2014 - 2015 Water Quality Monitoring Report North Fork Gunnison River Basin Delta County, Colorado
Prepared for – Western Slope Conservation Center 204 Poplar Ave. Paonia, Colorado 81428
Prepared by – Colorado Water and Energy Research Center University of Colorado Boulder, Colorado 80309
About the Colorado Water and Energy Research Center:
Colorado’s water and energy resources are tightly linked. The Colorado Water and Energy Research Center (CWERC) at the University of Colorado Boulder studies the connections between water and energy resources and the trade-offs that may be involved in their use. CWERC seeks to engage the general public and policymakers, serving as a neutral broker of scientifically-based information on even the most contentious “energy-water nexus” debates. Since CWERC’s launch in 2011, public interest and pressing scientific questions have steered the Center’s work specifically toward oil and gas extraction, groundwater resources, and the importance of baseline water quality and quantity data.
Colorado Water and Energy Research Center Mission: • • •
Provide neutral, scientific information on important energy and water resources issues for the general public and policymakers. Facilitate the exchange of information and expertise among researchers and regulators working on energy and water resources problems. Assist citizen scientists in environmental monitoring efforts, particularly as they relate to the collection of baseline groundwater quality and quantity data.
How CWERC Developed This Report:
This report was written as a summary of the water quality results gathered in 2014 and 2015 within the North Fork of the Gunnison River Watershed. Water samples were collected following the protocol outlined in the CWERC Guide, entitled Monitoring Water Quality in Areas of Oil and Natural Gas Development: A Guide for Water Well Users. This report is the result of collaboration between CWERC and the Western Slope Conservation Center (WSCC). It aims to strengthen the baseline water quality information that has been collected within the watershed by enhancing our understanding of the groundwater in the area in addition to springs and surface water information collected in past studies.
Authors of this Report: Jessica Lewand, BA, Environmental Studies Katya Hafich, MA, Geography Mark W. Williams, PhD, Geography and Institute of Arctic and Alpine Research
Contact:
[email protected]
ii
Executive Summary
The North Fork River Improvement Association (NFRIA) partnered with local volunteers to create a water quality monitoring network to gather surface water information throughout the North Fork of the Gunnison River watershed from 2001 to 2007, producing reports in both 2006 and 2009 (NFRIA-WSERC, 2009). In 2011, the monitoring network was expanded to include a baseline water quality study, completed by ERO Resources Corporation in Hotchkiss which included a water quality analysis of five surface water sites and three springs within the North Fork Gunnison River watershed (ERO, 2011). This study aims to expand upon the previous studies to monitor groundwater in the North Fork watershed that could be affected by current and proposed extraction of oil and natural gas by incorporating two additional rounds of water quality sampling that were conducted in September of 2014 and June of 2015. The purpose of this study is to provide a baseline of water quality conditions in the North Fork of the Gunnison River watershed to which future water quality conditions could be compared if and when new oil and gas wells are drilled and developed within the watershed. Water samples were tested for an extensive list of water quality parameters developed by the Colorado Water and Energy Research Center at CU Boulder, and include major ions, metals, and volatile organic compounds. Two rounds of water sampling seek to provide a range of concentrations typically seen in this snowmelt-dominated catchment, and measure any seasonal differences that may naturally occur. The following bullet points summarize the results of the monitoring project, which are explained in depth in the full report. • No BTEX compounds (benzene, toluene, ethylbenzene, and xylenes) were detected above the laboratory method detection limit (MDL) throughout the entire study. • Methane was detected in low concentrations in some samples, with many samples below the laboratory method detection limit. o Methane was generally detected during the September sampling event when low flow conditions were present. o Methane can naturally occur in groundwater and it is not hazardous to human health. • Water throughout the watershed can be characterized as having high alkalinity. o Groundwater and springs had significantly higher (p<0.05) alkalinity than surface water. o High alkalinity is likely caused by high carbonate and bicarbonate concentrations due to presence of marine sediments such as the Mancos shale.
•
•
• •
•
•
The pH values range from neutral in groundwater and springs to slightly basic in surface waters. o Surface water had significantly higher (p<0.05) pH than groundwater and springs. The results suggest that barium concentrations may typically be higher in the Muddy Creek area. o There was a significantly higher (p<0.05) concentration of barium in groundwater of Muddy Creek compared to groundwater down valley. o There was also a significantly higher (p<0.05) concentration of barium in surface water of Muddy Creek in the 2011 ERO study compared to the same site in this study. No other significant differences were found when comparing groundwater or surface water sites in Muddy Creek to down gradient sites. No analyte concentrations exceeded the National Primary Drinking Water Standards (NPDWS). A few analyte concentrations exceeded the National Secondary Drinking Water Standards (NSDWS) set in place for contaminants that are not health threatening, but are set to maintain aesthetic considerations, such as taste, color, and odor. No significant seasonal differences (p>0.05) were found for any analyte concentrations when separated by water source (groundwater, springs, and surface water). It is important to note that the sample size for this report is relatively small, and may not capture the full seasonal range, and/or fully characterize water quality throughout the watershed.
iv
Table of Contents Executive Summary ...................................................................................................... ii Introduction ................................................................................................................. 1 Background .................................................................................................................. 2 Site Description ........................................................................................................................................................... 2 North Fork Gunnison River Hydrogeology ..................................................................................................... 4 Surface Water Hydrology of the North Fork .................................................................................................. 6 Methods ...................................................................................................................... 6 Collection of Samples ................................................................................................................................................ 6 Data Analysis ................................................................................................................................................................ 7 Results ......................................................................................................................... 8 Volatile Organic Compounds ................................................................................................................................. 9 Methane .......................................................................................................................................................................... 9 General Water Quality Parameters ................................................................................................................. 10 Major Ions ................................................................................................................................................................... 12 Metals ........................................................................................................................................................................... 15 Comparison to Past Results ................................................................................................................................ 16 Summary and Conclusions ......................................................................................... 17 Bibliography ............................................................................................................... 19 Figures ....................................................................................................................... 22 Tables ........................................................................................................................ 28 Appendix ................................................................................................................... 35
Introduction
The purpose of the study is to provide a baseline of water quality conditions in the
North Fork of the Gunnison River Watershed to which future water quality conditions could be compared if and when new oil and gas wells are drilled and developed within the watershed. This study was initiated by the Western Slope Conservation Center (formerly NFRIA-WSERC) to continue the baseline study initiated by NFRIA-WSERC in 2001. In 2001, the North Fork River Improvement Association (NFRIA) partnered with local volunteers to create the North Fork Volunteer Water Quality Monitoring Network to gather surface water quality information throughout the North Fork of the Gunnison River watershed from 2001 through 2007 (NFRIA-WSERC, 2009). In 2011, the monitoring network was expanded to include a baseline report completed by ERO Resources Corporation in Hotchkiss, and included a water quality analysis of five surface water sites and three springs within the North Fork Gunnison River watershed (ERO, 2011). Sites were chosen to be downstream or down gradient from proposed gas permitting sites. This study utilizes three of the same sites as the ERO 2011 report, and the scope was expanded to include twelve groundwater samples from domestic wells in the North Fork watershed. This continued monitoring of water quality and addition of new sites consisted of two sampling campaigns, one in September of 2014 and one in June of 2015. Water samples were tested for an extensive list of water quality parameters developed by the Colorado Water and Energy Research Center at CU Boulder, which includes major ions, metals, and volatile organic compounds (Table 1). The scope of this study was expanded to include the Muddy Creek drainage, upstream of Paonia Reservoir, where presently there is limited oil and gas development. Since the initial baseline study was conducted, several oil and gas wells have been drilled in the Muddy Creek area. This report aims to summarize the results from the 2014-2015 sampling campaigns conducted in the area on surface water, springs, and groundwater samples. Two rounds of water sampling, in September and June, seek to provide a range of concentrations typically seen in this snowmelt-dominated catchment and any seasonal differences that may naturally occur. Results were also analyzed to detect significant differences between water sources,
1
and for any significant differences between Muddy Creek and downstream locations. It is important to note that this study has a relatively small sample size, and may not capture the complete seasonal range of analyte concentrations, and is not necessarily representative of the entire North Fork Gunnison River watershed. This report was produced through collaboration between the Western Slope Conservation Center (WSCC) and the Colorado Water and Energy Research Center (CWERC), based at the University of Colorado Boulder. The WSCC is a non-profit organization that seeks to create an active and aware community in order to protect and enhance the air, water and wildlife of the Lower Gunnison Watershed. CWERC aims to provide neutral, scientifically based information on important energy and water resources issues for the general public and policymakers. CWERC has also developed protocols for setting up a groundwater and/or surface water monitoring program to acquire baseline data, which were used in this study. The protocols presented in the guide have been reviewed by water quality scientists, Colorado based hydrologists, and industry regulators. This report discusses the average concentrations and range of concentrations measured in the watershed for analytes outlined in the Full Index of analytes from the CWERC Groundwater Monitoring Guide in Table 1 (Kroepsch & Williams, 2014a). These concentrations were compared to the human health standards set forth by the EPA, including the National Primary Drinking Water Standards (NPDWS), and the National Secondary Drinking Water Standards (NSDWS) and the human health standards set forth by the Colorado Department of Public Health and Environment (CDPHE). The concentrations were not compared to standards for aquatic life, fish ingestion, or agriculture.
Background
This section presents an overview of the study area including a general description
of the watershed and the hydrogeology.
Site Description
The North Fork of the Gunnison River watershed is 986 square miles, and is located
on the western slope in the northwestern corner of Gunnison County and eastern Delta
2
County. The study area includes the towns of Bowie, Somerset, Paonia, and Hotchkiss. The distribution of sample sites within the watershed can be seen below in Figure 1. The study includes two surface water, three springs, and twelve groundwater sites to capture a representative overview of the water quality the valley.
Figure 1: The North Fork of the Gunnison Valley watershed. Blue circles represent groundwater sample sites, green circles represent springs sample sites, and red circles represent surface water sites sampled.
3
North Fork Gunnison River Hydrogeology The North Fork Gunnison River watershed contains various geologic formations that include unconsolidated sediments and clastic materials, consolidated sedimentary rock, intrusive volcanic rock, coarse-grained sedimentary rock, and sedimentary and crystalline intrusive igneous rocks (Kolm & van der Heijde, 2013). An aquifer is defined as a saturated formation or group of formations that contain sufficient permeability to yield economic quantities of water to wells and springs (Sterrett, 2007). An aquitard is a geologic formation or group of formations through which virtually no water moves. Unconsolidated Quaternary sediments, consisting of silts, sands and gravels are located throughout the watershed and likely contain the greatest volumes of groundwater, but some of the aquifers within them may be of limited areal extent (Kolm & van der Heijde, 2013). Sandstone aquifers in the area are typically well cemented, and therefore transmit water via fractures. The shallow groundwater aquifers in the Quaternary unconsolidated sediments are a source of drinking and irrigation water for some of the municipalities and households in the North Fork watershed (Kolm & van der Heijde, 2013). In addition to the sandstone units in the area, other consolidated units include shale, claystone, siltstone, and intrusive volcanic rocks; because of low permeability, these units have limited ability to transmit water. These units may restrict the vertical movement of water, acting as aquitards that cause groundwater to perch and discharge to the surface as springs. In some cases, these units may confine underlying aquifers (Kolm & van der Heijde, 2013). The Cretaceous Mancos shale is an example of such a low permeability bedrock formation in the area. The Mancos shale is approximately 4,500 feet thick and acts as a major confining layer (Ackerman & Brooks, 1986; BLM, 2015). For more information on the local hydrogeology please refer to Kolm and van der Heijde (2013), found on the Delta County Environmental Health website to learn more. There are two types of aquifers found within the watershed, alluvial aquifers and multiple sedimentary bedrock aquifers in the southern Piceance Basin. The alluvial aquifers are thickest in the valley bottoms (usually less than 100 feet thick) with groundwater wells that range from 14 to 265 feet deep with an average of 65 feet deep (Ackerman & Brooks, 1986; BLM, 2015). The primary bedrock aquifers within the
4
watershed are the Dakota Sandstone and Burro Canyon Formation (Ackerman & Brooks, 1986; Kolm & van der Heijde, 2013). The depth of this formation varies throughout the watershed, but is generally deeper than 2,000 feet below the land surface (Ackerman & Brooks, 1986). Most potable wells exist in the alluvial aquifer and occur at depths less than 200 feet, but occasionally extend deeper (BLM, 2015). The groundwater wells sampled in this study typically ranged from 30 to 170 feet deep, with an average of 94 feet deep. There are additional secondary bedrock aquifers within the watershed including the Entrada formation at 8,900 feet deep and the Maroon formation at 9,000 to 9,500 feet deep (BLM, 2015). The higher cost of drilling deeper for groundwater is typically not rewarded with higher yields or better water quality (Watts, 2008), making these formations economically infeasible for groundwater supply. However, the deepest permeable formations can be used for deep injection of production wastewater produced by oil and gas development. In this region oil and gas development has recently increased (BLM, 2015). The shaded area in Figure 1 represents the oil and gas units present in the study area. The environmental impact statement of the Bull Mountain Unit describes the potential exploration and development of up to 146 natural gas wells, and 4 water disposal wells in the Muddy Creek area (BLM, 2015), which is in the northeast portion of the map in Figure 1 where sample site W-12 is located. There are currently 15 producing wells in the North Fork area, and the bulk of this production has occurred in the Mesa Verde Group. The primary targets for future development within this group includes the Cozzette and Corcoran Sandstone members within the Mount Garfield (or Iles) Formation at a depth of 4,570 – 5,380 feet (BLM, 2015). There is also potential for additional development of coalbed natural gas that exists in the South Canyon Coal and Cameo Coal units within the Williams Fork Formation (BLM, 2015). The above formations targeted for gas extraction range from 3,270– 4,360 feet deep (Table 2). There is also potential exploration of shale gas sources within the Mancos shale (BLM, 2015) which are typically 4,940 – 5,000 feet deep (Table 2). The groundwater aquifers that are currently used in the North Fork Valley occur at relatively shallow depths (less than 300 feet deep) compared to the depth of oil and gas formations that are being targeted for development. It is important to note that depths of these geologic units vary throughout the watershed. For more information on local
5
stratigraphy please refer to Ackerman and Brooks (1986), and Kolm and van der Heijde (2013).
Surface Water Hydrology of the North Fork
The North Fork of the Gunnison River begins at the confluence of Muddy Creek and
Anthracite Creek, and flows southwest for 33 miles until it joins the Gunnison River. It is a fourth order perennial stream predominantly fed by snowmelt (NFRIA-WSERC, 2009). Its major tributaries include Terror, Hubbard, Minnesota, Roatcap, Cottonwood, and Leroux Creeks. The average flow of the North Fork is approximately 3,000 cubic feet per second (cfs) during spring runoff, while the average flow can be less than 20 cfs in the late summer due to irrigation diversions and low flow conditions (NFRIA-WSERC, 2009). Surface water is the primary sources of irrigation for the area due to lack of a significant groundwater aquifer and little precipitation received during the summer. Sufficient water storage and availability has historically been a concern in the area (BLM, 2015).
Methods
This section describes the methods used to collect samples in the field, the statistical
analysis of the water quality results, and the laboratory analytical methods.
Collection of Samples
Water samples were collected following the protocol outlined in the CWERC Guide,
entitled Monitoring Water Quality in Areas of Oil and Natural Gas Development: A Guide for Water Well Users (Kroepsch & Williams, 2014a). Teams of undergraduate hydrology students from the University of Colorado at Boulder and community volunteers collected samples in September of 2014 and June of 2015. Each water source was tested for 24 of the 26 analytes listed as the Full Index in the CWERC Water Quality Monitoring Guide (Kroepsch & Williams, 2014a). The remaining two parameters are the depth to water (or level of the water table), and stable isotope analysis. Water level was measured at groundwater wells when possible, and stable isotopes of water were not measured. The 26 analytes were selected as a result of a review of the scientific literature, groundwater monitoring regulations from the Colorado Oil and Gas Conservation Commission, and
6
discussions with commercial laboratories that run water quality tests for well users, the oil and gas industry, and regulators. Before sampling, each well was purged following the CWERC Water Monitoring Guide protocol to get a representative sample of the aquifer instead of water residing in the pipes or well casing (Kroepsch & Williams, 2014a). The water was sampled directly from a fixture located between the well and a cistern to avoid any residence time in the cistern and/or onsite treatment. One of the sites, W-13, sampled in June 2015, was collected from a cistern because an upstream water source from a fixture was unavailable between the well and the cistern. All samples were analyzed at Accutest Mountain States Laboratories in Wheat Ridge, Colorado. Accutest provided the materials for sample collection, including sterile sample bottles. Samples were delivered to the lab within 48 hours of sampling. During transport, all water samples were kept on ice and all samples were measured below 6 °C upon arrival at the lab. The laboratory analytical methods for all of the sampled parameters are provided in Table 7.
Data Analysis
Once the data were compiled from both sampling campaigns, they were sorted
based on various criteria including the time of year sampled and water source (surface water, spring, or groundwater). Average and standard deviations were calculated for each analyte concentration within each water source and season. The seasonal data were used in the statistical program R to create box and whisker plots to visualize the distribution of each analyte’s seasonal concentrations and/or concentrations when separated by water source seen in Figures 3 – 10 and Figure 12 (R Core Team, 2013). In the plot a black line in the center of the box represents the median, while the bottom and top of the box represents the 25% and 75% quartile respectively. The whiskers represent 1.5 times the length of the box, and any circles above or below represents outlying values. A t-test was used to determine if there was a significant difference between these seasonal concentrations when separated by water source (Table 4); as well as for a comparison of
7
the water sources to determine if significant differences exist between the analyte concentrations in surface water, springs, or groundwater regardless of season (Table 5).
Many of the analyte concentrations were below the laboratory reporting limit or the
method detection limit. Accutest defines reporting limits (RL) as the lowest concentration standard in the calibration range of each compound analyzed, which is also the low limit unqualified quantitative data, or low limit at which target analytes are reported. The method detection limit (MDL) is defined as the lowest concentration of each compound that can be qualitatively identified by the method in use, which is determined via experimentation and verified through additional testing. The MDL is often much lower than the RL and it is a statistical calculation. Accutest utilizes a RL for each analyte tested in this study, and uses an MDL when determining the concentrations of methane and BTEX compounds. When an analyte is below the laboratory reporting limit or method detection limit, it does not necessarily mean that it is absent, but may mean the concentration is too low to detect. Because the statistical program R does not recognize values with a less than sign (<), to create the box and whisker plots all values that were below reporting limits were changed to concentrations equal to half of the reporting limit (Oblinger Childress et al., 1999). For the BTEX compounds, arsenic, chromium, lead, and selenium, every measurement was below the laboratory reporting limit. In these cases, statistical analyses were not performed, but it is important to note their absences or very low concentrations in the baseline samples. Comparisons of the laboratory reporting or method detection limit to the EPA standard for these analytes can be seen in Figures 11 & 12.
Results
This section presents a broad overview of the results gathered during the baseline
sampling campaigns, and discusses important water quality characteristics found within the North Fork Gunnison River watershed. The average concentrations found in groundwater and surface water samples for each analyte can be found in Table 3. The analytes are categorized as volatile organic compounds (VOCs), methane, general water quality parameters, major ions, and metals. The full data set can be found in Table 7.
8
Volatile Organic Compounds Water sources were tested for volatile organic compounds (VOCs) known as BTEX, (benzene, toluene, ethylbenzene, and xylenes). These compounds are present in gasoline, are used in the production of some consumer products and industrial solvents, and can be naturally occur in shales and other geologic formations (Kroepsch & Williams, 2014b). BTEX compounds can have negative health impacts after long term exposure to concentrations above the National Primary Drinking Water Standards (NPDWS), so it is important to monitor them (US EPA, 2015). The NPDWS for benzene, toluene, ethylbenzene, and xylene are 0.005 (mg/L), 1.0 (mg/L), 0.7 (mg/L), and 10.0 (mg/L) respectively. All samples collected during this project found BTEX concentrations to be below the laboratory method detection limits: 0.0002 mg/L, 0.0002 mg/L, 0.0002 mg/L, and 0.00046 mg/L respectively, which are all well below the EPA drinking water standards (Figure 2).
Methane
All samples were analyzed for methane concentrations. Methane can naturally occur in groundwater in two different forms, known as biogenic and thermogenic methane. Biogenic methane is created by anaerobic bacterial decomposition of organic material at shallow depths and low temperatures while thermogenic methane is associated with deep formations under high pressure and high temperatures (King 2012; Osborn et al. 2011). Thermogenic methane could naturally occur in groundwater as a seep from a deeper formation that has been connected over millions of years, and is more common where shale and coal-bearing formations interact with groundwater (King, 2012). In order to differentiate between the two types of methane one would need an isotopic analysis, which was not conducted in this study. Methane is not hazardous to human health, but the EPA recommends venting domestic wells once concentrations are near 10 mg/L because of the explosion hazard (Kroepsch & Williams, 2014b). The existing methane concentrations in the samples had an average of 0.697 ± 1.465 mg/L, with a maximum concentration of 4.26 mg/L found in a groundwater sample (Table 3). The majority of the samples had concentrations below the laboratory method detection limit and in general the detected concentrations were present in September (except for one
9
sample) when low flow conditions were present. Concentrations of methane did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found. The overall average in September was 0.291±1.06 mg/L, and the average in June was 0.035±0.129 mg/L. Methane concentrations in groundwater found in previous studies suggest that it is likely generated in underlying rocks (BLM, 2015; NFRIA-WSERC, 2010).
General Water Quality Parameters
The general water quality parameters tested for in this study were alkalinity, pH, specific conductance, and dissolved organic carbon. Alkalinity is defined as a waters’ ability to neutralize acids and buffer large changes in pH (Kroepsch & Williams, 2014b; US EPA, 2012). Results are reported as milligrams per liter of calcium carbonate (mg/L CaCO3). Strongly alkaline water has an objectionable “soda” taste and can dry out a person’s skin. The average concentration was 263.63 ± 88.72 mg/L, with range of 70.7 to 386.0 mg/L (Table 3). Alkalinity is not regulated in drinking water by the EPA or the CDPHE, but there is a recommended range for drinking water of 30 to 400 mg/L according to the Illinois Department of Public Health (Kroepsch & Williams, 2014b). All samples collected during this baseline study fell within this recommended range. There was no significant seasonal difference (p-value>0.05) found in alkalinity concentrations when separated by water source (Table 4 & 3A). The average alkalinity was 273.65 ± 77.01 mg/L in September and 253.60 ± 98.04 mg/L in June. However, there was a significantly (p<0.05) higher alkalinity concentration in surface water compared to groundwater, as well as compared to springs (Table 5, Figure 3B). The pH measured throughout the watershed averaged 7.64 ± 0.31, with a range of 7.14 to 8.48. It is important to note that the pH was not measured in the field, so some changes in pH may have occurred on the way to the lab. As seen in Table 3, the groundwater and springs were generally neutral, and the surface waters were slightly basic on average. This is supported by previous studies where the water ranged from neutral to slightly basic (ERO, 2011; NFRIA-WSERC, 2009; NFRIA-WSERC, 2010). There was a
10
significantly (p<0.05) higher pH in surface water compared to groundwater, as well as compared to springs (Table 5 & Figure 4B). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4 & Figure 4A). The average pH was 7.61 ± 0.34 mg/L in September and 7.66 ± 0.26 mg/L in June. The CDPHE human health standard for pH is 5.0 to 9.0. Values outside of this range may damage household plumbing (Kroepsch & Williams, 2014b). There was a significant difference (p-value<0.05) found between surface water and springs, as well as surface water and groundwater for both alkalinity and pH, which suggests that groundwater and springs have the same origin, while surface waters may originate from a different source or interact with different geology. Specific conductivity is a measure of water’s ability to conduct an electrical current, and it is an indication of the amount of inorganic dissolved solids in the water, as well as the salinity of the water. This parameter is not regulated in drinking water by the EPA or the CDPHE (Kroepsch & Williams, 2014b). The average specific conductance measured in this study was 726.1 ± 527.2 μS/cm, with a range of 147 to 2,680 μS/cm (Table 3); this large range in conductance was also found in a previous study (ERO, 2011). Concentrations of specific conductance did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 5B). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4 & Figure 5A). The average specific conductivity was 706.267 ± 444.247 mg/L in September and 745.933 ± 598.12 mg/L in June. The few high specific conductance concentrations in groundwater are associated with high sodium and sulfate concentrations (Table 7). The last general water quality parameter sampled for was dissolved organic carbon (DOC), which is a broad classification of organic molecules in water; this parameter is not regulated in drinking water by the EPA or the CDPHE. High DOC concentrations can be attributed to septic or fertilizer leakage, as well as rock formations containing large amounts of organic carbon (Kroepsch & Williams, 2014b). The concentration measured in the samples had averaged 2.60 ± 3.73 mg/L, with a range of 0.5 to 21.7 mg/L. The average dissolved organic carbon was 3.033 ± 5.057 mg/L in September and 2.16 ± 1.36 mg/L in June. Concentrations of dissolved organic carbon did not significantly differ (p-value>0.05)
11
when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 6B). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4 & Figure 6A).
Major Ions
There were eight major ions tested for in this study: calcium, chloride, fluoride,
magnesium, potassium, sodium, sulfate, and nitrate (Table 1). Most of these ions are commonly detected in groundwater due to local geology and/or human activities. Both chloride and fluoride ions were detected throughout the study in all water sources and all concentrations were below the EPA and CDPHE drinking water standards (Table 3). Chloride had an average concentration of 10.75 ± 22.98 mg/L with a range of 0.5 to 117 mg/L, and fluoride had an average concentration of 0.44 ± 0.28 mg/L with a range of 0.11 to 1.3 mg/L. Concentrations of both analytes did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 7). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4). The average chloride concentration was 9.073 ± 16.068 mg/L in September and 12.422 ± 28.160 mg/L in June. The average fluoride concentration was 0.445 ± 0.240 mg/L in September and 0.442 ± 0.306 mg/L in June. The two main cations that cause water hardness are calcium (Ca2+) and magnesium (Kroepsch & Williams, 2014b). Calcium and magnesium may leach into water from calcareous sandstone and shale deposits present in the North Fork area (Kolm & van der Heijde, 2013). The EPA and the CDPHE do not regulate either of these ions in drinking water. The average calcium concentration was 74.7 ± 58.7 mg/L, with a range of 15.9 to 242.0 mg/L (Table 3). The average magnesium concentration was 44.3 ± 46.5 mg/L, with a range of 4.26 to 231.0 mg/L (Table 3). The average calcium concentration was 72.72 ± 55.08 mg/L in September and 76.75 ± 62.08 mg/L in June, while the average magnesium concentration was 43.15 ± 38.56 mg/L in September and 45.53 ± 53.30 mg/L in June. Previous studies have also observed high concentrations of calcium and magnesium, and that the hardness of water in the watershed fluctuates between moderate to very hard throughout the year (ERO, 2011; NFRIA-WSERC, 2009). Concentrations of both analytes
12
did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 8). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4). Potassium is a common ion found in drinking water and is not regulated by the EPA or the CDPHE. The average concentration was 6.78 ± 7.04 mg/L with a range of 1.07 to 20.8 mg/L (Table 3). Concentrations of potassium did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 7). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4). The average potassium concentration was 6.72 ± 6.93 mg/L in September, and 6.42 ± 7.08 mg/L in June. Another common ion found in drinking water is sodium, which is not regulated by the EPA or the CDPHE; however, there is a recommended upper limit of 20.0 mg/L for people on a low sodium diet recommended by the EPA (Kroepsch & Williams, 2014b). In the samples collected, the average concentration was 36.26 ± 29.40 mg/L, with a range of 3.94 to 125.00 mg/L (Table 3). A similar range in concentrations were detected in the previous study, with higher concentrations typically in groundwater sources (ERO, 2011). Concentrations of sodium did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 7). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4). The average sodium concentration was 38.425 ± 26.371 mg/L in September, and 34.089 ± 31.991 mg/L in June. Sulfate is another common ion found in groundwater. Sources include atmospheric deposition, sulfate mineral dissolution, sulfide mineral oxidation, coal mines and power plants. The CDPHE drinking water standard and the NSDWS set by the EPA are both 250 mg/L; this is a secondary standard are set to maintain aesthetic considerations, such as taste, color, and odor. High sulfate concentrations can lead to a salty taste and diarrhea (US EPA, 2015c). In the samples collected, the average sulfate concentration was 171.00 ± 314.14 mg/L, with a range of 3.3 to 1,370 mg/L (Table 3). Concentrations of sulfate did not
13
significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 9B). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4 & Figure 9A). The average concentration in September was 163.68 ± 278.52 mg/L, and the average in June was 178.32 ± 345.95 mg/L. Some sulfate concentrations were above the secondary standard, which may be characteristic of the area because elevated sulfate concentrations were also found in previous studies (ERO, 2011; NFRIA-WSERC, 2009). In this study nitrogen was tested in the form of nitrate and nitrite; however, nitrite concentrations were very small (Table 3). Nitrogen is an important ion to look for in the baseline because it can help differentiate between potential contamination sources. This ion is more likely to come from agricultural operations or leakage of septic tanks than oil and gas (Kroepsch & Williams, 2014b). Farming is widespread in the valley and nitrogen is a major constituent of fertilizer, which can enter groundwater sources from surface infiltration (Kolm & van der Heijde, 2013). If there was a future increase in this ion concentration it may indicate that the contamination is unlikely to originate from oil and gas operations. Nitrogen sources may include erosion of some geologic deposits, liquid waste from septic tanks, and livestock manure (Kroepsch & Williams, 2014b; NFRIAWSERC, 2009). High nitrate concentrations can be harmful to the health of expecting mothers and infants; for this reason, the NPDWS and the CDPHE human health standards are set at 10 mg/L. All concentrations measured within the watershed were well below this standard. The average concentration within the watershed for nitrate was 0.763 ± 0.95 mg/L, with a range of 0.005 to 4.2 mg/L (Table 3). The average concentration in September was 0.635 ± 0.821 mg/L, and the average concentration in June was 0.883 ± 1.045 mg/L. Concentrations of nitrate did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 10B). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4 & Figure 10A). Previous studies found elevated total nitrogen concentrations in the winter months during low flow conditions (NFRIAWSERC, 2009).
14
Metals Concentrations of nine dissolved metals were measured during this sampling campaign. All concentrations of arsenic, chromium, lead and selenium were below the laboratory reporting limit. Figure 11 shows the comparison of the reporting limit and the EPA standard for these metals. Detectable concentrations of barium and boron were present in many of the samples, but all detections were below the standards set by the EPA and CDPHE (Table 3). Barium had an average concentration of 84.44 ± 76.32 μg/L with a range of 19.0 to 345 μg/L, and boron had an average concentration of 63.22 ± 55.04 μg/L with a range of 25 to 254 μg/L. Concentrations of both barium and boron did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 12). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4). There was a significantly higher (p<0.05) concentration of barium in groundwater of Muddy Creek compared to groundwater down valley, but still well below the EPA standard of 2,000 μg/L. Copper concentrations present in the samples were generally very low. All copper concentrations for surface waters and springs were below the reporting limit, while the average groundwater concentration was 21.83 ± 8.94 μg/L, with a range of 16.8 to 37.3 μg/L. Concentrations of copper did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 12). Samples from each source were also analyzed for seasonal differences and no significant differences (p>0.05) were found (Table 4). All copper concentrations were well below the CDPHE drinking water standard of 1,000 μg/L. Iron is soluble and pervasive in the environment and is commonly found in groundwater. The NSDWS and CDPHE secondary drinking water standard for dissolved iron is 300 μg/L, which is for taste, color and odor. The average concentration of dissolved iron was 109.01 ± 181.14 μg/L with a range of 11.9 to 641 μg/L. Most of the concentrations were less than the EPA’s secondary standards, except for a few concentrations found in groundwater that exceeded the secondary standard. High iron concentrations in water may
15
result in a rusty color, sediment buildup, metallic taste, and/or a reddish or orange staining (US EPA, 2015c). Iron was detected at seven sampled sites in September, and at three sites in June; this suggests that more dilute conditions may exist during spring snowmelt. Despite the concentration range and difference in detections in September and June, there was no significant difference (p-value>0.05) in seasonal iron concentrations (Table 4). Concentrations of iron did not significantly differ (p-value>0.05) when separated by source (groundwater, surface water, and springs) (Table 5 & Figure 12). Manganese is soluble and pervasive in the environment and commonly found in groundwater due to weathering of rocks, minerals, and soils. Manganese has a NSDWS and a CDPHE drinking water standard of 50 μg/L. The average concentration of dissolved manganese was 101.28 ± 150.54 μg/L with a range of 6.9 to 602 μg/L. There were six water samples with manganese concentrations that exceeded the secondary standards. High manganese concentrations are not hazardous to human health, but may have aesthetic effects such as a black or brown color in water, black staining on fixtures, or a bitter metallic taste (US EPA, 2015c).Concentrations of manganese did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 12). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4). The final metal tested for in this study was strontium, which is not currently regulated by the EPA. The average concentration was 780 ± 503 μg/L with a range of 201 μg/L to 2,290 μg/L. Concentrations of strontium did not significantly differ (p-value>0.05) when samples were separated by source (groundwater, surface water, and springs) (Table 5 & Figure 12). Samples from each source were also analyzed for seasonal differences, and no significant differences (p>0.05) were found (Table 4).
Comparison to Past Results This study shared three of the same sample sites as the 2011 baseline study conducted by ERO, including two springs (SP-01=Cave Spring, SP-02=Domestic Pipeline Spring) and one surface water location (SW-02=Muddy Creek). For the surface water location East and West Muddy Creek samples from the ERO study were averaged for
16
comparison to the Muddy Creek sample below the confluence in this study. This comparison was conducted to determine if any significant differences in analyte concentrations were found over time (Table 6). It is important to note that sample sizes for this comparison were extremely low, and each study tested for a slightly different index of analytes. There was a significantly higher sulfate concentration and sodium concentration at SP-01 (Cave Spring) in the 2011 study compared to concentrations found in the 2014 - 2015 study (p-value<0.05). There was a significantly higher barium concentration at SW02 (Muddy Creek) in the 2011 study compared to concentrations in the 2014 - 2015 study (p-value<0.05). Finally there was a significantly higher methane concentration at SP-01 (Cave Spring) in the 2014 - 2015 study compared to the concentrations found in the 2011 study, but a significantly higher methane concentration at SP-02 (Domestic Pipeline Spring) in the 2011 study compared to the 2014 - 2015 study. Both methane results had a p-value equal to 0.0 (Table 6), which is likely caused by multiple orders of magnitude difference in the results since the laboratory in each study utilized different method detection limits. Overall the analyte concentrations in each study remained very similar over time.
Summary and Conclusions Current water quality conditions in the water samples from North Fork watershed meet primary drinking water standards, and generally meet secondary drinking water standards. There were no sampled analytes that exceeded primary drinking water standards, but there were some sulfate, iron, and manganese concentrations that exceeded the secondary drinking water standards. All of the other sampled analytes were either under the laboratory reporting limit, or are not regulated by the EPA or CDPHE. The full index of analytes tested for within this study is listed in Table 1, the corresponding average concentrations and range of values are provided in Table 3, and the full data set is provided in Table 7. It is important to note that the sample size for this report is relatively small, and may not capture the full seasonal range of possible concentrations, and is not necessarily representative of the entire North Fork Gunnison River watershed. The findings suggest no
17
significant seasonal differences (p>0.05) exist between any of the analyte concentrations investigated in this study (Table 4). There was a significant difference (p-value<0.05) found between surface water and springs, as well as surface water and groundwater for both alkalinity and pH, which may suggest that groundwater is the main source of water to the springs in the North Fork watershed (Table 5). There was a significantly higher barium concentration in Muddy Creek groundwater compared to groundwater concentrations down valley, but still well below the NPDWS. No other significant differences were found between the analyte concentrations in Muddy Creek groundwater or surface water sites, close to existing oil and gas development (Figure 1), compared to concentrations down valley. The high barium concentration in the Muddy Creek area was also found during the comparison to past results, where the surface water in 2011 had a significantly higher concentration than surface water (SW-02) in the 2014 – 2015 study. This suggests high barium concentrations may be typical in this area of the watershed. The purpose of this report is to provide the North Fork watershed with a general picture of current water quality, to which future conditions can be compared. A general rule of thumb for future samples is to look for analyte concentrations that increase by 20 percent or more above the high end of the seasonal range (Kroepsch & Williams, 2014b). Future research on water quality in the area would benefit from continued sampling to increase understanding of variations in water quality that occur naturally in the area, and an analysis of isotopic signatures of methane where methane is detected.
18
Bibliography
Ackerman, D. J., & Brooks. (1986). Reconnaissance of ground-water resources in the North Fork Gunnison River Basin, southwestern Colorado. Dept. of the Interior, U.S. Geological Survey. BLM, US DOI (2015). Draft Environmental Impact Statement for the Bull Mountain Unit Master Development Plan (No. DOI-BLM-CO-S050-2013-0022-EIS). Uncompahgre Field Office. Retrieved from http://www.blm.gov/co/st/en/BLM_Information/ nepa/ufo/Bull_Mountain_EIS.html ERO. (2011). 2011 Oil and Gas Development Baseline Water Quality Monitoring Report - North Fork Gunnison River Basin Delta County, Colorado. Retrieved from http://theconservationcenter.org/wp-content/uploads/2013/04/2011-NFGWater-Quality-Monitoring-Rpt-SINGLEDOC.pdf King, G. E. (2012). Hydraulic Fracturing 101: What Every Representative, Environmentalist, Regulator, Reporter, Investor, University Researcher, Neighbor and Engineer Should Know About Estimating Frac Risk and Improving Frac Performance in Unconventional Gas and Oil Wells. Society of Petroleum Engineers. http://doi.org/10.2118/152596-MS Kolm, K. E., & van der Heijde, P. K. (2013). Groundwater Systems in Delta County, Colorado: North Fork Valley and Terraces Area (GIS-Based Hydrological and Environmental Systems Analysis and Formulation of Conceptual Site Models No. Contract # 2013CT-027). Delta County Board of County Commissioners, Colorado. Kroepsch, A., & Williams, M. W. (2014a). Monitoring Water Quality in Areas of Oil and Natural Gas Development: A Guide for Water Well Users | Colorado Water and
19
Energy Research Center | CWERC | cwerc.colorado.edu. Retrieved December 25, 2014, from http://cwerc.colorado.edu/ Kroepsch, A., & Williams, M. W. (2014b). Groundwater Quality Interpretation Guide: Understanding your Water Quality Results | Colorado Water and Energy Research Center | CWERC | cwerc.colorado.edu. Retrieved from http://cwerc.colorado.edu/docs/cwerc_interpretation_guide.pdf NFRIA-WSERC. (2009). North Fork River Improvement Association: Volunteer Water Quality Monitoring Network October 2004 - October 2007 Data Report. Retrieved from http://www.theconservationcenter.org/wp-content/uploads/2014/06/NFRIAFINAL-REPORT.pdf NFRIA-WSERC. (2010). North Fork of the Gunnison River Watershed Plan Update (North Fork River Improvemenet Association (NFRIA)). Retrieved from http://npscolorado.com/wp-content/supportfiles/North%20Fork%20of%20the%20Gunnison%20River%20Watershed%20Plan %20Update%20-%20NFRIA.pdf Oblinger Childress, C., Foreman, W., Connor, B., & Maloney, T. (1999). New Reporting Procedures Based on Long-Term Method Detection Levels and Some Considerations for Interpretations of Water-Quality Data Provided by the U.S. Geological Survey National Water Quality Laboratory (Open-File Report No. 99-193). Reston, Virginia: U.S. Geological Survey. Retrieved from http://nwql.usgs.gov/rpt.shtml?OFR-99-193 Osborn, S. G., Vengosh, A., Warner, N. R., & Jackson, R. B. (2011). Methane contamination of drinking water accompanying gas-well drilling and hydraulic fracturing. Proceedings
20
of the National Academy of Sciences, 108(20), 8172–8176. http://doi.org/10.1073/pnas.1100682108 Sterrett, R. J. (2007). Groundwater and wells (Third Edition). New Brighton, MN: Johnson Screens. Topper, R., Karen Spray, Cunningham, W., Hamilton, J., & Barkmann, P. (2003). Gound Water Atlas of Colorado (Special Publication 53). Colorado Geological Survey. US EPA. (2015a). 5.10 Total Alkalinity. Retrieved August 4, 2015, from http://water.epa.gov/type/rsl/monitoring/vms510.cfm US EPA. (2015b). National Primary Drinking Water Regulations. Retrieved July 31, 2015, from http://yosemite.epa.gov/water/owrcCatalog.nsf/SingleKeyword?Openview&Keyw ord=National+Primary+Drinking+Water+Regulations&count=2000 US EPA. (2015c). Secondary Drinking Water Regulations: Guidance for Nuisance Chemicals. Retrieved March 24, 2015, from http://water.epa.gov/drink/contaminants/secondarystandards.cfm US EPA. (2014). 10/20/2014: EPA Makes Preliminary Determination to Regulate Strontium in Drinking Water. Retrieved March 24, 2015, from http://yosemite.epa.gov/opa/admpress.nsf/6427a6b7538955c585257359003f023 0/327f339e63facb5a85257d77005f4bf9!OpenDocument Watts, K. R. (2008). Availability, sustainability, and suitability of ground water, Rogers Mesa, Delta County, Colorado—types of analyses and data for use in subdivision watersupply reports (Investigations Report No. 2008 - 5020) (p. 53 p.). U.S. Geological Survey. Retrieved from http://pubs.usgs.gov/sir/2008/5020/
21
Figures
BTEX Compounds: Comparison of MDL to NPDWS 12
10
Concentra3on (mg/L)
10
8
Benzene Toluene
6
Ethylbenzene Xylene
4
2 1
0
0.7
0.005
0.0002 0.0002 0.0002 0.00046
Method Detec3on Limit
Primary Standard
Figure 2: Comparison between the laboratory method detection limit and the National Primary Drinking Water Standard (NPDWS) set by the EPA for BTEX compounds (Benzene, Toluene, Ethylbenzene, and Xylene). All detection limits were well below the NPDWS. Alkalinity, Total as CaCO3: Seasonal Differences
Alkalinity, Total as CaCO3: Seasonal Differences Between Water Sources Between Water Types Alkalinity, Total as CaCO3
250 200
Concentration (mg/L)
300
350
Water Sources Groundwater Groundwater Springs Springs Surface Water Surface Water
100
150
300 250 200 150 100
Concentration (mg/L)
350
Sampling Events Fall – September 2014 September 2014 Spring – June 2015 June 2015
Groundwater Springs GW−Fall SP−Fall
Surface Water SW−Fall
GW
SP
SW
Sources of Water
A B Figure 3: A. This plot shows seasonal differences for alkalinity in each water source. Sample size is n=10 (GW-Sept), n=11 (GW-June), n=3 (SP-Sept), n=2 (SP-June), n=2 (SW-Sept), and n=2 (SW-June). B. This plot shows a significant difference (p-value<0.05) between SW and GW, as well as SW and SP. Sample size is n=21 (GW), n=5 (SP), and n=4 (SW).
22
Sampling Events Fall – September 2014 September 2014 Spring – June 2015 June 2015 8.4
pH
8.2 7.8
8.0
Water Sources Groundwater Groundwater Springs Springs Surface Water Surface Water
7.2
7.2
7.4
7.6
pH (standard units)
8.0 7.8 7.6 7.4
pH (standard units)
8.2
8.4
pH: Seasonal Differences Between Water Types
Groundwater Springs GW−Fall SP−Fall
Surface Water SW−Fall
GW
SP
SW
Sources of Water
A B Figure 4: A. This plot shows the seasonal differences for pH in each water source. Sample size is n=10 (GWSept), n=11 (GW-June), n=3 (SP-Sept), n=2 (SP-June), n=2 (SW-Sept), and n=2 (SW-June). B. This plot shows a significant difference (p-value<0.05) between SW and GW, as well as SW and SP. Sample size is n=21 (GW), n=5 (SP), and n=4 (SW). Specific Conductivity: Seasonal Differences Between Water Types
Specific Conductivity
Surface Water SW−Fall
2000 1500 1000 500
Specific Conductivity (uS/cm)
2000 1500 1000
Groundwater Springs GW−Fall SP−Fall
Water Sources Groundwater Groundwater Springs Springs Surface Water Surface Water
2500
Sampling Events Fall – September 2014 2014 Fall − September Spring – June 2015 2015 Spring − June
500
Specific Conductivity (uS/cm)
2500
Specific Conductivity: Seasonal Differences Between Water Types
GW
SP
SW
Types of Water
A
B Figure 5: A. This plot depicts the seasonal differences in specific conductivity in each water source. Sample size is n=10 (GW-Sept), n=11 (GW-June), n=3 (SP-Sept), n=2 (SP-June), n=2 (SW-Sept), and n=2 (SW-June). B. This plot shows specific conductance in each source of water. Sample size is n=21 (GW), n=5 (SP), and n=4 (SW).
23
Dissolved Organic Carbon (DOC):
Dissolved Organic Carbon (DOC): Seasonal Differences Between Water Types Dissolved Organic Carbon (DOC) Seasonal Differences Between Water Types
15
20
Water Sources Groundwater Groundwater Springs Springs Surface Water Surface Water
0
0
5
10
Concentration (mg/L)
15 10 5
Concentration (mg/L)
20
Sampling Events Fall – September 2014 2014 Fall − September Spring – June 2015 2015 Spring − June
Groundwater Springs GW−Fall SP−Fall
Surface Water SW−Fall
GW
SP
SW
Types of Water
A B Figure 6: A. This plot depicts the seasonal differences of dissolved organic carbon concentrations in each water source. Sample size is n=10 (GW-Sept), n=11 (GW-June), n=3 (SP-Sept), n=2 (SP-June), n=2 (SW-Sept), and n=2 (SW-June). B. This plot shows DOC concentrations in each source of water. Sample size is n=21 (GW), n=5 (SP), and n=4 (SW).
Major Ions Water Sources Groundwater Groundwater Springs Springs Surface Water Surface Water
120
Concentration (mg/L)
100 80 60 40 20
di
Sodium−SW
um Sodium−SP
Sodium−GW So
siu
Potassium−SW
Potassium−SPm
Po
ta Potassium−GW s
or
Fluoride−SW
ide Fluoride−SP
Flu Fluoride−GW
e
Chloride−SW
h
lor Chloride−SP id
Chloride−GW C
0
Figure 7: This plot shows the chloride, fluoride, potassium, and sodium concentrations within each water source. Sample size is n=21 (GW), n=5 (SP), and n=4 (SW) for each analyte.
24
Calciumof&Water Magnesium Main Cations Hardness 250
Water Sources Groundwater Groundwater Springs Springs Surface Water Surface Water
Concentration (mg/L)
200
150
100
Magnesium−SP
Magnesium−SW
Magnesium
Magnesium−GW
Calcium Calcium−SP
Calcium−GW
0
Calcium−SW
50
Figure 8: This plot shows the calcium and magnesium concentrations in each water source. Sample size is n=21 (GW), n=5 (SP), and n=4 (SW) for each cation.
Sampling Events Fall – September 2014 2014 Fall − September Spring – June 2015 2015 Spring − June
800 1000 600
Concentration (mg/L)
600 400
Water Sources Groundwater Groundwater Springs Springs Surface Water Surface Water
Groundwater Springs GW−Fall SP−Fall
Surface Water SW−Fall
0
0
200
400
800 1000
1400
Sulfate
200
Concentration (mg/L)
1400
Sulfate: Seasonal Water Types Sulfate: SeasonalDifferences DifferencesBetween Between Water Types
GW
SP
SW
Types of Water
A B Figure 9: A. This plot depicts the seasonal differences of sulfate concentrations in each water source. Sample size is n=10 (GW-Sept), n=11 (GW-June), n=3 (SP-Sept), n=2 (SP-June), n=2 (SW-Sept), and n=2 (SW-June). B. This plot shows sulfate concentrations in each source of water. Sample size is n=21 (GW), n=5 (SP), and n=4 (SW).
25
Nitrate: DifferencesBetween BetweenWater Water Types Nitrate: Seasonal Seasonal Differences Types Nitrate Water Sources Groundwater Groundwater Springs Springs Surface Water Surface Water
3 0
0
1
2
Concentration (mg/L)
3 2 1
Concentration (mg/L)
4
4
Sampling Events Fall – September 2014 2014 Fall − September Spring – June 2015 2015 Spring − June
Groundwater Springs GW−Fall SP−Fall
Surface Water SW−Fall
GW
SP
SW
Sources of Water
A B Figure 10: A. This plot depicts the seasonal differences in nitrate concentrations in each water source. Sample size is n=10 (GW-Sept), n=11 (GW-June), n=3 (SP-Sept), n=2 (SP-June), n=2 (SW-Sept), and n=2 (SWJune). B. This plot shows nitrate concentrations in each source of water. Sample size is n=21 (GW), n=5 (SP), and n=4 (SW).
Dissolved Metals: Comparison of RL to NPDWS 0.12
0.1
Concentra&on (mg/L)
0.1
0.08
Arsenic Chromium
0.06 0.05
0.05
0.05
0.05
Lead Selenium
0.04 0.025 0.02 0.01
0.01
0
Repor&ng Limit
Primary Standard
Figure 11: Comparison between the laboratory reporting limit and the National Primary Drinking Water Standard (NPDWS) set by the EPA. Note that for arsenic the reporting limit is greater than the NPDWS. Analysis for arsenic is expensive, and not all labs have the capability to detect it at low concentrations. Due to a limited budget, arsenic was not tested for at another lab. Both lead and selenium have a reporting limit equal to the NPDWS.
26
Dissolved Metals 2.5 1.5
0.4
1.0
0.3
0.5
0.2
Concentration (mg/L)
2.0
0.5
Water Sources Water Sources Groundwater Groundwater Groundwater Springs Springs Springs Surface Water Surface Water Surface Water
0.1
Ba riu Barium (GW) Barium (SP)m Barium (SW) Bo Boron (GW) ro Boron (SP)n Boron (SW) Co Copper (GW) pp Copper (SP)er Copper (SW) Iron (GW) Iro n Ma (SP) Iron Ironn(SW) ga ne Manganese (GW) s Manganese (SP)e Manganese S (SW) ro Strontium t(GW) nt ium Strontium (SP) Strontium (SW)
0.0
0.0
Concentration (mg/L)
0.6
Figure 12: This plot shows in the dissolved barium, boron, copper, iron, manganese, and strontium concentrations in each water type. Sample size is n=21 (GW), n=5 (SP) and n=4 (SW) for each analyte. The secondary y-axis refers to the strontium concentration.
27
Tables Table 1: Dissolved parameters tested in the North Fork watershed samples collected in September 2014 and June 2015 (Kroepsch & Williams, 2014a).
General Water Quality
Major Ions
Metals
Gas
Full Index Alkalinity Conductance pH Dissolved Organic Carbon Calcium Chloride Fluoride Magnesium Potassium Sodium Sulfate Nitrate + Nitrite (Total) Arsenic Barium Boron Chromium Copper Iron Lead Manganese Selenium Strontium Methane
Volatile Organic Compounds
BTEX compounds (Benzene, Toluene, Ethylbenzene, Xylene)
Table 2: Estimated depth and thickness of target formations in the current draft of the Bull Mountain Unit EIS (BLM, 2015). It is important to note that the depth of these formations varies throughout the watershed. These depths are specific to the Bull Mountain Unit/Muddy Creek area, and may not be representative of the depths of these formations down valley.
Name South Canyon Cameo Coal Cozzette – Corcoran Mancos Maroon (primary disposal target) Entrada (possible disposal target)
Depth (feet) 3,270 – 3,870 3,760 – 4,360 4,570 – 5,380 4,940 – 5,000 9,300 8,900
Thickness (feet) 100 100 300 3,000 100 200
28
Table 3: Averages and range of concentrations for each analyte tested, average values for each water source (groundwater (GW), springs (SP), and surface water (SW)) are listed concurrently. Standards or recommended limits provided by the EPA and CDPHE are listed for comparison. Analyte Average Range GW-Avg. SP-Avg. SW-Avg. Standard or (n=30) (n=21) (n=5) (n=4) Recommended Limit National Primary Drinking Water Standards BTEX: Benzene
ND
ND
ND
ND
ND
Benzene: 0.005 mg/L
Toluene
ND
ND
ND
ND
ND
Toluene: 1.0 mg/L
Ethylbenzene
ND
ND
ND
ND
ND
Ethylbenzene: 0.7 mg/L
Xylene
ND
ND
ND
ND
ND
Xylene: 10.0 mg/L
Methane (mg/L)
Alkalinity, Total as CaCO3 (mg/L)
Chloride (mg/L)
0.697 ±1.465
263.63 ±88.72
10.75 ±22.98
0.00044 – 4.26
70.7 – 386
0.5 – 117
0.965 ±1.659
286.73 ±73.06
13.57 ±26.89
0.059 ±0.0
280.20 ±39.94
5.02 ±3.60
0.00044 ±0.0
121.60 ±75.51
3.06 ±2.78
U.S. Department of the Interior recommends venting of water wells that contain greater than 10 mg/L of dissolved methane to minimize explosion hazard from methane gas building up inside a home. Not regulated in drinking water by EPA or CDPHE. A recommended range for drinking water is 30-400 mg/L, according to the Illinois Dept. of Public Health. National Secondary Drinking Water Standard: 250 mg/L CDPHE Drinking Water Standard: 250 mg/L
29
Analyte
Average (n=30)
Range
GW-Avg. (n=21)
SP-Avg. (n=5)
SW-Avg. (n=4)
Fluoride (mg/L)
0.44 ±0.28
0.11 – 1.3
0.47 ±0.29
0.51 ±0.15
0.24 ±0.21
Nitrogen, Nitrate (mg/L)
0.763 ±0.951
0.005 – 4.2
0.943 ±1.02
0.55 ±0.61
0.026 ±0.022
Nitrogen, Nitrate + Nitrite^a (mg/L)
0.764 ±0.95
0.007 – 4.2
0.945 ±1.02
0.560 ±0.61
0.022 ±0.022
Nitrogen, Nitrite^b (mg/L)
0.005 ±0.009
0.002 – 0.05
0.006 ±0.01
0.005 ±0.002
0.0025 ±0.0008
Specific Conductivity (µS/cm)
726.1 ±527.2
147 – 2,680
768.05 ±556.45
778.4 ±384.3
440.5 ±424.5
Sulfate (mg/L)
171.00 ±314.14
3.3 – 1,370
169.3 ±346.5
192.8 ±220.4
152.7 ±219.7
pH
7.64 ±0.31
7.14 – 8.48
7.57 ±0.19
7.45 ±0.27
8.22 ±0.16
Dissolved Organic Carbon (mg/L)
2.60 ±3.73
0.5 – 21.7
2.67 ±4.32
1.34 ±0.70
3.8 ±1.45
Standard or Recommended Limit National Primary Drinking Water Standard: 4.0 mg/L CDPHE Human Health Standard: 4.0 mg/L See the following limit.
National Primary Drinking Water Standard: 10 mg/L CDPHE Human Health Standard: 10 mg/L See the previous limit.
Not regulated in drinking water by EPA or CDPHE. Measured in micro Siemens per centimeter (µS/cm), informal estimates put drinking water in the range of 50-1,500 µS/cm. National Secondary Drinking Water Standard: 250 mg/L CDPHE Drinking Water Standard: 250 mg/L National Secondary Drinking Water Standard: 6.5- 8.5 CDPHE Human Health Standard: 5.0-9.0 Not regulated in drinking water by EPA or CDPHE.
30
Analyte
Average (n=30)
Range
GW-Avg. (n=21)
SP-Avg. (n=5)
SW-Avg. (n=4)
Arsenic (µg/L)
ND
ND
ND
ND
ND
Barium (µg/L)
84.44 ±76.32
19.0 – 345
96.88 ±87.25
48.58 ±17.12
63.95 ±22.56
Boron (µg/L)
63.22 ±55.04
25 – 254
63.64 ±60.99
71.64 ±28.08
50.50 ±44.17
Calcium (µg/L)
74,737 ±58,720
15,900 – 242,000
77,248 ±54,787
103,940 ±72,117
25,050 ±5,706
Standard or Recommended Limit National Primary Drinking Water Standard: 0.010 mg/L CDPHE Human Health Standard: 0.010 mg/L National Primary Drinking Water Standard: 2.0 mg/L. CDPHE Human Health Standard: 2.0 mg/L Not regulated by EPA. CDPHE Agricultural Standard for Groundwater: 0.75 mg/L Not regulated in drinking water by EPA or CDPHE. National Primary Drinking Water Standard: 0.1 mg/L CDPHE Human Health Standard: 0.1 mg/L EPA Action Level: 1.3 mg/L CDPHE Drinking Water Standard: 1.0 mg/L National Secondary Drinking Water Standard: 0.3 mg/L CDPHE Drinking Water Standard: 0.3 mg/L
Chromium (µg/L)
ND
ND
ND
ND
ND
Copper (µg/L)
21.83 ±8.94
16.8 – 37.3
21.83 ±8.94
ND
ND
Iron (µg/L)
109.01 ±181.14
11.9 – 641
169.74 ±239.81
96.1 ±0.0
36.33 ±14.51
Lead (µg/L)
ND
ND
ND
ND
ND
CDPHE Human Health Standard: 0.05 mg/L
Magnesium (µg/L)
44,338 ±46,532
4,260 – 231,000
50,324 ±52,317
37,280 ±17,952
21,733 ±26,556
Not regulated by EPA or CDPHE.
31
Analyte
Average (n=30)
Range
GW-Avg. (n=21)
SP-Avg. (n=5)
SW-Avg. (n=4)
Manganese (µg/L)
101.28 ±150.54
6.9 – 602
123.93 ±164.95
31.0 ±0.0
23.2 ±11.7
Potassium (µg/L)
6,779 ±7,039
1,070 – 20,800
8,362 ±7,647
2,496 ±183
2,843 ±2,444
Selenium (µg/L)
ND
ND
ND
ND
ND
Sodium (µg/L)
36,257 ±29,396
3,940 – 125,000
38,471 ±32,590
38,520 ±3,528
21,803 ±25,457
Strontium (µg/L)
780 ±503
201 – 2,290
783 ±452
921 ±575
588 ±591
Standard or Recommended Limit National Secondary Drinking Water Standard: 0.05 mg/L CDPHE Drinking Water Standard: 0.05 mg/L Not regulated in drinking water by EPA or CDPHE. National Primary Drinking Water Standard 0.05 mg/L CDPHE Human Health Standard: 0.05 mg/L Not regulated in drinking water by EPA or CDPHE. EPA informally recommends an upper limit of 20 mg/L for people on lowsodium diets. Not currently regulated in drinking water by EPA.
Footnotes: a Calculated as: (Nitrogen, Nitrate) + (Nitrogen, Nitrite) b Elevated detection limit due to matrix interference. ND = not detected above the reporting limit or method detection limit
Table 4: Output p-values from an ANOVA t-test for seasonal differences in analyte concentrations when separated into each water source. Sample size is n=10 (GW-Sept), n=11 (GW-June), n=3 (SPSept), n=2 (SP-June), n=2 (SW-Sept), and n=2 (SW-June).
Analyte Methane Alkalinity Specific Conductance pH Dissolved Organic Carbon Calcium Chloride Fluoride
GW: Sept/June 0.894 0.998 0.999 0.957 0.913 1.00 0.999 0.999
SP: Sept/June 1.00 0.999 0.993 0.999 0.999 0.989 0.999 0.997
SW: Sept/June 1.00 0.789 0.965 0.894 0.993 0.999 0.999 0.962
32
Analyte Magnesium Potassium Sodium Sulfate Nitrate Barium Boron Copper Iron Manganese Strontium
GW: Sept/June 0.999 0.999 0.999 0.999 0.998 0.999 0.999 0.796 0.730 0.702 1.00
SP: Sept/June 0.999 1.00 0.999 0.996 0.971 0.999 0.956 1.00 0.999 0.999 0.981
SW: Sept/June 0.992 0.998 0.939 0.985 1.00 0.999 0.952 1.00 0.999 0.999 0.799
Table 5: Output p-values from an ANOVA t-test for significant differences in analyte concentrations between the water sources regardless of season. The number of samples was n=21 (groundwater), n=5 (springs), and n=4 (surface water).
Analyte GW / SP SP / SW 0.848 0.999 Methane 0.982 0.008* Alkalinity 0.999 0.627 Specific Conductance 0.490 0.00002* pH 0.771 0.615 Dissolved Organic Carbon 0.624 0.120 Calcium 0.752 0.699 Chloride 0.951 0.317 Fluoride 0.849 0.879 Magnesium 0.212 0.998 Potassium 0.999 0.694 Sodium 0.988 0.982 Sulfate 0.733 0.711 Nitrate 0.436 0.953 Barium 0.958 0.849 Boron 0.613 1.00 Copper 0.936 0.985 Iron 0.633 0.998 Manganese 0.856 0.612 Strontium * - Indicates a significant difference (p<0.05)
GW / SW 0.859 0.0008* 0.518 0.00003* 0.854 0.237 0.999 0.342 0.526 0.245 0.580 0.995 0.196 0.721 0.909 0.665 0.992 0.728 0.773
33
Table 6: Output p-values comparing the sites utilized in this study and the ERO 2011 study. It is important to note that sample sizes were extremely small. SP-01: SP-02: SW-02: Analyte CWERC/ERO CWERC/ERO CWERC/ERO (n=2) (n=3) (n=8) 0.000* 0.000* NA Methane 0.839 0.999 0.507 Chloride 0.582 0.341 0.775 Nitrate 0.753 0.912 0.120 Specific Conductance 0.0000153 * 0.0614 0.636 Sulfate 0.999 0.999 0.322 pH NA NA 0.00752 * Barium 0.999 1.000 0.266 Calcium NA NA 0.708 Iron 0.999 0.998 0.568 Magnesium NA NA 0.383 Manganese 0.995 0.999 0.999 Potassium 0.0406 * 0.531 0.124 Sodium NA NA 0.0509 Strontium * - Indicates a significant difference (p<0.05) NA – ERO study did not sample for the analyte in all water sources
34
