Water Quality, Exposure and Health

, Volume 4, Issue 3, pp 123–135

Water Quality Assessment of Newton Creek and Its Effect on Hog Island Inlet of Lake Superior

Authors

    • Department of Natural SciencesUniversity of Wisconsin—Superior
Article

DOI: 10.1007/s12403-012-0071-1

Cite this article as:
Bajjali, W. Water Qual Expo Health (2012) 4: 123. doi:10.1007/s12403-012-0071-1
  • 207 Views

Abstract

A study of water quality was conducted for Newton Creek that flows in an urban area in the city of Superior and discharges into Hog Island Inlet (HII). Water quality data were collected from five locations located along the creek flow from its head to its mouth. Throughout the year there was a wide range in measured values for the temperature, dissolved oxygen (DO), electro conductivity (EC), and pH. These ranges showed there were numerous factors affecting these properties of the creek, with the impact from oil refinery being the greatest. Some of the recorded salinity was observed to be much greater than the acute criteria. The high salinity of the creek water is attributed to the chloride and sodium concentrations with relatively elevated concentrations of sulfate and nitrate. A geochemical model revealed that the water chemical composition of HII was the result of a mixing ratio of about 90 % of Lake Superior water and 10 % of Newton Creek water. Faxon Creek, which was monitored for comparison purposes, recorded salinity concentrations much higher than the salinity of Newton Creek. Both Newton and Faxon creeks revealed different chemical makeup. Faxon Creek revealed that all major cation and anion concentrations in its water were higher than in Newton Creek waters with the exception of potassium, sulfate and nitrate. Nitrate and phosphorous concentrations in Newton Creek are much higher than in Faxon Creek and all the surface water in the area, and they exceed the recommended concentration values set by U.S. Environmental Protection Agency (EPA).

The chemical makeup and the variation of the major ions in both creeks show that the source of water contamination in Newton Creek originates mainly from the discharge of oil refinery treated wastewater with minimum contribution from surface runoff. The source of contamination and mainly the elevated salinity in Faxon Creek water is mainly from urban surface runoff from the city of Superior.

Keywords

Urban streamWater qualityHydrochemistryContaminationWatershedLand use

Introduction

Newton Creek is a 1.6 mile water-way running through a mostly residential area of Superior, WI. It originates from an artificial pool that receives its main source of water from the treated wastewater of Murphy Oil Refinery. It discharges to HII, which is part of the greater Saint Louis River System’s Superior Bay and connects to Lake Superior (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig1_HTML.jpg
Fig. 1

Newton Creek and sample locations in Superior, WI

The creek water, surrounding soil, and HII were found to be highly contaminated with hydrocarbon byproducts (DNR 1995). The contaminated site was classified by the EPA as an “Area of Concern” (EPA 2008). Remediation was conducted to remove contaminants and restoration began. Between 1997 and 2005 a total of 60,175 tons of contaminated sediment was removed from the creek and HII. Despite this remediation, contaminates continued to exist within the creek environment. The success of the remediation has been investigated and found that the water quality does not meet the EPA standard for drinking and aquatic life. The value of salinity was found to be very high and the DO and pH levels in the stream water at some locations fluctuated beyond the limit that makes the water fit for aquatic life (Bajjali et al. 2007). This suggests either the previous remediation was insufficient or the contamination continues today. Contaminant sources, especially the high salinity, chloride, and nitrate concentration in the creek’s water, have been debated and attributed to various nonpoint sources within the urban environment. Even though the creek drains an area of wetlands, the majority of the stream flow comes from the refinery’s treated wastewater effluent. The creek flows through commercial, residential and recreational properties before discharging into HII. These properties, as well as storm water runoff and municipal sewer overflows may contribute to the creek’s contamination.

Other sources of salts in water resources that could increase the salinity and chloride concentration in water could come from different sources such as bedrock weathering, soil, atmospheric deposition, ocean, and volcanic activity (Feth 1981; Peters 1991). Since the creek flows through an urban environment, the most plausible source would be from treated wastewater or deicing salt. Sodium chloride (NaCl) is used in the city of Superior as a deicing agent in order to maintain road safety during the winter season. The salt is used at a rate of 3500 short tons/year (3175 metric tons/year). Na and Cl are mobile ions and will disperse into various parts of the roadside environment. They will be dispersed mainly as run-offs, while a minor portion may be forced into the air as small particles that will be transported by the wind and fall as precipitation. The application of road salt is a good practice to make the road safe, but this habit comes at a cost to the environments especially to the water quality of lakes and streams (Novotny et al., 2007; Mullaney et al. 2009). Elevated chloride concentrations in urban streams are very well recorded in runoff from roadways. In northeastern United States, the level of chloride concentrations in stream water has been recorded up to 5,000 mg/l (Kaushal et al. 2005). Storm sewers capturing snowmelt water from surface runoff can also increase the salinity and cause seasonal variations in water bodies in an urban environment (Ramstack et al. 2004). The salt becomes a pollutant when it is conveyed to storm drains or to surface water after application. Leaks and spills of these chemicals can also occur during handling and storage. Other elements such as nitrogen, sulfur, zinc can exist in the road salt (Marsalek 2003).

Hydrogen sulfide (H2S) gas is produced by various industrial processes such as petroleum refining. H2S contamination may be treated via biochemical, chemical, and physical methods, and the byproduct will oxidize to sulfate and therefore increase the sulfate concentration (Burgess et al. 2001). Oil storage tanks may also release H2S as a result of day-night temperature changes, volatilization, and filling operations. Produced water storage vessels may contain H2S dissolved in water that is brought up from the reservoir, or it may be produced by sulfate-reducing bacteria found in water and oil (EPA 1993). Produced water is water associated with oil and gas and contains some of the chemical characteristics of the geological formation and the hydrocarbon itself.

The other major constituent of sour water (water associated with refinery that contains hydrogen sulfide and ammonia) of any industrial refinery is nitrogenous compounds like ammonia. Wastewater from integrated petroleum refineries normally contains 20–80 mg/l ammonia nitrogen, which is harmful to the aquatic life receiving this water (Fang et al. 1993). The discharge waste with ammonia becomes subject to nitrification. Nitrification is a biological process during which nitrifying bacteria convert toxic ammonia to less harmful nitrate. This process usually occurs in an aerobic condition where dissolved oxygen is more than 0.5 mg/l and this process influences the concentration of the bicarbonate in the surface water (Belser 1984; Berounsky and Nixon 1993). The nitrification reaction consumes between 7 to 8.6 mg/l of alkalinity as CaCO3 for each mg/l of ammonia nitrogen oxidized (Scearce et al. 1980; Gujer and Jenkins 1974). The consumption of bicarbonate during nitrification process can be presented in this equation:
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Equa_HTML.gif
Phosphorus in stream water comes from phosphate-containing rocks, fertilizers, pesticides, and industry (Heathwaite 1997). Phosphates also are used widely in power plant boilers to prevent corrosion and the formation of scale. Phosphorus enrichment in rivers can degrade the plant community by altering the competitive balance between different aquatic plant species, including both higher plants and algae. This interaction has consequences for the whole ecosystem (Mainstone and Parr 2002). The total load to the river can be broadly divided into point source input and diffuse input. Point sources are typically dominated by sewage treatment work. Effluents from industrial waste can also be phosphorus-rich; the contribution from industry sources could reach 2.8 % (Parr et al. 1998).

Heavy metals such as arsenic, lead, copper, and zinc can originate from many sources within a refinery. The behavior and mobility of each trace element can vary with the pH. Zinc for example; is best absorbed with high pH, while arsenic absorbs better at low pH (Merill et al. 1986). Low pH can also allow toxic elements and compounds to become mobile and not needed for uptake by aquatic plants and animals.

This paper focuses on the sources of contamination and how the contaminants are changing the water quality of both Newton and Faxon Creeks in Superior, WI. It has five objectives: (1) determine how each individual creek, running in close proximity to each other and in the same environmental conditions, can have different chemical makeups; (2) determine the main source of contamination for each creek; (3) determine if the high concentrations of sodium and chloride in both creeks are originating from treated wastewater or surface runoff from the impervious layer of the watershed of each creek; (4) determine the relationship between the chemical elements in the creeks’ water; (5) verify if the water quality of HII is affected by the discharge from Newton Creek.

Site Overview

The study area is in Superior, Douglas County, in the northwestern corner of Wisconsin. It is part of the Lake Superior Lowland, which slopes gradually upwards toward the south from the shores of Lake Superior. Newton Creek, Faxon Creek, and the Nemadji River are considered an immediate part of the St. Louis River Watershed. The watershed is the largest U.S. tributary to Lake Superior and drains 3,634 square miles, entering the southwestern corner of the lake between Duluth, Minnesota and Superior, Wisconsin (Fig. 1). St. Louis River flows 179 miles and as it approaches Duluth and Superior, the river takes on the characteristics of a 12,000 acre freshwater estuary. The estuary consists mainly of forest (65 %), wetland (11 %), and agriculture (9 %), while the rest is open water, urban and suburban, industrial grassland, barren, and open land within the twin ports of Superior, Wisconsin and Duluth, Minnesota (Merryfield et al. 2000). The study area consists of a continental climate with varying temperatures throughout the year, largely due to northeast winds off of Lake Superior. The average annual temperature of northwest Wisconsin is 30 F and on average has an 80-day freeze-free average. Northwestern Wisconsin consists of mainly sedimentary rocks including sandstone, shale and glacial lacustrine red clay. The red clay in the study area is finely textured, resulting in very poor drainage and prevents the groundwater from recharging the surface water (DCLCC and LWCD 2009). Erosion due to glacial movements has played a major part in forming northwest Wisconsin’s shape and topology. Nearly half of the streams and rivers in northern Wisconsin including Nemadji, Newton, and Faxon, drain into Lake Superior.

Methodology

Various approaches were used in this study; chemical analysis, watershed delineation, and land use creation in GIS environment. The different water quality data were analyzed from the creeks’ water to meet the objectives of the study.

Water Collection

The first data set was conducted at five sites along the flow of both creeks: Newton and Faxon (Fig. 1). It was conducted in 2005 along Newton Creek, and in 2010 along Faxon Creek. The DO, EC, pH, and temperature of creek water were measured on a weekly basis using Omega devices; Dual pH/conductivity Handheld Instruments (PHH60 and PHH80 POCKET PAL) and Portable Digital Dissolved Oxygen Temperature Meters (PHDG80A). The water at the five sites was measured from the bank of the stream. Water of Newton Creek at site 1 was always running despite the drop of atmospheric temperature below zero. At the other four sites, the water can freeze when the temperature drops below zero and then the water cannot be measured. The second set of data was sampled for major cations, anions, and nutrients to provide more detailed information to explain the reason of high salinity in the creek’s water (Table 1). The sampling was performed twice for Newton Creek, the first sampling was in summer (August, 2009) in which all sites were sampled; the second time was in winter (November, 2011) in which only site 1, 3, and 5 were sampled. Thus, a total of eight samples were collected (Table 1). At the same time, analysis of the water was carried out for the nearby surface water bodies. Three samples were collected from Faxon Creek, two from HII, two from Lake Superior, and one from Nemadji River, these data will be used as reference regarding fresh water in this region. Table 1 shows all sampling sites, and all of the individual samples in the table match the site locations in Fig. 1.
Table 1

Average chemical analysis of the data

Sampling location

Date

Site #

Ca

Mg

Na

K

Cl

HCO3

SO4

NO3

P

No of samples

   

meq/l

mg/l

 

Newton Cr.

8/2009–11/2010

NC1, NC2, NC3, NC4, NC5

2.41

1.13

8.67

0.47

7.91

1.66

2.64

22.92

0.354

8

Faxon Cr.

11/2010

FC1, FC3, FC5

4.58

2.62

52.52

0.28

57.92

3.8

0.89

0.01

0.046

3

H.I.I

8/2009–11/2010

HHI

1.06

0.69

0.97

0.08

1.08

1.21

0.45

0.00

0.021

2

L. Superior

8/2009

LS1, LS2

0.96

0.78

0.48

0.05

0.33

1.31

0.37

0.00

0.028

2

Nemadji R.

8/2009

NR 0.89

0.48

0.14

0.03

0.13

1.05

0.2

0.00

0.022

1

 

Data Processing in GIS

A handheld Garmin GPS instrument was used to capture the coordinate system of all the sampling sites (Fig. 1). The error discrepancies of the latitude and longitude were taken into consideration with plus and minus 30 feet. The data points gathered were in database format (dbf). The dbf file of Garmin GPS was integrated into GIS and converted into shapefile format. The data underwent several processing steps to make them suitable for analysis, such as map projection, geoprocessing, and digitizing. The GPS-produced data were registered in geographic coordinate and World Geodetic System of 1984 (WGS84) Datum. In order to align the collected GIS data with an existing aerial photograph registered in Universal Transverse Mercator (UTM) Zone 15, the data were projected into UTM Zone 15 using the North American Datum (NAD83). In geoprocessing, the clip function was used in order to extract the digitized layers of the land cover/land use that were associated with the Newton Creek watershed. This was done by using the boundary of the watershed created in ArcGIS10 using the hydrology tool.

Watershed and Land Use Creation in GIS Environment

To determine the flow from a watershed, it was important to model the movement of surface water, which was done using the digital elevation model (DEM) in GIS technology. The DEM is elevation data in grid format necessary to delineate drainage basins, create and compute stream drainage basin data, and identify the surface drainage pattern (Olivera et al. 1998). Various steps were required to delineate the watershed. The hydrology tool of ArcGIS identified and filled any depressions that existed in the DEM. After filling all the sinks, a flow direction was determined by the elevation of the surrounding cell of the DEM in order to delineate the drainage network and drainage divide. Following the flow direction step, the flow accumulation, stream network, stream order, and stream link raster were created. The final step was to delineate the watershed. The watershed area for Newton Creek was delineated using a high resolution (10 meter) DEM obtained via The National Map Seamless Server. Figure 2 shows the final watershed of Newton Creek and all the land cover/land use inside of it. The second step was to create the land cover/land use inside the Newton Creek watershed. Land use implies a human component and refers to any human activities on the land (Meyer 1994). Therefore, all features covering the surface of the Newton Creek watershed, such as water bodies, wetlands, parking lots, buildings, and roads, were digitized using aerial photograph of high resolution (1 feet) obtained from the city of Superior. Knowledge about land use and land cover within the watershed will shed light on the sources and mechanisms that lead to the deteriorated water quality of Newton Creek. Within the watershed, there are different land uses that range from residential (neighborhoods, parks, roads, and houses), commercial, industrial, and wetland (Fig. 2). Roads, parking lots, rooftops, and sidewalks are considered impervious surfaces as they are covered by impenetrable materials such as asphalt and concrete. Many types of pollutant, originating from a variety of sources, accumulate over these surfaces and are carried away into the creek’s water during storm events as surface runoff. These surfaces could pose significant threats to Newton Creek by degrading its water quality and impairing the aquatic life.
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig2_HTML.jpg
Fig. 2

Land cover/land use map within Newton Creek watershed

Result

Water Quality

Newton Creek was sampled for DO, EC, temperature, and pH at five sites along its flow path from its head at Murphy Oil facilities to its mouth at HII beginning in February 2005 (Fig. 1). These parameters are important to aquatic life in the stream. The presence of DO in water is essential to aquatic life such as bottom dwelling invertebrate, fish, and other vertebrates. Bacteria utilize oxygen continuously to decompose organic matter. pH affects many chemical and biological processes in the water, which is important to the survival and reproduction of aquatic life such as fish. A pH higher than 8 causes stress and limits growth, while a pH lower than 5 can speed the mobility of certain trace elements that could harm plants and animals. High EC indicates pollution from different sources, such as urban runoff, road salt, wastewater, and others. Water temperature is a critical parameter for aquatic life and has an impact on DO concentration and bacterial activity in water. The temperature can determine which fish and macroinvertebrate species can survive in a given stream (Mackie 2001).

Dissolved Oxygen

The creek system both produces and consumes oxygen, and it is produced during photosynthesis and from atmosphere through aeration (Salisbury and Ross 1978). The amount of DO that a given volume of water can hold is a function of atmospheric pressure, water temperature, and the amount of other substances dissolved in the water. To sustain life, DO should be above the chronic criterion for growth: 4.8 mg/l (EPA 2000a). Somewhat lower concentrations put aquatic life under stress. There is clear variation in the DO concentration at the five sites (Fig. 3). This could be due to warmer summer temperatures and increased biological activity in terms of organic matter decomposition. Several sample sites recorded levels below chronic criterion for growth of aquatic life. Figure 3 shows that the extreme values below 4.8 mg/l were measured at all sites on different dates.
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig3_HTML.jpg
Fig. 3

Dissolved oxygen fluctuations (2005 to 2011)

Hydrogen Ion Concentration

(pH) The pH affects many chemical and biological processes in the water. Different organisms thrive within different ranges of pH and the majority of aquatic animals prefer a range of 6.5–8.0. Higher pH values pose stress on the organism and could reduce the diversity and reproduction of many organisms. The Newton Creek water showed fluctuation of the pH between acidic and alkaline along the stream at the five sampling sites (Fig. 4). Water with pH value above 10 were exceptional and may reflect contamination by a strong base such as NaOH and Ca(OH)2 (Langmuir 1997). It may also be a symptom of excess aquatic plant growth. Water plants are capable of using the bicarbonate ion if CO2 availability is limited, and using the bicarbonate ions causes the pH to rise. The range of the pH in the creek shows that at certain times the water is harmful to aquatic animals (Fig. 4). In addition to that, the low pH could make some toxic trace element to be released which may end up in the fat tissue of the organism in HII.
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig4_HTML.jpg
Fig. 4

pH fluctuations (2005 to 2011)

Temperature

Organisms are sensitive to temperature and will change location to select optimal temperature. If temperatures are outside this optimal range for a prolonged period of time, organisms are stressed and can cease to function. Causes of temperature change include weather, removal of shading from vegetation along the stream bank, and discharge of cool or warm water from industry. Site 1 never recorded temperature below zero throughout the year compared to the rest of the sites. Figure 5 shows the fluctuation of the temperature during the study period. You can observe that the temperature in the graph increases in summer and decreases in winter.
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig5_HTML.jpg
Fig. 5

Temperature fluctuations (2005 to 2011)

Electrical Conductivity

The EC is a measure of the ionic activity of a solution in terms of its capacity to transmit electrical current. Because EC depends on the overall ionic concentration in water, it has been used as an indirect measure of the presence of inorganic total dissolved solids (TDS) in aqueous solution. The ratio of TDS to EC for solutions of pure salts ranges from 0.4 to 0.7 for EC up to about 500 μS/cm, depending on the salt (Hem 1982). The creek water EC showed seasonal variations throughout the year (Fig. 6).
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig6_HTML.jpg
Fig. 6

EC fluctuation (2005–2011)

In general the EC value fluctuates sharply between 200 μS/cm and 2900 μS/cm along the flow of the creek. To quantify the EC values of the creek water and characterize the general trend of fluctuation, a graph was created to simplify the discussion of the creek’s salinity (Fig. 7).
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig7_HTML.jpg
Fig. 7

Average EC values on a monthly basis

The graph represents the average EC values on a monthly basis. The creek water EC showed, in general, a similar trend of fluctuation between the winter, summer, fall, or spring throughout the sampling years (2005–2011) at all site, except for site 5. The similar trends of EC seasonal fluctuation emphasize that the input sources of EC to creek water system behaves in a similar way throughout the year. In general, the average EC value at each site always increased in February, May, July–September, and November–December, while the value decreased in January, March–April, June, and October, with the exception being at site 5 (Fig. 7). Around 72.4 % of the average EC samples were higher than 1,000 μc/cm and only 27.6 % of the average EC samples were less than 1,000 μc/cm. Site 5 represented around 19.2 % of the total samples, and 12 % these samples had average EC values less than 716 μc/cm. The average high EC value ranged between 1,000 and 1,844 μc/cm, while the low EC value ranged between 564 and 995 μc/cm. The average EC values dropped slightly between site 1 and site 4, while it dropped dramatically at site 5. The relatively high EC value demonstrates that the source of water in the creek originates from polluted water. This source could originate mainly from the effluent of the treated wastewater of oil refinery. Another source, such as surface runoff from intense rain and snow melt, is possible especially in April, but not as a main contributor to the base flow. This is clearly demonstrated in Fig. 7 where the elevated EC in April is relatively low compared to the EC values in May, August, and September when there is no snow melt. The relatively low EC values found at site 5, the mouth of the creek at HII is in direct contrast to characteristics of a natural stream. This suggests that tidal influence from HII water dilutes the salinity of the water.

Nutrients

Nitrogen and phosphorus were analyzed twice in order to gain information about their concentrations in the creek. Nutrients play key roles in ecosystems and are key plant nutrients; however, they can contribute to water quality degradation when they are in excess. Phosphorus and nitrogen are delivered to the river system from a range of sources, varying in its bioavailability from source to source and could come from natural and non-natural sources, such as human and animal wastes. The tested water of the five sites had NO3 (nitrate) concentrations that ranged between 20.79 to 24.33 mg/l. The average concentration was much higher than the natural abundance of nitrate in natural water, which is less than 5 mg/l, but does not exceed the federal drinking water standards, which is 45 mg/l as NO3 (nitrate). The phosphorous concentrations observed in the Newton Creek water ranged between 0.29 and 0.46 mg/l. This concentration is three to four times the recommended maximum concentration values for rivers and streams set by EPA (2000b). The high phosphorous concentration could come mainly from the oil refinery. In the refinery industry, phosphorous is used as a common corrosion inhibitor. The phosphorus-containing formulations are generally more effective than the non-phosphorus, but they bring with them the concern about polluting the environment after disposing of them into the water body (Shalaby 2006). Another possible source of phosphorous load in the creek is from untreated sewage. In spring during intense rain and snow melt, certain areas in Superior experience flooding of the basement in houses. This flooding happens because the sewer system is unable to handle both wastewater and storm water that runs off impervious surfaces. As a result, the wastewater ends up flowing untreated into the creek. It is clear that the phosphorous concentration is much above the recorded phosphorus concentration in Lake Superior, HII, and Nemadji River, which is less than 0.1 mg/l and which represents the amount of phosphate-phosphorus that is usually found in most uncontaminated water bodies.

Hydrochemistry of Water

As part of an effort to obtain more comprehensive data regarding water quality of Newton Creek and the surrounding water body for comparison purposes, water samples for complete chemical analysis were collected from twelve sites (Fig. 1). The pH, DO, EC, and temperature were measured in the field whereas the concentration of major cations, anions, and nutrients were analyzed at the laboratory as per the standard analytical procedures at the University of Wisconsin Stevens Point laboratory. Table 1 shows the average concentration of the major ions that were analyzed.

A frequency histogram was created for all the samples in Table 1 to obtain information about the TDS distribution (Fig. 8). The graph reveals that there are huge discrepancies in the salinity between all the water sites despite the sample locations being within less than 2 miles away from each other. The salinity of distinct water locations clustered into different groups. The salinity of water in Faxon Creek is much higher than the salinity of water in Newton Creek in all sampled sites. Nevertheless, both creek waters recorded TDS much higher than the permissible level for drinking, which is 500 mg/l (EPA 2009). The average salinity for Newton Creek samples was 690.8 mg/l, and for the Faxon Creek it was 3670 mg/l. Much lower salinity values were recorded for the Nemadji River (107 mg/l) and Lake Superior (151.8 mg/l). The salinity of HII was 185 mg/l, which was almost 31 % higher than the salinity of Lake Superior sample.
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig8_HTML.jpg
Fig. 8

TDS histogram for all surface water

In order to shed light on the hydrochemistry of the water between all surface water sites, the major cations (Na+, K+, Ca2+, Mg2+) and anions (\(\mathrm{HCO}_{3}^{-}\), \(\mathrm{SO}_{4}^{2-}\) and Cl) are presented by plotting them on a Piper diagram (Fig. 9).
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig9_HTML.jpg
Fig. 9

Piper diagram for all surface water

The diagram reveals the similarity, dissimilarities, and different types of water in the study areas, which are identified and listed in Table 2. The diagram shows that the water is divided mainly into three hydrochemical facies. The concept of facies was developed in order to understand and identify the water composition in different classes (Back 1961). Group one is classified as sodium chloride type and is identical for the water of Newton and Faxon creeks. Group two is classified as calcium chloride and this type of water is alkaline earth water with increased portion of alkalies with prevailing chloride. This type of water belongs to HII. Group three is classified as calcium bicarbonate and this type of water is alkaline earth water with prevailing bicarbonate. Both Lake Superior and Nemadji River belong to this type of water.
Table 2

Type of water and average molal ratio

Site #

Location

Type

\(\frac{\mathrm{Cl}^{-}}{\mathrm{SO}_{4}^{-2}}\)

\(\frac{\mathrm{HCO}_{3}^{-}}{\mathrm{SO}_{4}^{-2}}\)

\(\frac{\mathrm{Na}^{+}}{\mathrm{K}^{-}}\)

NC1, NC2, NC3, NC4, NC5

Newton Creek

Sodium chloride

4.75

0.73

20.25

FC1, FC2, FC3

Faxon Creek

Sodium chloride

63.80

4.37

205.49

HII

Hog Island Inlet

Calcium chloride

2.48

2.71

12.41

LS1, LS2

Lake Superior

Calcium bicarbonate

0.88

3.50

10.41

NR

Nimadji river

Calcium bicarbonate

0.66

5.35

5.61

Source of Contamination in Faxon and Newton Creeks

Despite Faxon and Newton creeks being clustered in one location in the piper diagram, they have completely different chemical makeup and different molar ratios. The high salinity in both creeks is attributed mainly to the high chloride and sodium concentrations. Faxon Creek salinity is heterogeneous along the water flow and almost three and half times higher than the salinity of Newton Creek. Furthermore, the chloride content in Faxon Creek is almost five times higher than the chloride concentration in Newton Creek. The highest concentration was found at site 2, which is located close to the Superior Senior High School. At this location there is a snow storage area for the University of Wisconsin Superior. Site 2 receives all the melt water from this storage site.

All chemical elements of Faxon Creek water are higher than the water of Newton Creek with exception the K+, \(\mathrm{SO}_{4}^{2-}\) and \(\mathrm{NO}_{3}^{-}\) (Fig. 10). Furthermore, the different chemical makeup of both creeks’ water is very well observed in the molal ratio of different ion ratios (Table 2). The average Na/K, \(\mathrm{HCO}_{3}^{-}\)/\(\mathrm{SO}_{4}^{2-}\) and Cl/\(\mathrm{SO}_{4}^{2-}\) is 10, 6, and 13 times higher in Faxon Creek water than in Newton Creek water. With regard to the \(\mathrm{NO}_{3}^{-}\) concentration, it is less than 1 mg/l in Faxon Creek, but higher than 20 mg/l in Newton Creek. This huge discrepancy in nitrate, salinity, and variation in major ions leads to belief that the water for these two creeks originates from different sources despite their close proximity. The source of high salinity in Faxon Creek is mainly attributed to the chloride content, which likely originates from deicing compounds. The chemical substance that is used for deicing in the city of Superior is halite (NaCl). The main mechanism for accumulation of the salt in the creek water is during the snow melting period. The surface runoff during precipitation and snow melt in spring is the main source. The other source is from the salt that precipitates on streets and soil from evaporation effects. Some melting water in spring time will become subject to evaporation, leaving the various types of salt to precipitate on the ground. The precipitation and surface runoff will dissolve the salt from the streets and soil again and carry it to the creek, which will also contribute to the increase of salinity.
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig10_HTML.jpg
Fig. 10

Newton and Faxon Creeks chemical comparison

The source of water in Newton Creek could come mainly from oil refinery treated wastewater. The \(\mathrm{SO}_{4}^{2-}\) and \(\mathrm{NO}_{3}^{-}\) components in the water are identical to oil facility treatment plants. Separation of ammonia and hydrogen sulfide in a wastewater treatment plant leads to a high concentration of sulfate and ammonia. The absence of NH4 in Newton Creek water indicates that nitrification took place, which means the conversion of NH4 to \(\mathrm{NO}_{3}^{-}\) according to the following equation (Freeze and Cherry 1979):
$$\mathrm{NH}_{4} + 2\mathrm{O}_{2} \rightarrow \mathrm{NO}_{3}^{-} + 2\mathrm{H}^{+} + \mathrm{H}_{2}\mathrm{O} $$

This nitrification process influences the concentration of bicarbonate in the creek water by reducing it (Belser 1984). Therefore, the observed bicarbonate concentration in Newton Creek is almost less than half the concentration in Faxon Creek (Table 1). The \(\mathrm{SO}_{4}^{-2}\) concentration in Newton Creek is homogeneous along the entire flow and it is nearly three times higher than the concentration of \(\mathrm{SO}_{4}^{-2}\) in Faxon Creek.

Hydraulic Relationship and Mixing Pattern

The Ludwig diagram in Fig. 11 and Table 1 shows that HII has obvious variations in some cations and anions in comparison to freshwater in the region, and it plots between Newton Creek and Lake Superior. The HII, Lake Superior, and Newton Creek plotted on a straight line. In addition HHI demonstrates 2.7 times higher Cl concentration than Lake Superior (Table 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs12403-012-0071-1/MediaObjects/12403_2012_71_Fig11_HTML.jpg
Fig. 11

Ludwig diagram

Table 3

Mixed ratios between the Newton Creek and Lake Superior

Parameter

Newton Creek

Lake Superior

Mixed water

HII

Mixing ratio

10 %

90 %

  

Ca2+ (ppm)

41.80

18.20

21.20

20.56

Mg2+ (ppm)

12.94

8.72

23.50

25.35

Na+ (ppm)

163.50

10.00

90.00

86.00

K+ (ppm)

13.30

1.60

3.40

2.77

Cl (ppm)

181.00

10.10

28.50

27.19

\(\mathrm{SO}_{4}^{-2}\) (ppm)

130.68

16.95

23.83

28.32

\(\mathrm{HCO}_{3}^{-}\) (ppm)

96.00

76.00

76.00

78.00

This evidence suggests that the water of HII could be a mixture between the two waters in Newton Creek and Lake Superior. This notion was checked by mixing two types of water: different percentages of water from Lake Superior and Newton Creek. It was found that 90 % of Lake Superior and 10 % of Newton Creek represented the water chemistry in HII. The geochemical mass balance for this mixture was tested with the geochemical model (AquaChem) according to the formulation:
$$\sum m_i(\mathit{mixture})=\sum X1 m_{i,1}+\sum X2 m_{1,2} $$
where mi is the analytical molality of the species of the ith element and X1 and X2 are the proportion of water (by weight) in which solution 1 and 2 are mixed.

The output result shows that the calculated mixed water of the conservative ion of Cl concentration was relatively similar with the measured Cl concentration of HII (Table 3). Even the less conservative elements show an excellent correlation. These data suggest that Newton Creek could be responsible for the slight increase in salinity level of HII water.

Conclusion

The investigator found that water quality is severely affected by salt contamination. Some of the recorded parameters of the DO, pH, and salinity were observed to be much greater than the acute criteria. The elevated salinity in the water of Newton Creek is attributed to the Na+ and Cl with relatively high concentrations of \(\mathrm{SO}_{4}^{2-}\) and \(\mathrm{NO}_{3}^{-}\). The chemical makeup of the water in Newton Creek was found to be completely different than the surrounding water resources, Faxon Creek, Nemadji River, Lake Superior, and HII, due to its origin from industrially treated wastewater. The Newton Creek water affected the quality of water of HII. A geochemical model reveals that the water of HII is a mixture between Newton Creek and Lake Superior. The ratio of mixing is 90 % of Lake Superior and 10 % of Newton Creek. The source of high salinity in Faxon Creek is mainly attributed to the chloride content, which likely originates from deicing compounds. The current water quality in both Newton and Faxon Creeks does not stimulate the biodiversity and health of aquatic organism such as shellfish, macro-invertebrates, and fish species and effective establishment of wetland and riparian habitats. In order to make the water more habitable and unpolluted, the sources of contamination originating primarily from oil refinery discharge and deicing must be controlled and regulated.

Acknowledgements

The author is grateful to the Douglas County Land and Water Conservation Department for funding the project. I also extend my great thanks for all my students who participate in sampling and analyzing the water.

Copyright information

© Springer Science+Business Media B.V. 2012