website, allowing for efficient personnel response to equipment malfunctions or potential illicit discharge events.

February 2016 Fort Jackson Environmental Division 2

ENV’s continuous dataset is collected by a multi-parameter data sonde manufactured by Yellow Springs Incorporated (YSI). The specific instrument used to monitor Wildcat Creek is the 6600 V2-2 sonde, which collects 15-minute data for depth, dissolved oxygen (DO), turbidity, specific conductivity, temperature, and pH.

The water quality data collected at the ENV5 station is complemented and informed by grab sample bacteria data, meteorological data, and flow data. These datasets are provided by a Hydrologic Services tipping bucket rain gage and Vaisala WXT520 meteorological station located in the Fort and operated by ENV, as well as a Sontek IQ area-velocity meter installed in Wildcat Creek shortly downstream from the ENV5 station.

1.3 Watershed Description

Figure 1 shows the permanent ENV5 monitoring location and the drainage area corresponding to the monitoring station. The area in gray shows the drainage area captured by the permanent station, ENV5. This drainage area is located entirely within Fort Jackson, and drains approximately two square miles of the Fort’s urbanized area.

February 2016 Fort Jackson Environmental Division 3

Figure 1: Monitored Watershed Area

February 2016 Fort Jackson Environmental Division 4

2 Methods The section below describes the methods followed in the ENV monitoring program. It includes a description of the procedures for data collection and equipment maintenance, the techniques used to process the collected data, and the assumptions made in comparing the processed data to regulatory guidelines and standards.

2.1 Data Collection

The conclusions drawn from a water quality monitoring program are only as accurate and reliable as the quality of the collected dataset. As such, ENV places a high priority on maintaining its water quality equipment, and incorporating other data only from reliable sources.

2.1.1 Continuous Water Quality Data

The continuous water quality monitoring dataset analyzed in this report was collected by means of a Yellow Springs Instruments (YSI) 6600 V2-2 sonde equipped with sensors that measure the pH, specific conductivity, temperature, dissolved oxygen, turbidity, and depth of the creek at 15-minute intervals. The sonde was removed from the station periodically for calibration. Calibration procedures outlined by the manufacturer were followed to ensure that the water quality sensors were collecting accurate readings.

2.1.2 Grab Sample Water Quality Data

ENV began a grab sampling program at the location of the permanent monitoring station, ENV5, following the installation of this station. Under this program, bacteria grab samples are collected by field staff throughout storm event hydrographs, and samples are analyzed by a certified laboratory for E.coli. Samples can only be collected when conditions are safe for field personnel and when sufficient time is allowed for the samples to be delivered to the lab within the required maximum holding time for bacteria samples. For these reasons, the dataset of bacteria samples collected during this reporting period does not contain an adequate number of samples for statistical analysis. As additional samples are collected in the future, summary statistics will be calculated and reported.

2.1.3 Rainfall Data

A Hydrologic Services tipping bucket rain gage was installed by ENV on Fort Jackson property during July of 2015.

Prior to this time, rainfall data was obtained via a Weather Underground Personal Weather Station (PWS) (KSCCOLUM29) located within Ft. Jackson.1 This station is located near the corner of Semmes Rd and Chestnut Rd.

Following the installation of a dedicated ENV rain gage in July of 2015, rainfall data from this location was used to inform the analysis of water quality conditions at ENV5. Daily rainfall data for the monitored period was used for the analysis of water quality data on a seasonal scale. Following manufacturer recommendations, a verification test was performed on the Hydrologic Services tipping bucket rain gage in November of 2015 to ensure proper functioning.

2.1.4 Flow Data

Flow data was calculated using a combination of direct measurement and the stage-discharge relationship. Direct velocity and depth measurements were taken in real time using a SonTek IQ installed within a box culvert, and discharge was calculated using the continuity equation (Q=VA). When depth in the box culvert fell below the minimum operating depth of the IQ (0.4 feet), discharge measurements were performed using a SonTek FlowTracker, and a Campbell CS451 pressure transducer was used to determine stage in 5 minute intervals. The resulting stage-discharge relationship was used only during periods when the level of flow failed to exceed the IQ’s minimum operating depth.

1 http://www.wunderground.com/personal-weather-station/dashboard?ID=KSCCOLUM29

February 2016 Fort Jackson Environmental Division 5

2.2 Data Processing

Before performing any type of analysis on the continuous water quality data, the raw dataset was corrected using specialized software called AQUARIUS, specifically designed for continuous water quality data preprocessing. Time series datasets are corrected to ensure that erroneous data that are not representative of true in-stream conditions are not used in the data analysis process. Such erroneous data can be caused by sensor malfunctions, debris entrapment in the sonde stilling well, or other unpredictable equipment issues. Because this data is not representative of true in-stream conditions, it is removed before data analysis is performed to avoid introducing any bias into the dataset. Where needed, approximation methods such as linear interpolation and averaging techniques were used to fill data gaps of less than four hours. On days when the sonde was removed for calibration or redeployed after calibration, a complete dataset was not captured. In order to avoid influencing daily averages from the inclusion of partial days of data, the entire day of the dataset was discarded during these calibration periods.

2.3 Data Analyses and Water Quality Standards

Daily and seasonal analyses were performed on the final ENV5 corrected dataset. The data was then compared to a set of water quality standards set by SCDHEC, listed in Table 1 below. Notice should be taken that the 50 NTU water quality standard is not intended for continuous water quality data or for stormwater runoff. These standards are used here as a basis for assessing water quality conditions, but an exceedance of these values does not necessarily imply an exceedance of these standards.

Table 1: Water Quality Standards2

Parameter Limit/Range Applicability pH 6.0 – 8.5 Freshwaters

DO

Minimum 5 mg/L Daily average values

Minimum 4 mg/L Instantaneous values

Turbidity Maximum 50 NTU Not explicitly a stormwater standard

Temperature Maximum 90 °F This temperature cannot be exceeded as a result of the discharge of heated liquids

2 SCDHEC. R.61-68, Water Classifications & Standards. June 22, 2012.

February 2016 Fort Jackson Environmental Division 6

3 Results and Discussion The results presented below include data from the permanent station on Wildcat Creek (ENV5) during 2015. In order to most effectively analyze the dataset, each monitored parameter at each station was analyzed on both an annual scale and a seasonal scale. In addition, the collected turbidity dataset at the ENV station was compared against seasonal rainfall patterns.

The intent of the analysis of the 2015 data was to provide an understanding of the typical range and distribution of water quality values over the course of approximately one year at the ENV station. (The ENV station was installed on January 7, 2015.) Two main graphical approaches are used to display the data: a cumulative frequency distribution and a seasonal box and whisker plot. The cumulative frequency distributions contain the 2015 data in the format of daily average values. A cumulative frequency plot shows on the y-axis the percentage of the daily average values for which the parameter was at or below the corresponding level indicated on the horizontal x-axis.

For example, on each of these charts, 100% of the recorded daily average values were lower than the maximum daily average value, so the maximum daily average value corresponds to a cumulative frequency of 100%.

In order to provide a more refined understanding of water quality parameters over the monitored period at each station, a seasonal analysis of the dataset was also conducted. In the analysis, the 15-minute dataset at each station was divided into four seasons. For these purposes, the seasons were defined as follows: winter (January through March), spring (April through June), summer (July through September), and fall (October through December).

Because this analysis was conducted on the discrete, 15-minute dataset rather than a dataset of daily average values, it communicated the most extreme recorded values of each parameter. The data is presented in the form of box and whisker distributions. These distributions include whiskers which indicate the minimum and maximum recorded values, central boxes with lines representing the first quartile, median, and third quartile of the data distribution, and red triangles which show the average of each dataset. Additionally, where appropriate, the SCDHEC ambient water quality standard is displayed on this graph.

The figures displayed in the sections below were calculated from a dataset containing water quality values measured over the course of 2015 at the permanent monitoring station, ENV5.

3.1 ENV5 – Temperature

Figure 2 and 3 present the collected water temperature data at the ENV5 station. In 2015, this station recorded instantaneous temperatures ranging from 39°F to 89°F. The daily average temperature range spanned from 41°F to 86°F as shown in Figure 2. All of these recorded temperatures were below the SCDHEC water quality standard of 90°F. Figure 3 shows the seasonal distribution of temperatures; the water temperatures followed typical seasonal patterns, with the highest temperatures in the summer and the lowest temperatures in the winter.

February 2016 Fort Jackson Environmental Division 7

Figure 2: Daily Average Water Temperature Data Collected at the ENV5 Station in 2014

Figure 3: Discrete Water Temperature Data Collected at the ENV5 Station in 2015

40 45 50 55 60 65 70 75 80 85 90

C u m m u la ti ve

F re q u e n cy

Temperature (°F)

Cumulative Frequency Distribution ENV5 - Daily Average Water Temperature

January 2015 - December 2015

Winter '15 Spring '15 Summer '15 Fall '15

Te m p e ra tu re

°F

Seasonal Continuous Monitoring Statistics ENV5 - Discrete Temperature Data

January 2015 - December 2015

50th - 75th Percentile

25th - 50th Percentile

Mean

February 2016 Fort Jackson Environmental Division 8

3.2 ENV5 – Turbidity

Figure 4 illustrates the relatively low turbidity levels observed at the ENV5 station for much of the 2015 year, with daily average values below the SCDHEC standard of 50 NTU for approximately 92% of the year. The seasonal average turbidity levels, shown as red triangles in Figure 5, were also all below the 50 NTU threshold. (Note that the maximum turbidity values are omitted from Figure 5 in order to preserve the y-axis at a reasonable scale.)

The ENV5 station recorded significantly higher turbidity levels in the fall of 2015 compared to any of the other monitoring seasons, as shown in Figure 5. Turbidity is a rainfall driven parameter, and indeed the high turbidity levels in the fall of 2015 appear to be related to the extraordinarily high rainfall totals recorded from October through December of 2015, as shown in Figure 6. The high number of large storm events during this season likely lead to increased erosion in the watershed, driving turbidity levels up. The most significant event during this season occurred between October 2nd and 5th, when a historic rainfall event impacted the monitored watershed, bringing 16.47 inches of precipitation. This historic storm resulted in the failure of many dams in the Midlands of South Carolina, including the Semmes Lake dam, located a short way upstream from the ENV5 station. The failure of this dam and subsequent storm events contributed significantly to the elevated turbidity levels observed during the fall of 2015. The dam failure likely resulted in the immediate release of significant sediment quantities from the earthen dam as well as the erosion and failure of many stream banks along Wildcat Creek during the associated flooding. This watershed will likely require a significant period of time before the stream banks become stabilized and turbidity levels return to ambient conditions.

Figure 4: Daily Average Turbidity Data Collected at the ENV5 Station in 2015

1 10 100 1000

C u m m u la ti ve

F re q u e n cy

Turbidity (NTU)

Cumulative Frequency Distribution ENV5 - Daily Average Turbidity

January 2015 - December 2015

February 2016 Fort Jackson Environmental Division 9

Figure 5: Discrete Turbidity Data Collected at the ENV5 Station in 2015

Figure 6: Seasonal Rainfall Totals and Average Seasonal Turbidity at ENV5 during 2015

Winter '15 Spring '15 Summer '15 Fall '15

Tu rb id it y (N

TU

Seasonal Continuous Monitoring Statistics ENV5 - Discrete Turbidity Data

January 2015 - December 2015

50th - 75th Percentile

25th - 50th Percentile

Mean

Winter '15 Spring '15 Summer '15 Fall '15

Se as o n al A ve ra ge T u rb id it y (N

TU

R ai n fa ll To ta l ( in

Seasonal Rainfall and Average Turbidity ENV5 - January 2015 - December 2015

Rainfall

Average Turbidity

February 2016 Fort Jackson Environmental Division 10

3.3 ENV5 – pH

Figure 7 and 8 display the pH levels recorded at the ENV5 station during the 2015 monitoring year. As the cumulative frequency distribution in Figure 7 demonstrates, daily average pH levels were above the minimum pH standard of 6 for the vast majority of the monitoring year, with less 1% of the data (or 1 daily average value) falling below this limit. The lowest daily average pH level was recorded on October 4th, during the historic flooding event. In the Southeast, rainfall typically has a fairly low pH of 4.2 to 4.63. The low pH of subsequent runoff, in combination with potential impacts from acidic soils, leads to decreases in pH values during and shortly after rain events. Such a significant volume of rainfall as was recorded on October 4th unsurprisingly caused a drop in pH at the ENV5 location.

Figure 8 demonstrates a slight seasonal pattern in pH values, with higher levels in the spring and summer seasons at the ENV5 station. This may be attributable to rainfall patterns, or may also be indicative of increased photosynthetic production in Wildcat Creek during the summer months, which can cause decreased dissolved carbon dioxide levels and therefore increased pH readings.

Figure 7: Daily Average pH Data Collected at the ENV5 Station in 2015

3 US Department of the Interior. “An Evaluation of Trends in the Acidity of Precipitation and the Related Acidification of Surface Water in North America.” 1983.

5.8 6 6.2 6.4 6.6 6.8 7

C u m m u la ti ve

F re q u e n cy pH

Cumulative Frequency Distribution ENV5 - Daily Average pH

February 2016 Fort Jackson Environmental Division 11

Figure 8: Discrete pH Data Collected at the ENV5 Station in 2015

3.4 ENV5 – Specific Conductivity

Figures 9 and 10 display the specific conductivity data collected at the ENV5 station during the 2015 monitoring year.

The daily average specific conductivity levels followed a typical distribution, and no clear seasonal trend was noted in the data. However, the specific conductivity levels during the fall season were notably higher than the previous seasons. This appears to have been a result of the October flooding event. The elevated levels may have been a result of the dam breech at Semmes Lake, or due to washoff of significant pollutants during and after the event.

Future monitoring efforts will reveal whether these elevated levels can be expected to persist in the future, or if the ionic strength of Wildcat Creek will return to ambient levels.

5.5

6.5

7.5

8.5

9.5

Winter '15 Spring '15 Summer '15 Fall '15 p H

Seasonal Continuous Monitoring Statistics ENV5 - Discrete pH Data January 2015 - December 2015

50th - 75th Percentile

25th - 50th

February 2016 Fort Jackson Environmental Division 12

Figure 9: Daily Average Specific Conductivity Data Collected at the ENV5 Station in 2015

Figure 10: Discrete Specific Conductivity Data Collected at the ENV5 Station in 2015

0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12

C u m m u la ti ve

F re q u e n cy

Specific Conductivity (mS/cm)

Cumulative Frequency Distribution ENV5 - Daily Average Specific Conductivity

January 2015 - December 2015

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

Winter '15 Spring '15 Summer '15 Fall '15

Sp e ci fi c C o n d u ct iv it y

(m S/ cm

Seasonal Continuous Monitoring Statistics ENV5 - Discrete Specific Conductivity Data

January 2015 - December 2015

50th - 75th Percentile

25th - 50th

February 2016 Fort Jackson Environmental Division 13

3.5 ENV5 – Dissolved Oxygen

Figures 11 and 12 display the DO dataset collected at the ENV5 station during 2015. DO levels were well above SCDHEC minimum standards at this station throughout the year. As Figure 11 demonstrates, the daily average DO level remained well above the SCDHEC daily average minimum standard of 5 mg/L. Figure 12 also indicates that the SCDHEC instantaneous DO minimum was not violated either, with all recorded DO concentrations well above 4 mg/L.

The collected dataset indicates healthy DO levels at the ENV5 station during all seasons of the year.

A distinct seasonal trend is displayed in Figure 12, with higher DO levels in the winter and fall seasons and lower levels in the spring and summer seasons. DO concentrations are influenced by water temperatures. Colder water has a higher DO saturation concentration than warm water. In the fall and winter seasons, the DO concentrations were on average higher than in the spring and summer seasons, likely as a result of the water temperature fluctuations between seasons. DO levels at the ENV5 station may also be impacted by biological activity, which can cause increased DO levels during photosynthetic production and decreased DO levels as a result of biological respiration.

Figure 11: Daily Average Dissolved Oxygen Data Collected at the ENV5 Station in 2015

6 7 8 9 10 11 12

C u m m u la ti ve

F re q u e n cy

Dissolved Oxygen (mg/L)

Cumulative Frequency Distribution ENV5 - Daily Average Dissolved Oxygen

February 2016 Fort Jackson Environmental Division 14

Figure 12: Discrete Dissolved Oxygen Data Collected at the ENV5 Station in 2015

Winter '15 Spring '15 Summer '15 Fall '15

D is so lv e d

O xy ge n m g/ L)

Seasonal Continuous Monitoring Statistics ENV5 - Discrete Dissolved Oxygen Data

January 2015 - December 2015

50th - 75th Percentile

25th - 50th