103370 by UNIGIS América Latina

Master Thesis | Tesis de Maestría submitted within the UNIGIS MSc programme presentada para el Programa UNIGIS MSc at/en Interfaculty Department of Geoinformatics-Z_GIS Departamento de Geomática – Z_GIS University of Salzburg | Universidad de Salzburg

Water Quality in Alaska’s Kenai River, United States of America; Past, Present and Future by/por

Edgar Javier Guerron Orejuela 0969336 A thesis submitted in partial fulfilment of the requirements of the degree of Master of Science (Geographical Information Science & Systems) – MSc (GIS)

Riverview-United States of America, May 4, 2017

UNIGIS Master of Science in GIS

SCIENCE PLEDGE Por medio del presente documento, incluyendo mi firma personal certifico y aseguro que mi tesis es completamente el resultado de mi propio trabajo. He citado todas las fuentes que he usado en mi tesis y en todos los casos he indicado su origen.

Riverview, Florida; 05/04/2017 (Lugar, Fecha)

UNIGIS Master of Science in GIS

(Firma)

ABSTRACT The Kenai Watershed Forum and several governmental agencies formed a cooperative partnership to collect and analyze water samples from 13 locations along the Kenai River mainstem and from 8 of its tributaries every April and July from 2000 to 2014. Laboratory analysis were conducted on dissolved metals, total metals, nutrients, hydrocarbons, fecal coliform bacteria, and several other parameters. Data was analyzed using R statistical program, and the results were compared to the Alaska and federal water quality standards for freshwater aquatic life. Also, linear referencing, dynamic segmentation, inverse distance weighting and other GIS tools were used to characterize the water quality in the entire river as well as to assess the influence of land use on the water quality of the Kenai River. Results show a clear seasonality in the data, as well as a direct influence of the land use adjacent to the river and the water quality. Total metals had relatively few exceedances, excluding zinc levels in Slikok Creek and Soldotna Creek. Iron levels consistently exceeded the standard, especially in the Kenai River estuary and in the tributaries. Calcium and magnesium do not have applicable Alaska or federal standards. However, they were highest in the estuary and in Soldotna Creek. Nitrate concentrations decreased from Kenai Lake to the estuary while phosphorus increased. In the lower river, median hydrocarbon concentrations exceeded the Alaska standard during July. Total suspended solids and turbidity levels were highest in the estuary and in the Killey River. Water temperatures exceeded several standards in July, especially in the Moose River and other tributaries. Further studies and any necessary restoration should be considered for locations close to the confluences of the tributaries to the Kenai River as well as to the lower river since this presents the areas of greater development. In conclusion, the Kenai River is a very healthy system with better water quality in the spring than in the summer and with a clear difference between the water quality in the upper Kenai River and in the lower Kenai River, which is driven by land use. This study will allow managers to prioritize further studies and restoration actions to the sections of the river that present water quality concerns.

UNIGIS Master of Science in GIS

RESUMEN Kenai Watershed Forum y varias agencias gubernamentales formaron una asociación de cooperación para recoger y analizar muestras de agua en 13 lugares a lo largo del río Kenai y 8 de sus afluentes cada abril y julio desde el año 2000 hasta el 2014. Metales disueltos, metales totales, nutrientes, hidrocarburos, bacterias coliformes fecales, y varios otros parámetros fueron analizados en laboratorio. Los resultados de estos análisis son comparados en el presente documento a los estándares del estado de Alaska y los estándares federales de calidad del agua para la vida acuática. Los datos fueron analizados con la ayuda del programa estadístico R. También se utilizaron las siguientes herramientas espaciales: segmentación lineal, Inverse distance weighting y referenciación lineal. Estas permitieron caracterizar la calidad del agua en todo el río, así como también analizar la influencia que tienen los distintos tipos de uso de suelo en la calidad de agua del río. Los resultados muestran una clara influencia en la calidad del agua no solamente de las estaciones del año (primavera o verano), sino también de los usos de suelo aledaños al rio. Los metales totales tienen, relativamente, pocos casos en los que los estándares son excedidos, exceptuando los niveles de zinc en Slikok Creek y Soldotna Creek. Los niveles de hierro superan los estándares de manera regular, especialmente en la desembocadura del río Kenai y en los afluentes. El estado de Alaska y el gobierno federal de los Estados Unidos no tienen estándares para el calcio y el magnesio. Sin embargo, estos tuvieron valores elevados en el estuario y en Soldotna Creek. Las concentraciones de nitratos disminuyeron desde el lago Kenai al estuario, mientras que los valores de fósforo incrementaron. En la parte baja del río, las medianas de las concentraciones de hidrocarburos superaron el estándar de Alaska durante el mes de julio. Los sólidos suspendidos totales y los niveles de turbidez fueron más altos en el estuario y en el río Killey. Las temperaturas del agua superan varias normas en julio, especialmente en el río Moose y otros afluentes. Se recomienda realizar estudios adicionales para conocer mejor la razón por la cual los estándares son superados. Además, se recomienda fomentar esfuerzos de restauración en lugares cercanos a los ríos tributarios y a las zonas de mayor desarrollo. En conclusión, el río Kenai es un río muy

UNIGIS Master of Science in GIS

limpio, con calidad de agua más alta en la primavera que en el verano. Además, la calidad de agua de la parte alta del río es mejor que la de la parte baja del río. Este estudio permitirá a los administradores priorizar que secciones del río tienen mayor necesidad de esfuerzos de restauración o niveles de proyección.

UNIGIS Master of Science in GIS

ACKNOWLEDGEMENTS This thesis could not have been completed without the assistance and cooperation of the Kenai Watershed Forum and of many other institutions including the Alaska Department of Environmental Conservation (ADEC), the Alaska Department of Natural Resources (ADNR), and the Alaska Department of Fish and Game (ADFG). The Kenaitze Indian Tribe, Cook Inlet Aquaculture Association, the Nature Conservancy, Analytica Laboratories, Kenai Peninsula Trout Unlimited and Taurianen Engineering and Testing (previously known as Northern Testing Laboratories) also supported this project. Additional cooperation transpired with the United States Forest Service (USFS), the United States Fish and Wildlife Service (USFWS), the Kenai Peninsula Borough, the City of Soldotna, and the City of Kenai. Finally, many landowners graciously allowed access to the Kenai River and its tributaries from their property.

Thank you to all the staff at the Kenai Watershed Forum, the UNIGIS team and my advisor, for providing technical support for the completion of my thesis. Special thanks to my parents, my brother and my beautiful wife who were next to me through this process.

UNIGIS Master of Science in GIS

ACRONYMS ADEC: Alaska Department of Environmental Conservation ADFG: Alaska Department of Fish and Game BTEX: Benzene, Toluene, Ethylbenzene, and Xylenes DRO: Diesel Range Organics GIS: Geographic Information System GPS: Global Positioning System GRO: Gasoline Range Organics HUC: Hydrologic Unit Code IDW: Inverse Distance Weighting KPB: Kenai Peninsula Borough KWF: Kenai Watershed Forum MDL: Method Detection Limit NHD: National Hydrography Dataset PCA: Principal Component Analysis RRO: Residual Range Organics SFPI: Single Factor Pollution Index SPARROW: Spatially Referenced Regressions on Watershed Attributes SWAT: Soil and Water Assessment Tool USEPA: United States Environmental Protection Agency USFWS: United States Fish and Wildlife Service USGS: United States Geological Survey WASP: Water Quality Analysis Simulation Program WBD: Watershed Boundary Dataset WQI: Water Quality Index

UNIT ABBREVIATIONS CFU/100m = coliform forming units per 100 milliliters NTU= nephelometric turbidity unit mg/L = milligrams per liter µg/L = micrograms per liter µS/cm = microsiemens per centimeter UNIGIS Master of Science in GIS

TABLE OF CONTENTS FIGURES .............................................................................................................................. 10 TABLES ................................................................................................................................ 12 CHAPTER 1: INTRODUCTION ..................................................................................... 13

1.1 Background ...................................................................................................................13 1.2 Motivation ......................................................................................................................14 1.3 Objectives ......................................................................................................................16 1.3.1 General Objective ...........................................................................................................................16 1.3.2 Specific Objectives .........................................................................................................................16 1.4 Research Questions ..................................................................................................16 1.5 Hypothesis.....................................................................................................................16 1.6 Justification ..................................................................................................................17 1.7 Scope...............................................................................................................................18

CHAPTER 2: LITERATURE REVIEW ......................................................................... 20 2.1 Basic Concepts ..................................................................................................................20 2.1.1 Water Quality Standards .............................................................................................................20 2.1.2 National Hydrography Dataset.................................................................................................20 2.1.3 Water Quality Index........................................................................................................................21 2.2 Water Quality .......................................................................................................................23 2.2.1 Dissolved Metals ..............................................................................................................................23 2.2.1.1 Arsenic ...............................................................................................................................................23 2.2.1.2 Cadmium ..........................................................................................................................................24 2.2.1.3 Chromium ........................................................................................................................................24 2.2.1.4 Copper ...............................................................................................................................................24 2.2.1.5 Lead ....................................................................................................................................................25 2.2.1.6 Zinc ......................................................................................................................................................25 2.2.2 Total Metals ........................................................................................................................................26 2.2.2.1 Calcium .............................................................................................................................................26 2.2.2.2 Iron .......................................................................................................................................................26 2.2.2.3 Magnesium .....................................................................................................................................26 2.2.3 Nutrients ...............................................................................................................................................27 2.2.3.1 Nitrate .................................................................................................................................................27 2.2.3.2 Phosphorus.....................................................................................................................................27 2.2.4 Hydrocarbons ....................................................................................................................................27 2.2.4.1 Diesel Range Organics ............................................................................................................27 2.2.4.2 Gasoline Range Organics ......................................................................................................28 2.2.4.3 Residual Range Organics ......................................................................................................28 2.2.4.4 Total BTEX ......................................................................................................................................28 2.2.5 Fecal Coliform Bacteria................................................................................................................29 2.2.6 pH .............................................................................................................................................................30 2.2.7 Specific Conductance ...................................................................................................................30 2.2.8 Total Suspended Solids ...............................................................................................................30 2.2.9 Turbidity ................................................................................................................................................31 2.2.10 Water Temperature .....................................................................................................................31 2.3 Water Quality and GIS analysis ....................................................................................36 2.3.1 Linear referencing ...........................................................................................................................38 2.3.2 Dynamic Segmentation ................................................................................................................39 2.3.3 Inverse Distance Weighted Interpolation ...........................................................................39

CHAPTER 3: METHODOLOGY .................................................................................... 40

3.1 Study Area ............................................................................................................................40

UNIGIS Master of Science in GIS

3.1.1 Sampling Locations ........................................................................................................................42 3.2 Sample Collection and Laboratory Analysis ...........................................................46 3.3 Data Digitalization and Geographical Analysis ......................................................48

CHAPTER 4: RESULTS AND DISCUSSION ............................................................ 61

4.1 Results ...................................................................................................................................61 4.1.1 Parameters..........................................................................................................................................61 4.1.1.1 Arsenic ...............................................................................................................................................61 4.1.1.2 Cadmium ..........................................................................................................................................63 4.1.1.3 Chromium ........................................................................................................................................63 4.1.1.4 Copper ...............................................................................................................................................65 4.1.1.5 Lead ....................................................................................................................................................67 4.1.1.6 Zinc ......................................................................................................................................................69 4.1.1.7 Calcium .............................................................................................................................................71 4.1.1.8 Iron .......................................................................................................................................................73 4.1.1.9 Magnesium .....................................................................................................................................75 4.1.1.10 Nitrate ..............................................................................................................................................77 4.1.1.11 Phosphorus ..................................................................................................................................79 4.1.1.12 Diesel Range Organics .........................................................................................................81 4.1.1.13 Gasoline Range Organics ...................................................................................................81 4.1.1.14 Residual Range Organics....................................................................................................81 4.1.1.15 Total BTEX ...................................................................................................................................82 4.1.1.16 Fecal Coliform Bacteria.........................................................................................................82 4.1.1.17 pH ......................................................................................................................................................84 4.1.1.18 Specific Conductance ............................................................................................................86 4.1.1.19 Total Suspended Solids ........................................................................................................88 4.1.1.20 Turbidity .........................................................................................................................................90 4.1.1.21 Water Temperature .................................................................................................................92 4.1.2 Water Quality .....................................................................................................................................94 4.2 Discussion............................................................................................................................99

CHAPTER 5: CONCLUSIONS ..................................................................................... 107 CHAPTER 6: REFERENCES ....................................................................................... 109

UNIGIS Master of Science in GIS

FIGURES Figure 1: Development of Water Quality in Alaskaâ&#x20AC;&#x2122;s Kenai River project. ............ 15 Figure 2: Location of the Kenai Peninsula in relation to the State of Alaska, USA. .................................................................................................................................................... 41 Figure 3: Baseline project sampling stations along the Kenai River. ....................... 43 Figure 4: Flow chart of methodology for geographical analysis. ................................ 48 Figure 5: Six categories of land use available along the estuary and lower sections of the Kenai River. ................................................................................................... 51 Figure 6: Developed and undeveloped land in the estuary and lower sections of the Kenai River.............................................................................................................................. 52 Figure 7: Six categories of land use available along the middle section of the Kenai River. ..................................................................................................................................... 53 Figure 8: Developed and undeveloped land in the middle section of the Kenai River. ................................................................................................................................................... 54 Figure 9: Six categories of land use available along the middle section of the Kenai River. ..................................................................................................................................... 55 Figure 10: Developed and undeveloped land in the middle section of the Kenai River. ................................................................................................................................................... 56 Figure 11: Six categories of land use available along the upper section of the Kenai River. ..................................................................................................................................... 57 Figure 12: Developed and undeveloped land in the upper section of the Kenai River. ................................................................................................................................................... 58 Figure 13: Six categories of land use available along the upper section of the Kenai River. ..................................................................................................................................... 59 Figure 14: Developed and undeveloped land in the upper section of the Kenai River. ................................................................................................................................................... 60 Figure 15: Levels of arsenic in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 62 Figure 16: Levels of chromium in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 64 Figure 17: Levels of copper in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 66 Figure 18: Levels of lead in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 68 Figure 19: Levels of zinc in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 70 Figure 20: Levels of calcium in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 72 Figure 21: Levels of iron in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 74 Figure 22: Levels of magnesium in the Kenai River during the spring and summer sampling events. ....................................................................................................... 76 Figure 23: Levels of nitrate in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 78 Figure 24: Levels of phosphorus in the Kenai River during the spring and summer sampling events. ....................................................................................................... 80 UNIGIS Master of Science in GIS

Figure 25: Levels of fecal coliform bacteria in the Kenai River during the spring and summer sampling events. ............................................................................................. 83 Figure 26: Levels of pH in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 85 Figure 27: Levels of specific conductance in the Kenai River during the spring and summer sampling events. ............................................................................................. 87 Figure 28: Levels of total suspended solids in the Kenai River during the spring and summer sampling events. ............................................................................................. 89 Figure 29: Levels of turbidity in the Kenai River during the spring and summer sampling events. ........................................................................................................................... 91 Figure 30: Levels of water temperature in the Kenai River during the spring and summer sampling events. ....................................................................................................... 93 Figure 31: Cluster analysis of all the measured parameters. ....................................... 95 Figure 32: PCA performed using the entire set of data. Group 1 represents all the data collected in Spring and Group 2 represents data collected in summer. ............................................................................................................................................. 96 Figure 33: Cluster analysis of land use values and WQI. ............................................... 96 Figure 34: Overall water quality in the Kenai River during the spring and summer. ............................................................................................................................................. 98

UNIGIS Master of Science in GIS

TABLES Table 1: Water Quality ranges. ...................................................................................................... 22 Table 2: Summary of Alaska state and federal water quality standards for all parameters presented in this study. .................................................................................. 35 Table 3: Exact sampling locations of all 13 sites along the Kenai River mainstem. ......................................................................................................................................... 45 Table 4: Exact sampling locations of all 9 tributary sites. ............................................... 46 Table 5: Water quality ranges that have been modified in order to better adjust for the data in this project. ....................................................................................................... 49

UNIGIS Master of Science in GIS

CHAPTER 1: INTRODUCTION 1.1

Background It is widely accepted that life wouldn’t exist without water, yet water is a finite resource. Over 700 million people in 43 different countries suffer from water scarcity and these numbers could double by 2025 (WWAP, 2012). In other regions of the World, water scarcity is not yet a problem, in fact water is abundant. The World Resources Institute (n.d) developed a water risk mapping tool that helps understand where water problems are arising and opportunities that may be present. It’s important to mention that, in order for water to be available to sustain life, water quantity and quality need to meet minimum standards (World Resources Institute, n.d.).

Alaska is a unique and special place with thousands of rivers, lakes, wetlands, snowfields and glaciers. Alaska comprises 40% of the United States’

surface

waters

(USFWS,

2010).

With

this

mind

and

acknowledging the relatively young development history of the State, it is important to know and understand what the state of the waters is and what makes it change.

The Kenai River is a glacial stream located on the Kenai Peninsula of Alaska. Worldwide known as a prime fishing spot for many species of Pacific Salmon, Trout and Steelhead, the Kenai River receives thousands of anglers every year (ADEC, n.d). With salmon as the key species that drives the economy of the area, and the Kenai River and its watershed as the main spawning rivers of the peninsula, it is important to preserve and protect these ecosystems for generations to come (ADEC, n.d).

UNIGIS Master of Science in GIS

1.2

Motivation

As part of a collective effort to preserve the water quality in the Kenai River, the Kenai Watershed Forum (KWF) has consistently collected water quality parameters at 21 sites on the Kenai River since the year 2000 (KWF, 2015). Because all salmon species are very sensitive to changes in water temperature, turbidity, flow, conductivity, etc. it is important to know and control the water parameters in the river (KWF, 2015). Furthermore, for management purposes it is important to know which factors affect the water quality and how and where it affects it. It is important to note that water quality in lakes, streams and rivers can vary between different systems and within the same system; and it can be driven by multiple factors such as geology, hydrology, land use, etc. (Ashton, 1998). Management efforts of an entire water body are costly, time consuming and many times impracticable, which is why it is essential to be able to spatially localize and prioritize these efforts (Clark, 1970). This project was developed to provide land managers with a tool that will allow prioritization of different segments in the Kenai River based on the water quality in each segment. Figure 1 represents how this project will allow land managers to better target areas of concern in the Kenai River in order to prioritize efforts.

UNIGIS Master of Science in GIS

Figure 1: Development of Water Quality in Alaskaâ&#x20AC;&#x2122;s Kenai River project.

UNIGIS Master of Science in GIS

1.3

Objectives

1.3.1 General Objective – To determine which sections of the Kenai River (Alaska, United States of America) should be prioritized for management efforts, using temporal and spatial data collected by the KWF between years 2000 - 2014.

1.3.2 Specific Objectives •

To compare the field and laboratory measurements obtained in collected data to State and Federal Water Quality Standards.

•

To analyze the spatial differences of the river water quality.

•

To evaluate if seasonality plays a role in the water quality of the river.

1.4

Research Questions •

How is the water quality in the Kenai river?

•

Is the water quality in the Kenai river different during different seasons like spring and summer?

•

Can GIS be used to analyze and monitor water quality?

•

What is the advantage of using GIS as a complement of in situ water quality monitoring?

1.5

Hypothesis

The water quality in the Kenai River (Alaska, United States of America) is lower in areas of high development due to the high use of the river.

UNIGIS Master of Science in GIS

1.6

Justification The Kenai Peninsula accounts for about 8% of the total fisheries revenue of the State of Alaska (Knapp, 2012; University of Alaska Center for Economic Development, 2016). The commercial fish industry does not only consist of fishing; a very big sector of this industry is the fish processing (Knapp, 2012; University of Alaska Center for Economic Development, 2016). In 2010, a total of 2,863 people were employed in the fishing industry (Knapp, 2012; University of Alaska Center for Economic Development, 2016). This includes permit holders who fish plus commercial crew license holders (Knapp, 2012; University of Alaska Center for Economic Development, 2016). The estimated 2010 ex-vessel income by Kenai Peninsula Borough â&#x20AC;&#x201C; based fisherman was $122,140,353 (Knapp, 2012; University of Alaska Center for Economic Development, 2016). Ex-vessel is a technical term used specifically when talking about a commercial fisheries boat and the money it receives for the catch (ADFG, n.d.). In 2009, the seafood processing industry provided 1,846 jobs generating a total of $11,590,049 in wages

(Knapp,

2012;

University of

Alaska Center for Economic

Development, 2016). In addition to direct harvester and processor works, fisheries related jobs include fuel, accountants, consultants, air and water travel, air cargo, government jobs, tourism, etc. (Knapp, 2012; University of Alaska Center for Economic Development, 2016). In conclusion, the commercial fishing industry can be seen as a key component in the Kenai Peninsula Boroughâ&#x20AC;&#x2122;s economy, generating earnings that circulate in the local economy (Knapp, 2012; University of Alaska Center for Economic Development, 2016).

The Kenai Peninsula has experienced a 17% growth in its population over the last 15 years (United States Census Bureau, 2016). This trend is expected to continue its upward trajectory in the years to come (United States Census Bureau, 2016; University of Alaska Center for Economic Development, 2016). With an increase of population there is an expected increase in infrastructure to accommodate this increment (United States Census Bureau, 2016). In 2015 alone, over 74,000 square meters of built

UNIGIS Master of Science in GIS

space were added to the Kenai Peninsula (United States Census Bureau, 2016).

In order to protect the economy in the Kenai Peninsula, there is a need to protect its natural resources and more specifically its salmon populations. There are multiple efforts being done to regulate and control the fisheries of these species, as well as protecting the habitat and water they live in. But because of the size and the remoteness of the Kenai Peninsula it is difficult to allocate resources for restoration and conservation efforts. The Kenai River is the most heavily fished river in Alaska, with an average of 275,000 fishing days being recorded (ADEC, n.d). The river flows through private developed and undeveloped land, the Chugach National Forest, the Kenai National Wildlife Refuge, the cities of Soldotna and Kenai, among others. Due to the complexity of this system, it is necessary to develop a method that allows the assessment of water quality of the entire system and is able to provide water quality for different sections of the river which in turn will serve to better allocate resources for restoration. 1.7

Scope

This project analyzes the data of the river water quality collected between July 2000 and July 2014 from 21 sampling locations in the Kenai River mainstem and its tributaries. Local, state, federal, and tribal government entities, as well as several local area non-profits, formed a cooperative partnership so that sampling teams from various agencies were able to collect water samples twice per year, once in late April and once in late July. This effort continues beyond the timeline presented in this study. The locations of the sampling sites are identified with maps, GPS coordinates, and photographs. Trends in the data are highlighted, and the results are compared to the Alaska and Federal water quality standards for freshwater aquatic life.

The water quality data focuses on metals, nutrients, hydrocarbons, fecal coliform bacteria, and various field parameters. Arsenic, cadmium,

UNIGIS Master of Science in GIS

chromium, copper, lead, and zinc are the dissolved metals that have been analyzed, and calcium, iron, and magnesium were reported as total metals. Additionally, the study focuses on the nutrients nitrate and phosphorus. Specifically, the hydrocarbons that were collected and analyzed include diesel range organics, gasoline range organics, residual range organics, benzene, toluene, ethylbenzene, m,p-xylene, and o-xylene. Fecal coliform bacteria, pH, specific conductance, total suspended solids, turbidity, and water temperature are the remaining parameters that have been included in the analysis. Utilizing geographic information from the Kenai Peninsula Borough, United States Geological Survey (USGS) and the KWF, spatial analysis is conducted to better understand the variability in water quality in the Kenai River.

UNIGIS Master of Science in GIS

CHAPTER 2: LITERATURE REVIEW 2.1 Basic Concepts 2.1.1 Water Quality Standards Water Quality Standards are the maximum allowable standards of pollutant present in water, and are set by the United States Environmental Protection Agency (USEPA) as mandated by the Clean Water Act. By setting water quality standards the protection of waterbodies health can be attained (USEPA, 2015c).

2.1.2 National Hydrography Dataset There are many Federal, State, and local agencies as well as academia and private parties conducting research and monitoring of surface water. This leads to multiple sets of data being managed by different people who have different objectives, jurisdictions, etc. For better management of water resources, the data generated by all parties should be integrated and shared using a universal data linking system.

In the United States of

America, the National Hydrography Dataset (NHD) is the common national framework for surface-water geospatial information. The USGS utilizes a linear referencing system to link information, collected by surface water researchers and professionals, to the NHD allowing the information provided in a particular study to be linked to a comprehensive national flow network which in turn will allow for more powerful analysis providing historical information, flow direction, dams, etc.

The NHD and Watershed Boundary Dataset (WBD) are used to store information, describe, and illustrate surface water on The National Map (USGS, 2014, 2015). “The NHD represents the drainage network with features such as rivers, streams, canals, lakes, ponds, coastline, dams, and stream gages” (USGS, 2015, ¶1). On the other hand, “the WBD represents

UNIGIS Master of Science in GIS

drainage basins as enclosed areas in eight different size categoriesâ&#x20AC;? which are then assigned a hydrologic unit code (HUC) (USGS, 2014, 2015, Âś1).

The NHD and WBD are resources often used by GIS professionals since it is the official database for hydrologic information for the United States. An addressing system based on reach codes and linear referencing are used to link specific information about the water such as water discharge rates, water quality, and fish population (USGS, 2014, 2015).

2.1.3 Water Quality Index Water Quality Index (WQI) is a tool used by many environmental and water resource scientists and professionals to summarize and describe the quality of a water body. There is no WQI that could be applied or used for every water body, in fact there are many water quality indices that have been developed based on the specific needs of the study being conducted and are therefore limited in the application and use (Srebotnjak, Carr, de Sherbinin, & Rickwood, 2012; Stoner, 1978).

As previously mentioned, WQIs are a tool developed to summarize and evaluate water quality in a scientific defendable way as well as to simplify the writing and reporting of water quality information. WQIs are developed based on the needs of a particular study and on the geographic area they will be applied to. Furthermore, the parameters used to calculate a WQI also depend on the goals and objectives of the index. Hence, there are some parameters that are widely accepted as the basic parameters needed to calculate the index: Biochemical Oxygen Demand (mg/L), Dissolved Oxygen (%), Temperature (Celsius degrees), Fecal Coliform (# of colonies/100 ml), Nitrate (mg/L), Phosphorus (mg/L), Turbidity (NTU), Total Suspended Solids (mg/L) and pH (ph units) (North Carolina Division of Water Quality Watershed Assessment Team, 2009; Rickwood & Carr, 2007; Litchfield & Kyle, 1992; Simeonov et al., 2003; Srebotnjak et al., 2012).

UNIGIS Master of Science in GIS

In general, for the calculation of most WQI there is a normalization or standardization

each

parameter

according

the

expected

concentrations of such parameter in the system being analyzed and an interpretation of good or bad concentrations (Rickwood & Carr, 2007; Srivastava & Kumar, 2013; Stoner, 1978). All parameters are also weighted according to their perceived importance of the overall quality of the water. The index will most likely be a number that falls between 0 and 100 and qualitative values will be assigned to specific ranges prior calculation of the index (Rickwood & Carr, 2007; Srivastava & Kumar, 2013; Stoner, 1978). Table 1 shows the water quality ranges valid for incomplete sets of data. In this case, the water quality will be one of five possible categories depending on the value of the calculated index.

Index Ranges

Water Quality

0-25

Very Bad

25-50

Bad

50-70

Medium

70-90

Good

90-100

Excellent

Table 1: Water Quality ranges. Source: Srivastava & Kumar (2013)

UNIGIS Master of Science in GIS

2.2 Water Quality

Many governments, world organizations, academics, and the general public agree that freshwater is not only a finite resource but it is essential for life on Earth. Without enough water of good quality health risks can arise rapidly, sustainable development is impossible and we could damage or even loose industries such as agriculture, fisheries, tourism, etc. (Bartram & Ballance, 1996; UNEP, 2008; UNW, 2012; USEPA, 2013). The Kenai River water quality, like every other river, is driven by natural factors such as climate, geology, vegetation, etc., as well as by human activities (Brabets, Nelson, Dorava, & Milner, 1999).

2.2.1 Dissolved Metals

2.2.1.1 Arsenic

Natural sources of arsenic in the Cook Inlet Basin, in South Central Alaska USA, include volcanic ash, glaciation, and mineral deposits, and only a minimal contribution of arsenic results from human activities like wood preservation (Agency for Toxic Substances and Disease Registry, 2007c). Arsenic is naturally present as a compound in rocks within the Kenai River Watershed, and as a dissolved metal, it can be acutely or chronically toxic to fish (Glass, 1999, 2001; Glass & Frenzel, 2001). The Alaska Department of Environmental Conservation (ADEC) and the USEPA have set the standard at 150 micrograms per liter (Âľg/L) for freshwater aquatic life chronically exposed to arsenic (ADEC, 2008; USEPA, 2014).

UNIGIS Master of Science in GIS

2.2.1.2 Cadmium

Cadmium is a rare elemental metal that can occur naturally in freshwater at concentrations

less

than

0.1

µg/L,

but

slightly

increased

concentrations, it can be toxic to aquatic life (USEPA, 2000). Additional cadmium can enter the hydrologic cycle as a component of fertilizer, pesticide, pigment, and as a result of iron and steel production, coal combustion, and mining waste (USEPA, 2000). The ADEC and the USEPA have set the standard for cadmium at a range of 0.0650 µg/L to 0.64 µg/L, depending on hardness, for chronically exposed freshwater aquatic life (ADEC, 2008; USEPA, 2014).

2.2.1.3 Chromium

In rivers and streams, chromium is an elemental metal that typically exists as hexavalent or trivalent chromium (USEPA, 1980a). Non-natural sources of chromium salts include the metal finishing industry, textile manufacturing, leather tanning, paint, fungicides, and wood preservatives (USEPA, 1980a). At a concentration of 21 µg/L of hexavalent chromium, river algae cannot photosynthesize, and the growth in weight of Chinook salmon can be reduced by approximately ten percent at a concentration of 16 µg/L (USEPA, 1980a). Consequently, the ADEC and the USEPA have set the standard for hexavalent chromium at 11 µg/L and at a range of 15.52 µg/L to 230.67 µg/L for trivalent chromium, depending on hardness, for chronically exposed freshwater aquatic life (ADEC, 2008; USEPA, 2014).

2.2.1.4 Copper

Typically, present in surface waters, naturally low concentrations of copper are essential as micronutrients for plants and animals, but elevated levels can be toxic to certain aquatic species (USEPA, 2007). Concentrations of copper can increase in surface waters due to discharges from mining, leather industry, electrical equipment, and fabricated metal products

UNIGIS Master of Science in GIS

(USEPA, 2007). Copper is present in the brake pads of vehicles and can enter surface waters in storm water runoff (USEPA, 2015a). The standard for copper set by the ADEC, ranges from 1.75 µg/L to 29.28 µg/L, depending on hardness, for chronic exposure to aquatic life in freshwater (ADEC, 2008).

2.2.1.5 Lead

Metallic lead and lead minerals are not classified as soluble in water, but they can be solubilized by certain acids, and a selection of industrial lead compounds are water soluble upon production (USEPA, 1980b). Lead is a component in electroplating, metallurgy, construction materials, plastic, and electronic equipment, and it can enter surface water through precipitation, dust, street runoff, and wastewater discharges (USEPA, 1980b). Exposure to lead can cause delayed embryonic development, reduced growth, and suppressed reproduction in fish, and spinal deformities in rainbow trout fry (USEPA, 1980b). Determined by the ADEC and the USEPA, the standard for lead ranges from 0.300 to 10.94 µg/L, depending on hardness, for the chronic exposure of aquatic life in freshwater (ADEC, 2008; USEPA, 2014).

2.2.1.6 Zinc

Zinc enters surface water naturally through the weathering of bedrock and is an essential micronutrient for all plants and animals (USEPA, 1987). Additional zinc can enter surface water because it is widely used for galvanizing steel, as an alloy, in rubber, and in paint (USEPA, 1987). More recent studies suggest that higher concentrations of Zinc in suburban areas can be caused by tire wear (Councell, Duckenfield, Landa, & Callender, 2004). The ADEC and the USEPA have set the standard for zinc ranging from 23.4 to 382.4 µg/L, depending on hardness, for the chronic exposure of freshwater aquatic life (ADEC, 2008; USEPA, 2014).

UNIGIS Master of Science in GIS

2.2.2 Total Metals

2.2.2.1 Calcium

Both plants and animals require calcium as an essential element, and calcium is important to water quality because it is directly linked to the hardness of water (Glass, 2001). There is no ADEC or USEPA standard for the chronic exposure of freshwater aquatic life to calcium.

2.2.2.2 Iron

Naturally present in many rocks and soils, iron is required by plants and animals for metabolism (Glass, 2001). Sources of detrimental levels of iron are industrial waste, mining, and iron-rich groundwater, and when high concentrations of iron react with dissolved oxygen, precipitates form that can harm salmon eggs and other aquatic life (USEPA, 1976). The ADEC and the USEPA have set the iron standard for the chronic exposure of freshwater aquatic life at 1 mg/L (ADEC, 2008; USEPA, 2014).

2.2.2.3 Magnesium

As an essential element, magnesium concentrations can significantly impact plants and animals, and also influences water quality because it contributes to the water hardness (Glass, 2001). Neither the USEPA nor the ADEC has set a water quality standard for magnesium concentrations in freshwater for chronically exposed aquatic life.

UNIGIS Master of Science in GIS

2.2.3 Nutrients

2.2.3.1 Nitrate

Inorganic nitrogen is generally present in well-aerated, natural streams in the form of nitrate (Glass, 1999). Nitrate enters the hydrologic cycle as a result of precipitation, plant residue, natural minerals, fertilizer, and septic tanks (Glass, 2001). Neither the USEPA nor the ADEC have set a nitrate standard for chronically exposed freshwater aquatic life.

2.2.3.2 Phosphorus

Plants and animals require the essential element phosphorus for growth, and most concentrations are not toxic to aquatic or human life (Glass, 1999). As a nutrient, phosphorus is significant to water quality because in high concentrations, it can lead to excessive algal growth and eutrophic conditions (Glass, 1999). The ADEC and the USEPA do not have a standard for phosphorus for chronic exposure to aquatic life in freshwater.

2.2.4 Hydrocarbons

2.2.4.1 Diesel Range Organics

Diesel range organics (DRO) consist of diesel fuels and associated byproducts, and include the n-alkane range from C10 to C25 (Geosphere Inc, 2006). Although the ADEC and the USEPA do not have a standard for the chronic exposure of freshwater aquatic life to DRO, it was included in this study to cover a broad range of hydrocarbons and narrow down potential sources.

UNIGIS Master of Science in GIS

2.2.4.2 Gasoline Range Organics

Gasoline Range Organics (GRO) are comprised of gasoline fuels and byproducts, including benzene, toluene, ethylbenzene, and xylene (BTEX), and GRO corresponds to the alkane range of C6 to C10 (Geosphere Inc, 2006). Although there is an ADEC standard specifically for BTEX, there is not an ADEC or USEPA standard applicable to GRO as a whole for the chronic exposure of freshwater aquatic life.

2.2.4.3 Residual Range Organics

Heavy fuel products, like asphalt or Bunker C fuel, are referred to as residual range organics (RRO) and include the n-alkane range of C25 to C36 (Geosphere Inc, 2006). RRO was included in this study to cover a broad range of hydrocarbons, however, there is no ADEC or USEPA standard for freshwater aquatic life chronically exposed to RRO.

2.2.4.4 Total BTEX

Benzene, toluene, ethylbenzene, m,p-xylene, and o-xylene are the aromatic hydrocarbons that are commonly referred to as BTEX. Volcanoes and forest fires are natural sources of benzene, and benzene is also a component in cigarette smoke, crude oil, and gasoline (Agency for Toxic Substances and Disease Registry, 2007a). Benzene breaks down slowly in water and soil, but concentrations do not build up in plants and animals (Agency for Toxic Substances and Disease Registry, 2007a). A natural part of crude oil, toluene is used to produce gasoline, paint, paint thinner, lacquer, and adhesives (Agency for Toxic Substances and Disease Registry, 2001). Toluene can enter surface water due to petroleum or solvent spills and engine exhaust, but does not build up to high concentrations in animals (Agency for Toxic Substances and Disease Registry, 2001). Ethylbenzene is a colorless liquid found naturally in coal tar and petroleum, and is often used as a solvent and as a component in fuel, ink, insecticide, and paint (Agency for Toxic Substances and Disease Registry, 2007b). Xylene has

UNIGIS Master of Science in GIS

three forms or isomers called meta-xylene, para-xylene, and ortho-xylene, which are reported in this study as m,p-xylene and o-xylene. As a natural part of petroleum and coal tar, xylene has many uses as a solvent, cleaner, paint thinner, and a small component of gasoline (Agency for Toxic Substances and Disease Registry, 2007c). Once xylene enters surface water, it evaporates quickly into the air, but a small amount can concentrate in aquatic life (Agency for Toxic Substances and Disease Registry, 2007c). The ADEC standard is 10 Âľg/L for total aromatic hydrocarbons in freshwater for the growth and propagation of fish, shellfish, other aquatic life, and wildlife (ADEC, 2012). Additionally, the ADEC requires that surface waters do not have any floating oil, film, sheen, or discoloration due to hydrocarbons and that in shoreline or bottom sediments, there cannot be any concentration of petroleum hydrocarbons that cause deleterious effects to aquatic life (ADEC, 2012).

2.2.5 Fecal Coliform Bacteria

The presence of fecal coliform bacteria in surface water indicates fecal contamination from warm-blooded animals, which is linked to diseasecausing viruses and bacteria (Glass, 1999). Sources of fecal coliform bacteria include waste from septic systems, domestic animals, waterfowl, and other wildlife (Glass, 1999). The ADEC and USEPA standards for fecal coliform bacteria have two types of criteria, a 30-day geometric mean and a no more than 10% of the samples can exceed a specified value criteria. The geometric mean criterion was not evaluated in this study because not enough samples were collected during any 30-day period. For reference, the ADEC fecal coliform drinking water standard states that in a 30-day period, the geometric mean of samples may not exceed 20 CFU/100ml and not more than 10% of the total samples may exceed 40 CFU/100ml (ADECâ&#x20AC;&#x2122;s single sample limit). The ADEC fecal coliform secondary recreation standard states that in a 30-day period, the geometric mean of samples may not exceed 200 CFU/100ml and not more than 10% of the total samples may exceed 400 CFU/100ml (ADEC, 2012).

UNIGIS Master of Science in GIS

2.2.6 pH

The concentration of hydrogen-ion activity is represented by a pH measurement on a logarithmic scale ranging from 0 to 14 with acidic conditions closer to 0 and alkaline conditions closer to 14 (Glass, 1999). Values of pH ranging from 6.5 to 8.0 are typical in unpolluted river water, and these values can be affected by both natural and human processes (Glass, 1999). The ADEC standard for pH maintains that freshwater may not have a pH less than 6.5 or greater than 8.5 for the growth and propagation of fish, shellfish, other aquatic life and wildlife (ADEC, 2012). Additionally, the standard stipulates that the pH may not vary more than 0.5 pH units from natural conditions and that in any water with a pH naturally outside the specified range, variation of pH must be toward the specified range (ADEC, 2012).

2.2.7 Specific Conductance

The capacity of water to conduct electricity is measured by specific conductance, and the level of specific conductance also correlates with the concentration of calcium, dissolved solids, and the waterâ&#x20AC;&#x2122;s hardness (Glass, 1999). There is not currently an ADEC or USEPA standard for specific conductance for aquatic life in freshwater.

2.2.8 Total Suspended Solids

The concentration of total suspended solids is a way of measuring the amount of mineral and organic particles that are transported in the water column (Bash, Berman, & Bolton, 2001). Erosion increases the amount of suspended sediment in surface water and can be naturally caused by glaciers, fires, and floods (Glass, 1999). In the Cook Inlet Basin, glaciers are the main cause of erosion, but mining, logging, construction, and recreation can also contribute to elevated levels of suspended sediment (Glass, 1999). High concentrations of suspended sediment can be lethal to post-larval fish, and incubating eggs can suffer severe mortality rates UNIGIS Master of Science in GIS

because settled sediment prevents the exchange of oxygen (Glass, 1999). As such, the ADEC standard requires that fine sediment in the range of 0.1 mm to 4.0 mm cannot increase over 5% by weight higher than natural conditions in gravel beds used by anadromous or resident fish for spawning, and the sediment can never exceed a maximum of 30% by weight in freshwater for the growth and propagation of fish, shellfish, other aquatic life, and wildlife (ADEC, 2012). More generally, the ADEC standard also states that the amount of deposited or suspended sediment cannot cause adverse effects on aquatic life, including their habitat and reproduction (ADEC, 2012).

2.2.9 Turbidity

Turbidity measures the degree to which suspended and dissolved materials cause light to scatter instead of being transmitted in straight lines, and these materials consist of silt, clay, chemicals, microscopic organisms, and fine organic or inorganic matter (Bash, Berman & Bolton, 2001). Turbidity varies naturally; higher turbidities are typically found in watersheds fed by glacial melt water, and turbidity generally increases from headwater tributaries to mainstems and estuaries (Bash et al., 2001). Even though salmonids can naturally live in turbid water systems, they do not always cope well with increases in suspended sediments, and these high levels in suspended sediments can cause fatalities, while lower levels can result in difficulty finding food, reduced growth, increased stress, and difficulty migrating (Bash et al., 2001). The ADEC standard states that turbidity may not exceed 25 Nephelometric Turbidity Units (NTU) above natural conditions in freshwater or 5 NTU above natural conditions in lakes for the growth and propagation of fish, shellfish, other aquatic life, and wildlife (ADEC, 2012).

2.2.10 Water Temperature

Water temperature varies seasonally and as a result of glacial activity and anthropogenic sources. Low water temperatures between 0Â°C and 4Â°C can result in low growth rates for fish, but much higher water temperatures can UNIGIS Master of Science in GIS

encourage disease, competitors, predators, mortality, and an oxygendeprived habitat (Kyle & Brabets, 2001). The ADEC has five temperature standards for the growth and propagation of fish, shellfish, other aquatic life, and wildlife in freshwater; the water temperature must remain below 20°C in all areas, below 15°C in rearing areas and migration routes, and below 13°C in egg and fry incubation and spawning areas (ADEC, 2012).

Table 2 summarizes Alaska state and federal water quality standards for all parameters presented in the previous sections. Some parameters don’t have a federal standard but the State of Alaska has developed a standard based on the minimum physiological requirements of sensitive species.

UNIGIS Master of Science in GIS

Parameter Arsenic

ADEC Standard 150 μg/L for aquatic life, fresh water, and chronic exposure.

Cadmium

(e7.409(ln hardness)-4.719) (1.1101672[(ln hardness) (0.041838)]) for aquatic life, fresh water, and chronic exposure.

Chromium III

(e0.819(ln hardness)+0.6848) (0.860) for aquatic life, fresh water, and chronic exposure.

Chromium VI

11 μg/L for aquatic life, fresh water, and chronic exposure.

Copper

(e0.8545(ln hardness)-1.702)(.960) for aquatic life, fresh water, and chronic exposure. (e1.273(ln hardness)-4.705) (1.46203[(ln hardness)(0.145712) for aquatic life, fresh water, and chronic exposure.

Lead

Zinc

(e0.8473(ln hardness)+0.884)(0.986) for aquatic life, fresh water, and chronic exposure.

Calcium

None applicable to aquatic life, fresh water, and chronic exposure

Iron

1 mg/L for aquatic life, fresh water, and chronic exposure.

Magnesium

None applicable to aquatic life, fresh water, and chronic exposure.

Nitrate

None applicable to aquatic life, fresh water, and chronic exposure.

UNIGIS Master of Science in GIS

USEPA Standard 150 μg/L for a priority toxic pollutant, fresh water, and Criterion Continuous Concentration (CCC). (e7.409(ln hardness)-4.719) (1.1101672[(ln hardness) (0.041838)]) for a priority toxic pollutant, fresh water, and CCC. (e0.819(ln hardness)+0.6848) (0.860) for a priority toxic pollutant, fresh water, and CCC. 11 μg/L for a priority toxic pollutant, fresh water, and CCC. See EPA, 2007. (e1.273(ln hardness)-4.705) (1.46203[(ln hardness)(0.145712) for a priority toxic pollutant, fresh water, and CCC. (e0.8473(ln hardness)+0.884)(0.986)

for a priority toxic pollutant, fresh water, and CCC. None applicable to a priority or non-priority pollutant, fresh water, and CCC. 1 mg/L for a non-priority pollutant, fresh water, and CCC. None applicable to a priority or non-priority pollutant, fresh water, and CCC. None applicable to a priority or non-priority pollutant, fresh water, and CCC. 33

Parameter Phosphorus

ADEC Standard None applicable to aquatic life, fresh water, and chronic exposure.

Diesel Range Organics

None applicable to aquatic life, fresh water, and chronic exposure.

Gasoline Range Organics

None applicable to aquatic life, fresh water, and chronic exposure.

Residual Range Organics

None applicable to aquatic life, fresh water, and chronic exposure.

BTEX

For fresh water growth and propagation of fish, shellfish, other aquatic life, and wildlife: total aromatic hydrocarbons (TAH) in the water column may not exceed 10 Îźg/L. There may be no concentrations of petroleum hydrocarbons, animal fats, or vegetable oils in shoreline or bottom sediments that cause deleterious effects to aquatic life. Surface waters and adjoining shorelines must be virtually free from floating oil, film, sheen, or discoloration. Not applicable to the sampling methods used in this study For fresh water growth and propagation of fish, shellfish, other aquatic life, and wildlife: May not be less than 6.5 or greater than 8.5. May not vary more than 0.5 pH units from natural conditions None applicable to the aquatic life, fresh water, and chronic exposure.

Fecal Coliform Bacteria pH

Specific Conductanc e

UNIGIS Master of Science in GIS

USEPA Standard None applicable to a priority or non-priority pollutant, fresh water, and CCC. None applicable to a priority or non-priority pollutant, fresh water, and CCC. None applicable to a priority or non-priority pollutant, fresh water, and CCC. None applicable to a priority or non-priority pollutant, fresh water, and CCC. No comparable standard for a priority or non-priority pollutant, fresh water, and CCC.

Not applicable to the sampling methods used in this study. 6.5-9 for a non-priority pollutant, fresh water, and CCC.

None applicable to a priority or non-priority pollutant, fresh water, and CCC.

Parameter Total Suspended Solids

ADEC Standard For fresh water growth and propagation of fish, shellfish, other aquatic life, and wildlife: The percent accumulation of fine sediment in the range of 0.1 mm to 4.0 mm in the gravel bed of waters used by anadromous or resident fish for spawning may not be increased by more than 5% by weight above natural conditions. In no case, may the 0.1 mm to 4.0 mm fine sediment range in those gravel beds exceed a maximum of 30% by weight.

USEPA Standard For fresh water fish and other aquatic life: Settleable and suspended solids should not reduce the depth of the compensation point for photosynthetic activity by more than 10% from the seasonably established norm for aquatic life.

Turbidity

For fresh water growth and propagation of fish, shellfish, other aquatic life, and wildlife: May not exceed 25 NTU above natural conditions. For all lake waters, may not exceed 5 NTU above natural conditions.

Water Temperatur e

For fresh water growth and propagation of fish, shellfish, other aquatic life, and wildlife: May not exceed 20°C at any time. The following maximum temperatures may not be exceeded, where applicable: Migration routes 15°C Spawning areas 13°C Rearing areas 15°C Egg & fry incubation 13°C

For fresh water fish and other aquatic life: Settleable and suspended solids should not reduce the depth of the compensation point for photosynthetic activity by more than 10% from the seasonably established norm for aquatic life. Not applicable in Alaska to a non-priority pollutant, fresh water, and CCC.

Table 2: Summary of Alaska state and federal water quality standards for all parameters presented in this study.

UNIGIS Master of Science in GIS

2.3 Water Quality and GIS analysis

The complexity of hydrological processes is one of the limiting factors for water quality and water quantity studies (Khairy, Hannoura, Cothren, & McCorquodale, 2000). When water quality problems arise, there are several widely-accepted techniques and tools that are used to minimize and suppress such problems. Models have become an essential tool for land and resource managers to assess the quality of water in streams and lakes, and allow for adaptive, real-time, management efforts (Chang, 2008; Khairy et al., 2000; Quinn & Hanna, 2003; Tsihrintzis, Hamid, & Fuentes, 1996). Models are often used to identify water quality problem areas and to evaluate the effectiveness of hypothetical solutions (Fan, Fleischmann, Collischonn, Ames, & Rigo, 2015; Yan et al., 2015). Depending on the type of questions that need to be answered, based on the scale, on the data and on the desired output there are several widely-used models: QUAL2kw, water quality analysis simulation program (WASP), single factor pollution index (SFPI), among others (Fan et al., 2015; Pelletier, Chapra, & Tao, 2006; USEPA, 2015b; Yan et al., 2015).

These are effective tools for

identifying problem areas of water quality; however, the use of these models is limited to a small region water system and is difficult to apply on large watershed systems (Fan et al., 2015; Pelletier et al., 2006; USEPA, 2015b; Yan et al., 2015).

Other models such as the soil and water assessment tool (SWAT) or spatially referenced regressions on watershed attributes (SPARROW) have been developed to work at larger scales and are able to assess the water quality in large basins (Gassman, Reyes, Green, & Arnold, 2007; Schwarz, Hoos, Alexander, & Smith, 2006). These types of models are limited by very specific data requirements, high speed technology needs and necessary human expertise to run the models (Gassman et al., 2007; Schwarz et al., 2006).

UNIGIS Master of Science in GIS

Aside from the above presented models there are other tools used to assess the quality of a water body, especially when not monitoring industrial and sewage treatment plants. For nonpoint source pollution management, it is of critical importance to maximize the use of the available data to derive the water quality in unmonitored locations and optimize the monitoring network (Strager et al., 2010; Tsihrintzis et al., 1996; Xu, Xu, Wu, & Tang, 2012; Zhou & Zhao, 2011).

One viable option, although not necessarily geographic, is to estimate empirical relationships between watershed characteristics and water quality parameters. Understanding these relationships could help assess the condition of unmonitored locations and identify human actions that could be related to the water quality. Watershed characteristics that have been used for such studies are slope, land use and land cover, population density, road density, etc. However, the main method used for estimating the relationship between watershed characteristics and water quality is the Ordinary Least Square, which requires all observations to be independent from each other. This requirement is not met due to the spatial correlation between sampling events. A similar analysis to the ordinary least square that takes the spatial correlation into account is the spatial regression method. This method has been used in conjunction with geostatistical methods such as kriging, and provides a very promising alternative for studying the impact of watershed characteristics to water quality (Tsihrintzis et al., 1996; Yang & Jin, 2010).

A different approach is presented by Yan et al. in his study in the Honghe River Watershed. This study seeks not only to assess the water quality at a watershed level but also to have the ability to identify polluted risky areas. With a study area of over 12,300 Km2, the authors desigend the study based on 67 sampling stations in the entire watershed and and six water quality parameters. Using a single factor pollution index and ordinary Kriging spatial interpolation, this study shows that the GIS approach can be successfully applied to evaluate the water quality in a River (Yan et al., 2015).

UNIGIS Master of Science in GIS

A similar approach is presented by Kadhem in his study in the River Tigris. The study is set up to assess the water quality in an entire river based on the chemical analysis results of 8 monitoring stations. Using Inverse Distance Weighted for spatial interpolation of water parameters, determining cell values using a linear weighted combination set of sample points. The weight assigned is a function of the distance of an input point from the output cell location, in other words, the greater the distance, the less influence the cell has on the output value (Kadhem, 2013).

In 2000, Palacios completed her study analyzing the potential of dynamic segmentation for aquatic ecosystem management. Her study provided a descriptive resource management tool and since conception it has evolved in a more prescriptive analysis tool, utilizing historical data, utilizing social and habitat data among others, allowing managers to make better and more informed decisions. Aside from the results of her study, perhaps the most valuable contribution is that she began the shift of the GIS tool from being a solely mapping descriptive tool to showing the immense potential it has to be a prescriptive tool performing spatial analysis and modelling, thus assisting resource managers in the decision-making process.

In addition to water quality parameters, it is important to include geographic information to the analysis. This will allow managers evaluate possible sources of contamination as well as prioritize resources for restauration and conservation purposes (Flynn, 1999).

2.3.1 Linear referencing

Storing geographic locations by using their relative location along a measured line feature is known as linear referencing (ESRI, 2010). The way it works is that this method makes it possible to locate a certain event in a linear feature based on the distances measured in such feature; furthermore, this method allows to associate multiple sets of attributes to a linear feature without splitting such feature each time the value changes (ESRI, 2010).

UNIGIS Master of Science in GIS

2.3.2 Dynamic Segmentation

The process of transforming events (or linearly referenced data) that are stored in a table into a feature that can be visualized and analyzed in a map is known as dynamic segmentation (Cadkin, 2002). In other words, dynamic segmentation allows to change the values of attributes in a line feature without needing to split such feature (ESRI, 2010). To run a dynamic segmentation process there are two requirements on the data. The first requirement is that each event in an event table must have a unique identifier and position along a linear feature. The second requirement is that each linear feature must have a unique identifier and measurement system (Cadkin, 2002).

2.3.3 Inverse Distance Weighted Interpolation

This spatial interpolation method is widely used because the interpretation of its results is simple and very informative. This method operates under two assumptions, the first one is that the value of an un-sampled point is equivalent to the weighted average of the known values that are located in the neighboring area. The second assumption is that the weights of the measured values are inversely related to the distances of the predicted (Bartier & Keller, 1996; Lu & Wong, 2008; Zimmerman, Pavlik, Ruggles, & Armstrong, 1999).

UNIGIS Master of Science in GIS

CHAPTER 3: METHODOLOGY 3.1 Study Area

Located in southcentral Alaska, the Kenai River is part of the Cook Inlet Basin (inlet to the Pacific Ocean) and is intricately linked to the surrounding communities through sport and commercial fishing, tourism, recreation, and the propagation of fish and wildlife. Figure 2 shows the location of the Kenai River watershed and the Kenai River, where this study took place as well as the location of the Kenai Peninsula in relation to the State of Alaska, USA.

UNIGIS Master of Science in GIS

Figure 2: Location of the Kenai Peninsula in relation to the State of Alaska, USA.

UNIGIS Master of Science in GIS

Running 82 river miles westward from the Kenai Lake, through the Kenai National Wildlife Refuge, the Kenai River is a melt water river that drains the Central Kenai Peninsula. With a watershed of roughly 5,200 Km2, five species of pacific salmon flourish in the Kenai River Watershed, comprising 30% of the commercial Chinook harvest and 40% of the commercial sockeye harvest (Glass, 1999). Surface runoff, groundwater composition, natural minerals, aquatic plants and animals, and human activities can affect the water quality in this area (Glass, 1999). Potential sources of pollution from humans include gasoline powered boat engines, agriculture, mining, street runoff, and perforated septic tanks. For this study, the Kenai River was divided into four reaches, the Estuary (No Name Creek to River Mile 6.5) the Lower Kenai River (River Mile 6.5 to 21), the Middle Kenai River (River Mile 21 to 50) and the Upper Kenai River (River Mile 50 to 82).

3.1.1 Sampling Locations

Water samples were collected at 13 locations along the Kenai River mainstem and from 9 tributaries near their confluence points (see Figure 3). These locations were chosen by dozens of participants in order to accurately represent the Kenai River Watershedâ&#x20AC;&#x2122;s ambient water quality conditions (Ashton, 1998). Because of the different sources of funding for this project and the variety of organizations participating in this effort, the sampling station at Juneau Creek was only sampled four times during the projectâ&#x20AC;&#x2122;s timeline. Due to the small sample size for Juneau Creek, the data collected at this station was not used for the analysis presented in this study, making the total sampling stations 13 in the mainstem and 8 in the tributaries. Sampling occurred in late April and late July each year beginning in July 2000 and is still an ongoing effort, although for this study data collected until July 2014 was used.

UNIGIS Master of Science in GIS

Figure 3: Baseline project sampling stations along the Kenai River.

UNIGIS Master of Science in GIS

The Kenai River mainstem

Mile 82 site is near the Kenai Lake Outlet and Kenai Lake Bridge. Samples are typically collected downstream of the boat launch. Mile 70 site is near Jim’s Landing. The sample is typically collected 40 feet downstream of the boat launch. Mile 50 site is near the Skilak Lake Outflow. Samples are typically collected between the swan signs off of the south bank. Mile 43 site is upstream of Dow Island. The samples are typically collected 100 feet upstream of the point of Dow Island. Mile 40 site is near Bings Landing. Samples are typically collected in front of the boat launch near the center of the river. Mile 31 site is near Morgan’s Landing. Sampling typically occurs down the abandoned steep road behind the headquarters building. Mile 23 site is near Swiftwater Park. Samples are typically collected mid-channel in front of the ramp. Mile 21 This site is near the Soldotna Bridge. Samples are typically collected 20 feet downstream of the bridge on the south bank. Mile 18 site is near Poacher’s Cove. Samples are typically collected midchannel just downstream of an island. Mile 12.5 site is near the Pillars Boat Launch. Samples are typically collected toward the center of the river across from the dock. Mile 10.1 site is upstream of Beaver Creek. Samples are typically collected 200 yards upstream of the Beaver Creek and Kenai River confluence. During July 2000, April 2001, and July 2001, samples were collected downstream of the Kenai River and Beaver Creek confluence, and no samples were collected from this site in April 2002. Mile 6.5 site is near Cunningham Park. Sampling typically occurs straight out from the public-use boardwalk and can vary due to the tidal stage. Mile 1.5 site is near the City of Kenai Dock. Samples are typically collected at the north end of the public fueling dock. For exact GPS coordinates see Table 3.

UNIGIS Master of Science in GIS

Site

Longitude

Latitude

Mile 82

60.492007 N

149.810844 W

Mile 70

60.481392 N

150.115020 W

Mile 50

60.467517 N

150.507789 W

Mile 43

60.489844 N

150.636905 W

Mile 40

60.515441 N

150.702069 W

Mile 31

60.498284 N

150.863121 W

Mile 23

60.480338 N

151.030847 W

Mile 21

60.476634 N

151.082099 W

Mile 18

60.502005 N

151.106973 W

Mile 12.5

60.533743 N

151.099258 W

Mile 10.1

60.539279 N

151.142263 W

Mile 6.5

60.5408100 N

151.182780 W

Mile 1.5

60.543680 N

151.222940 W

Table 3: Exact sampling locations of all 13 sites along the Kenai River mainstem.

Kenai River Tributaries

At the Juneau Creek station, the sample is typically collected 40 feet downstream of the boat launch at Alaska Wildlands. At the Russian River site, samples are typically collected 90 feet upstream of the sanctuary sign. At the Killey River site, sampling typically occurs 100 yards upstream from the Kenai River confluence across from the fish table. At the Moose River station, sampling typically occurs upstream of the parking area. At the Funny River site, samples are typically collected 75 feet downstream of the bridge. At the Soldotna Creek site, sampling typically occurs mid-channel. At the Slikok Creek station, samples are typically collected in the midchannel. At Beaver Creek samples are typically collected approximately 100 yards upstream from the Kenai River confluence. At No Name Creek site, samples are typically collected approximately 500 feet upstream of the confluence with the Kenai River, just upstream of the footbridge. Precise coordinates are provided in Table 4.

UNIGIS Master of Science in GIS

Site

Longitude

Latitude

Juneau Creek

60.481392 N

150.115020 W

Russian River

60.484622 N

149.993955 W

Killey River

60.481518 N

150.632498 W

Moose River

60.536870 N

150.754724 W

Funny River

60.489963 N

150.860982 W

Soldotna Creek

60.483364 N

151.057656 W

Slikok Creek

60.482318 N

151.127053 W

Beaver Creek

60.548029 N

151.143240 W

No Name Creek

60.550888 N

151.268417 W

Table 4: Exact sampling locations of all 9 tributary sites.

3.2 Sample Collection and Laboratory Analysis

After a half-day training session, staff from governmental and nongovernmental agencies dispersed to the sampling locations in teams of two or more to collect samples. All samples were collected on the same day, and the timing of the sampling coincided with an outgoing tide, near low tide, to reduce the potential of collecting water from Cook Inlet.

Typically, the individual collecting the sample waded into the water until the water level was around two feet deep, and the sample was collected while facing upstream. If the individual collected the sample using a boat, the samples were collected from the bow while the boat faced upstream. The bottles were placed approximately one foot below the surface to collect the water samples and then preserved for transportation to the laboratory. Beginning in April 2002, two duplicate samples were collected for quality control. These procedures follow the protocols established in a Quality Assurance Project Plan that was approved by the ADEC in 2001 (KWF, 2001).

After collection, most samples were sent to a private laboratory that is Alaska State certified for drinking water and waste water parameters in

UNIGIS Master of Science in GIS

Anchorage, Alaska in order to be processed and analyzed. Here, all samples were prepared and analyzed according to USEPA or equivalent methods outlined in the Methods for the Determination of Metals in Environmental Samples, EPA/600/R-91/111 (USEPA, n.d.). All analytical results provided by the laboratory underwent strict quality control and quality assurance protocols. The samples taken to analyze for Fecal Coliform bacteria were processed and analyzed at the Soldotna waste water treatment plant during the spring sampling events and at a private testing facility in Soldotna during the summer sampling events. In both cases the fecal coliform bacteria analysis was done by membrane filter.

In order to maintain the most consistency possible during the entire project, a basic suite of parameters was collected at all the sampling stations during the entire project. This basic suite of parameters consists of: Arsenic, Cadmium, Chromium, Copper, Lead, Zinc, Calcium, Iron, Magnesium, Nitrate, Phosphorus, Fecal Coliform Bacteria, pH, Specific conductance, Total Suspended Solids, Turbidity and Temperature.

Due to budgetary

constraints, parameters such as Diesel Range Organics, Gasoline Range Organics, Residual Range Organics and Total BTEX were only collected during the sampling efforts of Spring 2001-2005 and Summer 2000-2007.

To meet the first specific objective of this study, all the results from every sampling event were compared to standards set by the State of Alaska and the Federal government.

Additionally, a sub-set of parameters were used to assess the water quality in the river. Although all the above presented parameters are important to determine the water quality of the river, there are some parameters that provide more comprehensive information because they allow scientist to calculate a water quality index which in turn assigns one numeric value that represents the quality.

UNIGIS Master of Science in GIS

3.3 Data Digitalization and Geographical Analysis

Figure 4: Flow chart of methodology for geographical analysis.

Figure 4 represents the pathway followed to combine all the information and make it available in a GIS setting in order to perform Geospatial analysis. Using the results obtained from the laboratory, the field measurements, statistical analysis, GIS information regarding land use, road system, and hydrography, dynamic segmentation and inverse distance weighting (IDW) analysis were conducted to better assess the water quality in the entire Kenai River. The layers for political boundaries and road system were downloaded from the database of the Kenai Peninsula Borough (KPB). The KPB also provided a Geodatabase with lot lines, lot parcels, polygons, tax return information, address, etc. that allowed for the creation of a land use map. The hydrographic information was downloaded from the National Hydrography Dataset (NHD). The exact sampling locations were provided by the KWF. Because this project was done in the State of Alaska, the Projected

Coordinate

System

used

the

NAD_1983_StatePlane_Alaska_4_FIPS_5004_Feet.

UNIGIS Master of Science in GIS

For the statistical analysis, the programs used were Minitab 16 and R. A principal component analysis (PCA) was performed using the entire set of data to allow finding patterns that could help summarize and simplify the data set. Also, a hierarchical cluster analysis by means of the average linkage method, using arithmetic averages between the distance values of pares of cases as a measure of similarity, was performed with all parameters in order to identify if there was any redundancy between parameters and to understand how parameters behave between each other (Bartier & Keller, 1996; Bengraı̈ ne & Marhaba, 2003; Ding et al., 2015; Fan, Cui, Zhao, Zhang & Zhang 2010; Xu et al., 2012). Furthermore, a water quality index (WQI) was calculated using Srivastava’s and Kumar’s 2013 calculation of the WQI with missing parameters using the following parameters: Nitrate, pH, Temperature, Turbidity, Total Suspended Solids (TSS), Fecal Coliform Bacteria and Phosphorus. It was necessary to use this particular calculation of a WQI because in some instances the results were reported to be below the MDL or MRL. Once a WQI was available for every station and date, all the indices for each location were averaged. These values were then used to perform geographic analysis producing results of water quality in the entire system as opposed to specific stations in the river. The Water Quality Ranges used for this study are a modification of the ranges presented by Srivastava’s and Kumar’s in their 2013 study (see Table 5).

Index Ranges

Water Quality

Level of Concern

< 50

Moderate

High

50-60

Good

Medium

60-70

Excellent

Low

Table 5: Water quality ranges that have been modified in order to better adjust for the data in this project.

For the geographic analysis, the first step was to develop two excel spreadsheets that had the following information: a unique identifier for each event, seasonality (spring or summer), site locations, number of times a water quality standard was exceeded in each location, the land use adjacent to the river and the WQI. Originally, based on the information UNIGIS Master of Science in GIS

provided by the KPB, the land use was assigned one of five different categories: Residential, Commercial, Institutional, Gravel Pit and Vacant. Itâ&#x20AC;&#x2122;s widely accepted that although the entire watershed influences the water quality, quantity and geomorphology of a river or stream, the riparian zone is of critical importance due to the direct proximity to the water. In consequence, for this study, a radius of 0.5 mile from the sampling point was used to delimit the land use area (Bonansea, Ledesma, & Rodriguez, 2016; Yan et al., 2015; Yang & Jin, 2010; Yu, Xu, Wu, & Zuo, 2016). Due to the nature of the data being analyzed, it became clear that a simplification of the land use categories would allow for a better representation of the impact on the water quality of the river. This led to a simplification of the land use categories, from Residential, Commercial, Institutional, Gravel Pit and Vacant to developed or non-developed. The developed category encompasses any land or parcel that has had some type of human activity or disturbance. On the other hand, the undeveloped category contains all the parcels and land that are untouched, protected, pristine or have little to no human traffic (see Figures 5-14). Based on the number of times water quality standards were exceeded on each sampling station and the land use adjacent to the river, each sampling station was assigned a qualitative value. When comparing the Kenai River with other rivers in the United States, its condition and water quality are considered pristine (ADEC, 2015). Due to this, values from excellent to moderate were assigned without considering a â&#x20AC;&#x153;badâ&#x20AC;? value for the quality of the river. It is important to remember that the purpose of this analysis is to obtain values of water quality for the entire river utilizing the sampling stations as the start and end of each segment of the river.

UNIGIS Master of Science in GIS

Figure 5: Six categories of land use available along the estuary and lower sections of the Kenai River.

UNIGIS Master of Science in GIS

Figure 6: Developed and undeveloped land in the estuary and lower sections of the Kenai River.

UNIGIS Master of Science in GIS

Figure 7: Six categories of land use available along the middle section of the Kenai River.

UNIGIS Master of Science in GIS

Figure 8: Developed and undeveloped land in the middle section of the Kenai River.

UNIGIS Master of Science in GIS

Figure 9: Six categories of land use available along the middle section of the Kenai River.

UNIGIS Master of Science in GIS

Figure 10: Developed and undeveloped land in the middle section of the Kenai River.

UNIGIS Master of Science in GIS

Figure 11: Six categories of land use available along the upper section of the Kenai River.

UNIGIS Master of Science in GIS

Figure 12: Developed and undeveloped land in the upper section of the Kenai River.

UNIGIS Master of Science in GIS

Figure 13: Six categories of land use available along the upper section of the Kenai River.

UNIGIS Master of Science in GIS

Figure 14: Developed and undeveloped land in the upper section of the Kenai River.

UNIGIS Master of Science in GIS

CHAPTER 4: RESULTS AND DISCUSSION 4.1 Results 4.1.1 Parameters 4.1.1.1 Arsenic

None of the samples exceeded the Alaska or federal standard for freshwater aquatic life at any sampling location in April or July. The highest level detected in the mainstem was 46.5 Âľg/L at Mile 1.5 in May 2007, and arsenic was not detected on many occasions below the method detection limit (MDL) of 0.25 Âľg/L. In the mainstem higher arsenic levels occurred in the spring samples, while for the tributaries levels where higher during the summer and there were more detected levels between the years 2007-2014 than any of the previous years. Of the tributaries, Soldotna Creek had the highest median level, while No Name Creek had the fewest incidences of arsenic detection of all the tributaries. Furthermore, using GIS tools such as dynamic segmentation and IDW, along with the means and medians for the values of arsenic during spring and summer sampling events, maps were developed to assess the concentration of this parameter along the Kenai River. From this analysis, it can be inferred that the highest levels of Arsenic for spring can be found in the estuary portion of the Kenai River. During the summer the highest levels of arsenic were found in the middle section of the river (see Figure 15). Also, it is important to note that the highest levels of arsenic, regardless of seasonality, were always found in sections of the river that presented development.

UNIGIS Master of Science in GIS

Figure 15: Levels of arsenic in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.2 Cadmium

In the mainstem, concentrations of cadmium ranged from the lowest levels that were below the MDL of 0.062 µg/L that occurred in multiple locations to a high level of 8 µg/L at Mile 21 in July 2000. The highest level of cadmium ever detected in the tributaries was 63 µg/L at Soldotna Creek in July 2002, and the lowest was below the MDL of 0.062 µg/L. In spring, cadmium was only reported four times, in contrast to summer samples in which it was reported ten times. Between 2001 and 2005, the method reporting limits (MRL) were mainly higher than the standard, so it is unknown whether these samples exceeded the standard. No geographic analysis was performed with this parameter due to the fact that there were so many results under the MDL or MRL.

4.1.1.3 Chromium

None of the medians for the mainstem or the tributaries exceeded the freshwater aquatic life standard for hexavalent or trivalent chromium, although single samples did detect concentrations above the hexavalent chromium standard at Mile 6.5, Mile 10.1, Mile 21, and Beaver Creek. The standard for trivalent chromium was not exceeded at any site along the mainstem or tributaries during any sampling

event. The highest

concentration of chromium in the mainstem was 25 µg/L at Mile 21 in July 2000, and the lowest levels occurred below the MDL of 0.36 µg/L. In the tributaries, higher levels of chromium were reported in July compared to April, while a seasonal trend was difficult to distinguish in the mainstem.

Additionally, using the means and medians for the values of chromium, and with the help of tools like dynamic segmentation and IDW, two maps were developed to assess the concentration of this parameter along the Kenai River. This analysis shows that the highest levels of chromium for spring can be found in the estuary portion of the Kenai River. Meanwhile, during the summer the highest levels of arsenic were found in the lower section of the river (see Figure 16). It is important to note that the highest levels of chromium, were found in sections of the river that presented development. UNIGIS Master of Science in GIS

Figure 16: Levels of chromium in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.4 Copper

The highest concentration of copper in the mainstem was reported at 443 µg/L at Mile 21 in July 2000, and the lowest levels were below the MDL of 0.12 µg/L. Higher levels occurred in the estuary, while the lowest concentrations were detected at Mile 40, Mile 43, and Mile 50. No exceedances were recorded for the mainstem during the sampling events in spring. Concentrations in the tributaries ranged from below the MDL of 0.12 µg/L that occurred in many locations during this project to 13 µg/L in the Killey River in April 2002. Median concentrations and the number of exceedances were generally higher in July than in April in the tributaries and the mainstem.

Using GIS tools such as dynamic segmentation and IDW and the means and medians for the values of copper during spring and summer sampling events, maps were developed to assess the concentration of this parameter along the Kenai River. These maps show that the highest levels of copper during the spring sampling events are found in the estuary portion of the Kenai River. During the summer sampling events the highest levels of copper were found at mile 21 between the lower and middle section of the river (see Figure 17). Also, it is important to note that the levels of copper are higher during the summer sampling events. Also, during spring and summer sampling events the highest levels of copper are found in developed areas.

UNIGIS Master of Science in GIS

Figure 17: Levels of copper in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.5 Lead

The highest concentration of lead in the mainstem was 468 µg/L and was sampled at Mile 31 during July 2005, and the lowest levels occurred below the MDL of 0.030 µg/L at many locations in April and July of the sampling years. The concentration of lead in the tributaries ranged from a high of 3.92 µg/L at Beaver Creek in July 2006, and the lowest levels were below the MDL of 0.030 µg/L at many locations in April and July sampling events. In most cases the MDLs were higher than the standard, so it is unknown whether exceedances occurred in many of these samples.

Dynamic segmentation, IDW and the means and medians for the values of lead during spring and summer sampling events were used to develop maps to assess the concentration of this parameter along the Kenai River. These maps show that the highest levels of lead during the spring sampling events are found in the estuary and lower portion of the Kenai River. During the summer sampling events the highest levels of lead were found in the middle section of the river (see Figure 18). Also, it is important to note that the levels of lead are higher during the summer sampling events. The highest levels of lead in the river were found in areas with high development.

UNIGIS Master of Science in GIS

Figure 18: Levels of lead in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.6 Zinc

The highest level recorded in the mainstem was 2,900 Âľg/L in July 2000 at Mile 21, and the lowest level occurred at Mile 31 in July 2006 with a concentration of 0.521 Âľg/L. In April, exceedances occurred at every sampling station except Mile 40 and Mile 50. The Kenai River estuary had higher median concentrations than the rest of the river in spring, during the summer events this pattern was not clear. In April and July 2003, the MRL of 50 Âľg/L was higher than most of the standards, so incidences of exceedance are unknown in many of these samples.

Furthermore, dynamic segmentation, IDW and the means and medians for the values of zinc during spring and summer sampling events were used to develop maps to assess the concentration of this parameter along the Kenai River. These results show that the highest levels of zinc for spring can be found in the estuary and lower portion of the Kenai River. During the summer the highest levels of zinc were found in the middle section of the river (see Figure 19). Also, it is important to note that the highest levels of zinc, regardless of seasonality, were always found in sections of the river that presented development.

UNIGIS Master of Science in GIS

Figure 19: Levels of zinc in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.7 Calcium

The highest value in the mainstem occurred during April 2002 at Mile 1.5 with a concentration of 174 mg/L, and the lowest value was 0.447 mg/L at Mile 43 in July 2010. In the mainstem, calcium levels were relatively higher in spring than in summer. In the tributaries, the lowest concentration was 3 mg/L sampled from the No Name Creek during May 2013, and the highest concentration occurred in the Moose River at 25.7 mg/L during July 2004.

In addition, with the help of GIS tools such as dynamic segmentation and IDW, along with the means and medians for the values of calcium during spring and summer sampling events, maps were developed to assess the concentration of this parameter along the Kenai River. From this analysis, it can be inferred that the highest levels of calcium for spring can be found in the estuary portion of the Kenai River. During the summer the highest levels of calcium were found in the middle section of the river (see Figure 20).

UNIGIS Master of Science in GIS

Figure 20: Levels of calcium in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.8 Iron

In the mainstem, the highest concentration of iron was 128 mg/L at Mile 6.5 during April 2006, and 0.03 mg/L was the lowest concentration that occurred at Mile 70 during spring 2013. There was a general upward trend in iron concentration from Kenai Lake to the estuary, especially in July. The concentrations in the tributaries ranged from a high of 20.5 mg/L in Beaver Creek in spring 2006 to Russian River, which had the lowest concentration at below the MDL of 0.0027 mg/L. In both the tributaries and the mainstem, iron levels were higher in April than July.

Dynamic segmentation, IDW and the arithmetic means and medians for the values of iron during spring and summer sampling events were used to develop maps to assess the concentration of this parameter along the Kenai River. These maps show that the highest levels of iron during the spring sampling events are found in the estuary and lower portion of the Kenai River. During the summer sampling events the highest levels of iron were found in the middle section of the river (see Figure 21). The highest levels of iron in the river were found in areas with high development.

UNIGIS Master of Science in GIS

Figure 21: Levels of iron in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.9 Magnesium

In the mainstem, the highest concentration of magnesium was 582 mg/L at Mile 1.5 during April 2011, while the lowest level was 0.729 mg/L at Mile 50 in July 2001. In both April and July, Mile 1.5 had the highest median followed by Mile 6.5. There was a general upward trend in magnesium levels from Kenai Lake to the estuary in both April and July. In the mainstem, concentrations were higher in April than in July. In the tributaries, Magnesium ranged from the high of 21.4 mg/L in April 2010 at No Name Creek to 0.746 mg/L in April 2004 at the Killey River. In the tributaries, concentrations were higher in July than in April.

Using GIS tools like dynamic segmentation, IDW plus the means and medians for the values of magnesium during spring and summer sampling events were used to develop maps to assess the concentration of this parameter along the Kenai River. These maps show that the highest levels of magnesium during the spring sampling events are found in the estuary portion of the Kenai River. During the summer sampling events the highest levels of magnesium were also found in the estuary section of the river (see Figure 22). The highest levels of magnesium in the river were found in areas with high development.

UNIGIS Master of Science in GIS

Figure 22: Levels of magnesium in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.10 Nitrate

In the mainstem, the highest concentration of nitrate was 0.706 mg/L at Mile 70 during April 2005, and in many instances, nitrate levels were below the MDL of 0.015 mg/L. In April, the mainstem had a general downward trend from Kenai Lake to the estuary, excluding the higher medians at Mile 70 and Mile 1.5. Overall, nitrate levels were slightly lower in April than July for the mainstem. The tributaries generally had lower levels of nitrate than the mainstem, with the exception of Russian River. The highest level of nitrate in the tributaries occurred at Russian River with a concentration of 1.11 mg/L during April 2005, and the lowest levels occurred at many locations below the MDL of 0.015 mg/L. Tributaries had too many values below the MDL and MRL to establish a clear seasonal trend. Using dynamic segmentation, IDW and the arithmetic means and medians for the values of nitrate during spring and summer sampling events maps were developed to assess the concentration of this parameter along the Kenai River. These maps show that the highest levels of nitrate during the spring sampling events are found in the upper portion of the Kenai River. During the summer sampling events the highest levels of nitrate were also found in the upper section of the river (see Figure 23).

UNIGIS Master of Science in GIS

Figure 23: Levels of nitrate in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.11 Phosphorus

The highest level of phosphorus ever recorded occurred at Mile 1.5 in April 2005 with a concentration of 4 mg/L, and the lowest levels occurred on many occasions below the MDL of 0.0020 mg/L in April and July. The estuary had significantly higher medians and variance than the rest of the mainstem in July and April. The lowest concentrations of phosphorus occurred between Mile 40 and Mile 82. In the mainstem, there was a general upward trend in phosphorus concentration from Kenai Lake to the estuary. In the tributaries, phosphorus ranged from 64 mg/L in Slikok Creek during May 2013 to below the MDL of 0.0020 mg/L at many locations in April and July.

Dynamic segmentation, IDW and the means and medians for the values of phosphorus during spring and summer sampling events were used to develop maps to assess the concentration of this parameter along the Kenai River. These maps show that the highest levels of phosphorus during the spring sampling events are found at Slikok Creek. During the summer sampling events the highest levels of phosphorus were found in the estuary section of the river (see Figure 24).

UNIGIS Master of Science in GIS

Figure 24: Levels of phosphorus in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.12 Diesel Range Organics

In the mainstem, DRO concentrations ranged from a high level of 0.29 mg/L that occurred at Mile 1.5 and Mile 31 during April 2001 to the lowest levels that occurred at multiple locations at concentrations below the MDL of 0.0060 mg/L. In April, DRO was reported only in 2001 at Mile 1.5, Mile 10.1, and Mile 31, and the reported levels were only slightly higher than the MRLs. The concentrations observed in the tributaries were higher than those detected in the mainstem. No geographic analysis was performed with this parameter due to the fact that there were so many results under the MDL and/or MRL.

4.1.1.13 Gasoline Range Organics

The highest level of gasoline range organics was 38.3 Âľg/L at Mile 1.5 in July 2002, and many samples were below the MDL of 3.0 Âľg/L. Gasoline range organics have only been detected in July at Mile 1.5, Mile 6.5, and Mile 10.1 in the mainstem and were never detected in the tributaries. Although no concentrations were reported in July 2005, this may be due to the exceptionally high MRL of 100 Âľg/L. No geographic analysis was performed with this parameter due to the fact that there were so many results under the MDL and/or MRL.

4.1.1.14 Residual Range Organics

The only instance of detection in the mainstem occurred at Mile 23 with a concentration of 4.75 during July 2000, and the lowest levels were reported in numerous locations below the MDL of 0.032 mg/L in April and July. The highest concentration of RRO ever detected was 5.67 mg/L in Soldotna Creek during July of 2000. No geographic analysis was performed with this parameter due to the fact that there were so many results under the MDL and/or MRL.

UNIGIS Master of Science in GIS

4.1.1.15 Total BTEX

The highest level of BTEX in the mainstem occurred at Mile 1.5 at a level of 15.2 Âľg/L in July 2002, and the lowest levels were reported as non-detect because they were below the MDLs and MRLs. No BTEX was reported in any of the April samples, but high levels were detected in the lower river during July. In the tributaries, concentrations ranged from 6.65 Âľg/L in the Moose River in July 2002 to non-detects reported in many instances. None of the tributaries had reported levels above the standard. No geographic analysis was performed with this parameter due to the fact that there were so many results under the MDL and/or MRL.

4.1.1.16 Fecal Coliform Bacteria

The highest level recorded was 2,980 CFU/100m at Mile 6.5 during July 2002, however this sample may be unreliable because the duplicate sample was below the MDL of 1 CFU/100ml. The highest median in the mainstem occurred at Mile 6.5 in April and at Mile 1.5 in July. The concentration of fecal coliform was generally higher in July than in April in the mainstem and the tributaries. In the tributaries, the concentration of fecal coliform ranged from a low of 0 CFU/100ml at multiple sites to a high of 520 CFU/100ml in Soldotna Creek during April 2001.

Dynamic segmentation, IDW and the means and medians for the values of fecal coliform bacteria during spring and summer sampling events were used to develop maps to assess the concentration of this parameter along the Kenai River. These maps show that the highest levels of fecal coliform bacteria during the spring sampling events are found in the lower portion of the Kenai River. During the summer sampling events the highest levels of fecal coliform bacteria were also found in the lower section of the river (see Figure 25). Also, during spring and summer sampling events the highest levels of fecal coliform bacteria are found in developed areas.

UNIGIS Master of Science in GIS

Figure 25: Levels of fecal coliform bacteria in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.17 pH

In the mainstem, the pH ranged from a high of 8.52 at Mile 50 in July 2002 to a low of 6.16 at Mile 10.1 in July of 2010. All the medians along the mainstem were between 7 and 8 for April and July, so they did not exceed the upper and lower limits. The pH was generally lower in April in comparison to July in both the mainstem and the tributaries.

Dynamic segmentation, IDW and the means and medians for the values of pH during spring and summer sampling events were used to develop maps to assess the concentration of this parameter along the Kenai River. These maps show that the during the spring sampling events multiple pockets of high pH levels are found along the Kenai River. During the summer sampling events the highest levels of pH were found in the middle section of the river (see Figure 26).

UNIGIS Master of Science in GIS

Figure 26: Levels of pH in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.18 Specific Conductance

The highest level of specific conductance in the mainstem was 29,398 µS/cm, which was sampled at Mile 1.5 in April 2011, and the lowest level was 37.7 µS/cm at Mile 50 in April 2004. Specific conductance was higher in April than in July throughout the mainstem, except at Mile 50 where specific conductance was higher during the summer. In the tributaries, specific conductance ranged from a high of 1,088 µS/cm in No Name Creek during April 2010 to the lowest level of 18 µS/cm in the Killey River during July 2004.

Using dynamic segmentation, IDW and the arithmetic means and medians for the values of specific conductance during spring and summer sampling events maps were developed to assess the concentration of this parameter along the Kenai River. These maps show that the highest levels of specific conductance during the spring sampling events are found in the estuary portion of the Kenai River. During the summer sampling events the highest levels of specific conductance were also found in the estuary section of the river (see Figure 27).

UNIGIS Master of Science in GIS

Figure 27: Levels of specific conductance in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.19 Total Suspended Solids

The concentration of 2,073 mg/L at Mile 6.5 during April 2006 was the highest level of total suspended solids recorded in the mainstem, and the lowest concentration was less than the MDL of 0.48 mg/L during spring and summer sampling events. The estuary had much higher medians than the rest of the mainstem in both July and April In the tributaries, the concentration of total suspended solids ranged from a high of 748 mg/L in Beaver Creek in April 2009 to a low of less than the MDL of 0.48 mg/L.

Using dynamic segmentation, IDW and the arithmetic means and medians for the values of total suspended solids during spring and summer sampling events maps were developed to assess the concentration of this parameter along the Kenai River. The resulting maps show that the highest levels of total suspended solids during the spring sampling events are found in the estuary portion of the Kenai River. During the summer sampling events the highest levels of total suspended solids were also found in the estuary section of the river (see Figure 28).

UNIGIS Master of Science in GIS

Figure 28: Levels of total suspended solids in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.20 Turbidity

The highest level of turbidity occurred at Mile 6.5 in April 2006 and was detected at 3,200 NTU, and the lowest level in the mainstem was 1 NTU at Mile 70 in April 2009. The highest medians and degree of variance for turbidity were in the estuary for both April and July, and the lowest median occurred at Mile 50. The April medians in the estuary and Upper River were higher than the July medians, but the rest of the mainstem had higher medians in July. In the tributaries, the highest level of turbidity was 336 NTU at Beaver Creek in April 2010, and the lowest level was at Russian River with the level of 0.02 NTU from July 2006.

In addition, with the help of GIS tools such as dynamic segmentation and IDW, along with the arithmetic means and medians for the values of turbidity during spring and summer sampling events, maps were developed to assess the concentration of this parameter along the Kenai River. From this analysis, it can be inferred that the highest levels of turbidity for spring can be found in the estuary and lower portion of the Kenai River. During the summer the highest levels of turbidity were also found in the estuary and lower portion of the river (see Figure 29).

UNIGIS Master of Science in GIS

Figure 29: Levels of turbidity in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

4.1.1.21 Water Temperature

The highest temperature recorded on the mainstem occurred at Mile 1.5 in July 2014 with a value of 17.02°C, and the lowest recorded temperature along the mainstem was –0.15°C at Mile 12.5 in April 2002. In April, two of the temperatures exceeded the standard for rearing areas and migration routes. There was a general upward trend from Mile 82 to Mile 1.5. It is important to note that during the sampling event of April 2014, twelve out of thirteen sites exceeded the standard for egg and fry incubation and spawning areas. In the tributaries, the highest temperature was 19°C in the Moose River during July 2003, and the lowest temperature was -0.21°C in the Killey River during April 2002. In April, No Name Creek, Soldotna Creek and Russian River exceeded the standards for rearing areas and migration routes. Also, Slikok Creek, Funny River and Moose River exceeded the standard for egg and fry incubation and spawning areas. All tributaries had medians below 5°C, with the highest at Russian River and the lowest at Funny River. During July, the medians of Beaver Creek, Soldotna Creek, and Moose River exceeded the standards for egg and fry incubation and spawning areas, and the median at Moose River also exceeded the standards for rearing areas and migration routes. The median water temperatures were higher in July than in April for all sampling locations in both the mainstem and the tributaries. For this analysis, there were no values available for temperature during the sampling event of April 2012.

Furthermore, using dynamic segmentation, IDW and the arithmetic means and medians for the values of temperature during spring and summer sampling events maps were developed to assess the levels of this parameter along the Kenai River. These maps show that the highest levels of temperature during the spring sampling events are found in the estuary, lower and middle portion of the Kenai River. During the summer sampling events the highest levels of temperature were found in the middle section of the river, followed by the lower section and the estuary (see Figure 30).

UNIGIS Master of Science in GIS

Figure 30: Levels of water temperature in the Kenai River during the spring and summer sampling events.

UNIGIS Master of Science in GIS

Furthermore, this study shows that the Kenai River is overall a very healthy system and that itâ&#x20AC;&#x2122;s water quality is driven by two main components, seasonality and land use. The estuary and lower Kenai River present a high-density development with land being used for multiple purposes. On the other hand, the middle and upper Kenai River present far less development with very small pockets of development.

After analyzing all of the data and corresponding trends there are six areas of concern for aquatic life in the Kenai River Watershed:

1. Tributaries and their confluence with the Kenai River need to be monitored because they are a key factor for the water quality in the river. 2. The Moose River requires special attention, during the summer, at its confluence with the Kenai River the following parameters were high compared to other sections of the river: pH, conductance and temperature. 3. Water temperatures in the Moose River typically exceeded two of Alaskaâ&#x20AC;&#x2122;s standards, and two of the standards were regularly exceeded at Beaver Creek, and Soldotna Creek during the summer. 4. Road crossings need to be monitored more carefully especially for parameters like Zinc and Cooper. 5. Iron median levels regularly exceeded the standard in the estuary and in No Name Creek, Beaver Creek, Slikok Creek, Soldotna Creek, Funny River, Moose River, and the Killey River, especially in the spring. 6. Zinc levels are on the rise since 2010. Also, Name Creek and Slikok Creek were often higher levels than the state and federal standards.

4.1.2 Water Quality Performing a hierarchical cluster analysis of all the water quality parameters measured in the field and in the laboratory, allowed to distinguish which parameters are strongly related to each other and how they can be grouped in order to better explain the water quality of the river. The resulting dendrogram allows to visualize how the different parameters group themselves and how similar they are between each other. Parameters with

UNIGIS Master of Science in GIS

high levels of similarity means that they interact in the same way with other parameters (see Figure 31). Furthermore, with the WQI values for each location and sample date a PCA confirmed that seasonality plays a key role in the water quality of the Kenai River. Figure 32 shows how data collected during spring and summer occupy different areas and are mainly explained by the first component. A second cluster analysis, which includes all five values of land use and the WQI value, shows that the data presents two major clusters. On one side, there is the cluster of parcels under the category vacant, and on the other side all the remaining categories for land use together. This means that the fact that land is being used is reason enough to drive the water quality of the river, which allowed to group the land use in two categories, developed and undeveloped (see Figure 33).

Water Quality Parameters

Similarity

61.72

74.48

87.24

100.00

ic te en itra s N Ar

l r y c s pH tur e ium i um nce ium Ir on idit T SS pe Zin eca ru ead p o c s a L a l b F t m h e r Co ur Ca gn duc hro sp pe T o a m C M C on Ph Te

Parameters

Figure 31: Cluster analysis of all the measured parameters.

UNIGIS Master of Science in GIS

Score value of Water Quality Indices for the Kenai River 4

Group 1 2

Second component

2 1 0 -1 -2 -3 -4 -7.5

-5.0

-2.5 0.0 First component

2.5

5.0

Figure 32: PCA performed using the entire set of data. Group 1 represents all the data collected in Spring and Group 2 represents data collected in summer.

Sampling Sites 78.92

Similarity

85.95

92.97

100.00

e m Na

k 31 er ek er 23 18 ek .5 .5 ek .1 50 70 er 43 82 er ek .5 21 40 v v v v 2 e 1 6 r e 10 Ri re Ri Ri Ri r e 1 Cr C y k C se y uC n r a n e a e i n ll a n ko o ot av ss Ki ne Fu Sl i Mo ld Be Ru Ju o S

ee Cr

Sampling Sites

Figure 33: Cluster analysis of land use values and WQI.

UNIGIS Master of Science in GIS

Based on the statistical analysis and in order to simplify and better assess the influence of land use to the water quality of the river; the land use categories were reduced from six (vacant, residential, institution, Farm/Agr, Commercial

and

Gravel

Pit)

two

categories

(developed

and

undeveloped).

The overall water quality in the Kenai River ranges from moderate to excellent. Calculated WQI in spring and summer are presented in Figure 34 which was developed using IDW. In the spring, there is a clear difference of water quality between the lower and the upper Kenai River, the latter one presenting excellent water quality while the lower river starts presenting some decay in the water quality from river mile 44 (Killey River) with good water quality to moderate water quality in the mouth of the river (No Name Creek). During the summer the water quality of the Kenai River is more uniform along the entire system. There are however a few pockets of the river with excellent water quality (river miles 82, 70 and 50) which shows that the upper Kenai River still presents better water quality than the lower Kenai River.

UNIGIS Master of Science in GIS

Figure 34: Overall water quality in the Kenai River during the spring and summer.

UNIGIS Master of Science in GIS

4.2 Discussion Dissolved metals were generally reported in small concentrations with few exceedances of state or federal standards, many reported below laboratory detection methods. Arsenic levels were generally higher in the tributaries, but no samples exceeded the aquatic life standard at any time. Concentrations of arsenic are generally lower in surface streams than in groundwater, which is typically the source of drinking water (Glass & Frenzel, 2001). The USEPA set the criterion for arsenic in drinking water at 10 Âľg/L because arsenic has been linked to cancer, skin damage, and circulatory problems (USEPA, 2003). Although the levels of arsenic reported in this study do not exceed the national criterion for the health of an

aquatic

community

freshwater,

groundwater

may

contain

concentrations that are hazardous to human health, and all sources of drinking water should be tested for arsenic.

Laboratory methods for detecting cadmium did not measure concentrations to low enough levels to determine if there were exceedances for many sampling events. However, even when detection procedures improved, very few incidences of detection occurred. Since so few cadmium samples were detected, trends were difficult to determine and should be interpreted with caution.

Trivalent chromium was never reported at levels above the standard. It is possible that hexavalent chromium may have exceeded the standard on four occasions, but the laboratory analysis did not distinguish between the two isomers of chromium.

In the early sampling years, the number of exceedances of the copper criteria are unknown due to laboratory detection limits, some exceedances occurred when methods improved. More detected concentrations occurred in sampling events after 2005, but this may be partially due to the large fluctuation in MDLs and MRLs. From July 2000 to July 2003, the MDLs or MRLs were higher than the standard, so it is unknown if many of these UNIGIS Master of Science in GIS

samples exceeded the standard.

Lead had more exceedances at the Kenai River mainstem than in the tributaries. Of the dissolved metals, zinc had the most exceedances in both the Kenai river mainstem and the tributaries as well as during spring and summer events.

The total metals, calcium, iron, and magnesium, display a spatial trend, increasing from Kenai Lake outlet to the Kenai River estuary, and they all had higher concentrations in the spring. There is no Alaska or federal water quality standard that applies to calcium and magnesium concentrations for freshwater aquatic life, so neither experienced any exceedances. Calcium levels were highest in the estuary, most likely due to tidal influence, and concentrations were also high in the Upper Kenai River. Magnesium levels dropped at the Skilak Lake outlet and then rose throughout the remainder of the river heading downstream. The tributaries generally had higher concentrations of magnesium and iron than the mainstem (excluding the estuary). Iron levels usually exceeded the standard in all tributaries except for the Russian River, the Killey River, and Juneau Creek. Iron levels generally increased downriver, and exceedances of the iron standard on the mainstem were typical from Mile 10.1 through the estuary.

The two nutrients included in this study, nitrate and phosphorus, displayed very different spatial trends. From Kenai Lake to the estuary, nitrate levels decreased. Beaver Creek had relatively low levels of nitrate, while Russian River had a very high concentration of nitrate. In contrast, Russian River had very low levels of phosphorus, and there was a general upward trend from Kenai Lake to the estuary with high levels of phosphorus in the Killey River and Beaver Creek.

Sampling for diesel range organics, gasoline range organics, and residual range organics was useful in narrowing down the main source of hydrocarbons present: benzene, toluene, ethylbenzene, and xylene (BTEX). Very few detections occurred for any of the range organics. The laboratory

UNIGIS Master of Science in GIS

100

detection limits were too high during several sampling events to detect exceedances of gasoline range organics and residual range organics. All of the detected gasoline range organics occurred downstream of Mile 10.1 during the summer. BTEX concentrations showed an upward trend from the Middle Kenai River to the estuary, with no median exceeding the standard. All of the aromatic hydrocarbons of BTEX are typically associated with gasoline (Oasis Environmental, Inc., 2004). Litchfield and Kyle (1992) suggest that motorized boats could be the primary contamination source since peak times of outboard motor use coincided with peak concentrations of BTEX. In 2003, the KWF determined that the majority of hydrocarbon contamination resulted from unburned refined gasoline product with outboard motors as the most likely source. No detections of BTEX occurred in the spring, indicating outboard motors as the primary source of contamination.

Fecal coliform bacteria, pH, and specific conductance all had few to zero exceedances. All of the median levels of fecal coliform bacteria were below 200 CFU/100m, although the highest levels occurred at Mile 6.5, Slikok Creek, Beaver Creek, and No Name Creek in the summer. Most of the exceedances of the pH standards occurred below the lower limit in the spring, but overall, there were very few samples outside of the acceptable range. In average, the pH range was lower during the spring months than in the summer months. This could be associated with the difference between the amount of water flowing through the system between April and July as well as the source of the water, glacial or nonglacial water (Litchfield & Kyle, 1992). Also, the pH values for the upper Kenai River remain fairly constant during the spring and the summer, in contrast with the fluctuating pH values in the lower section of the Kenai River. This is important to mention because the lower Kenai River is more developed than the upper Kenai River, and during the summer monthsâ&#x20AC;&#x2122; people are actively using the river more in this section (see Figure 26). As expected, specific conductance was higher in the tidally influenced portion of the River (No Name Creek to Mile 6.5) for both spring and summer. The range for conductance values was larger in the spring than in

UNIGIS Master of Science in GIS

101

the summer. This could be due to miss-coordination between the sampling event and the tidal cycle. During the spring, the conductance above Mile 6.5 remains fairly constant, while in the summer, almost the same pattern is visible but there is a pocket of higher conductance in the confluence of the Moose River (see Figure 27).

The remaining characteristics of total suspended solids, turbidity, and water temperature had high levels in several areas. The concentration of total suspended solids was highest in the estuary in the spring and the summer, along with high levels in the Killey River in the summer. Russian River had the lowest median for total suspended solids and turbidity. In the spring, turbidity values were below 10 NTU for all stations above Mile 10.1. In the summer, turbidity displayed a general increasing trend moving downriver, with the highest levels at the estuary and the Killey River (see Figure 29). Overall, turbidity in the main stem during the spring months is lower than the turbidity measured in the main stem during the summer months, which can be attributed to the concentration of total suspended solids in the system for the same period as well as the fact that there if more river use during the summer. In 2012, the KWF released a turbidity study conducted on the Kenai River suggesting that there is a direct correlation between boat traffic and turbidity in the water. Water temperature also increased from the Kenai Lake Outlet to the estuary both in spring and summer. As expected, seasonality plays an important role in the water temperature; the temperature range in the river was lower during the spring and higher during the summer (Ouyang, Nkedi-Kizza, Wu, Shinde, & Huang, 2006). There were exceedances in the mainstem and tributaries during the spring and the summer for rearing areas and migration routes as well as for egg/fry incubation and spawning areas. In the summer, medians in the estuary were very close to exceeding the upper limits for spawning areas and egg and fry incubation. The medians exceeded the upper limits for spawning areas and egg and fry incubation in Beaver Creek and Soldotna Creek, and the Moose River in the summer. Additionally, the Moose River also exceeded the standards for rearing areas and migration routes (see Figure 30).

UNIGIS Master of Science in GIS

102

The calculation of WQI showed to be a viable tool that allowed the grouping of similar parameters, thus providing one numeric water quality value per site per date without losing vital information. Although this simplification proved to be correct for this particular analysis showing expected patterns of water quality, there is the need for the development of a more specific WQI that is focused on the Kenai Peninsula and is based on the ecological, hydrological and climatological reality of the region in order to obtain better results.

According to the KPB records, there are five distinct land use categories in the Kenai Peninsula: Residential, Vacant, Commercial, Institutional and Gravel Pit. Figure 11 shows that these five categories can be summarized into two categories, Developed and Undeveloped. This categorization can be explained because of the overwhelming percentage of vacant land compared to all the other categories. The estuary, lower and middle portion of the river flow through the cities of Soldotna and Kenai and the township of Sterling and is being used for many different activities that take place in a relatively small area which creates a high-density development (see Figures 5, 6, 7, 8, 9 &10). The upper river flows mainly through the Kenai Wildlife Refuge and the Chugach National forest, with only a few pockets of development that represent the Russian River Falls campground, a few tourist lodges and the town of Cooper Landing (see Figures 11, 12, 13 & 14).

Non-point source pollution is defined by USEPA as pollution that comes from a variety of sources like land runoff, rain, drainage, etc. (USEPA, 2016). This

is one of the biggest problems that the United States of

America faces in terms of water quality, not only because assigning water chemistry values to a non-point source can be very challenging since the source of the information is not a point but rather an entire area but also because there are gaps of information that need to be explained in order to make any type assessment or take a management action (Bach, Rauch,

UNIGIS Master of Science in GIS

103

Mikkelsen, McCarthy, & Deletic, 2014; Ding et al., 2015; Glass, 1999; Grunwald & Qi, 2006; Tsihrintzis et al., 1996; Yang & Jin, 2010).

The overall water quality of the Kenai River is good to excellent during the spring season and it is mostly good during the summer, with a clear difference in water quality in the estuary area for both spring and summer (see Figure 34). During the spring, there is a clear division of the quality of the water in the Kenai River between the estuary, the lower and upper river. The same division can be seen in the land use between the different sections of the river. The lower river, that has significantly more development along the river, presents water with lower quality. The upper river has very little development along the river and presents an excellent water quality. These results are to be expected in a system like the Kenai River, more development means there is more potential for impervious surfaces and storm water runoff, which leads to contamination. On the other hand, during the summer, the only division in the water quality of the river is between the estuary and the rest of the river. The reason for this change in water quality is most likely due to the heavy use of the river during the fishing season. Many of the lodges and hotels along the River are busy with visitors from all around the world and there is a constant boat traffic. Also, itâ&#x20AC;&#x2122;s important to note that the tributaries to the Kenai River, which have larger flows during the summer months, have a very big influence on the Rivers water quality. Figure 34 shows that, especially during summer months, aside from the estuary area the other areas that require attention are all the confluences of the tributaries. This study has shown that the water quality of the Kenai River shows seasonality shifts, being lower during the summer months compared to the spring months. These highlighted areas of concern require more intensive monitoring and any necessary restoration, so that the Kenai River can continue to support fishing, recreation, tourism, and the propagation of fish and wildlife.

In addition, this study has provided further evidence that GIS technology can and should be used for water quality studies and management efforts; especially for efforts were information has not been collected or to fill certain

UNIGIS Master of Science in GIS

104

gaps of information in a reliable manner (Strager et al., 2010; Tsihrintzis et al., 1996; Xu et al., 2012; Zhou & Zhao, 2011). This tool will continue to allow managers maximizing the available data they have and in the long run it could prove to be a more budget friendly approach (Fan et al., 2015; Yan et al., 2015).

The geographic analysis performed in this study is very

valuable for a variety of reasons. First, it allows to transform the water quality information of 21 points into information pertaining the water quality in the entire Kenai River. This fact alone is incredibly valuable, otherwise there would be the need to sample the river in infinite locations which is economically and logistically impossible. Second, it allows to pinpoint sections of the river that present a potential problem or concern, making it easier for land managers to prioritize their actions and budgets in a more effective manner. Third, it provides a tool that can be enhanced and improved on future studies and analysis.

Even though the current analysis allows the identification of some key trends and sections of the river that need attention, further sampling and analysis should be done before presenting a final product to decision makers and land managers. Future studies for the Kenai River watershed should focus on running a Kriging analysis with this set of data to see if the results are smoothened, thus allowing for more clear conclusions (Yan et al., 2015). Additionally, future data collection should be planned in a way that it can be useful to continue the line of this study but also that it is suitable to run more complex modelling structures such as WASP, SFPI or even large scale watershed assessment models such as SPARROW and SWAT that would allow to better understand the role the tributaries play in the Kenai River water quality (Fan et al., 2015; Gassman et al., 2007; Pelletier et al., 2006; Schwarz et al., 2006; USEPA, 2015b; Yan et al., 2015). Currently only a few tributaries to the Kenai river are being sampled and based on this study, tributaries are playing a bigger role than expected in the water quality of this system so collection of this data is necessary. Additionally, a land use analysis should be conducted for all the tributaries in order to better understand the water quality of these tributaries and how this affects the Kenai River. These land use analyses should take into

UNIGIS Master of Science in GIS

105

consideration attributes such as road crossings, culverts, river use, population density, etc. that would help strengthen the results. Based on the results in this study, future land use studies should seek to combine sociodemographic characteristics with satellite imagery technology over time so that a comparison can be made between the change in water quality and the change in land use making it possible to further understand the correlations between them (Ding et al., 2015; Mena et.al., 2011).

UNIGIS Master of Science in GIS

106

CHAPTER 5: CONCLUSIONS Water quality data was collected from the Kenai River and its tributaries from summer 2000 to summer 2014, and sampling efforts continue past the completion of this study. Several local non-profit organizations, along with local, state, federal, and tribal governments contributed to the sampling effort by providing teams to collect samples at 13 mainstem and 9 tributary sites (Juneau Creek was sampled on four occasions). Samples were analyzed for metals, nutrients, hydrocarbons, and several other water quality parameters. The results were compared to state and federal water quality standards that apply to freshwater aquatic life. Using this information, a water quality index was calculated for each sampling site. The results show that seasonality is a major factor when analyzing the water quality of the Kenai River; with the summer presenting a lower water quality than the spring. Overall the Kenai River presents a moderate to excellent water quality in both the spring and the summer. Additionally, the proportion of development along the river was proven to lead to a lower quality of water in the river.

By meeting all its objectives, this study has demonstrated that a combination of water sampling events, GIS research/application and multivariate statistical techniques not only provide an overview of the relationship between land use and water quality, but enable the characterization of the water quality of the entire system, thus creating a valuable tool that allows land managers to prioritize their efforts and budgets in a more effective way. Additionally, this study supports the hypothesis that the water quality in the Kenai River (Alaska, United States of America) is lower in areas of high development.

The results and methodologies presented in this study can be used by land managers and natural resources managers to prioritize sections of the river that need attention or sections that have the potential of needing attention in the future. However, these results should only be part of decision making process since there are variables that have not been taken into account for UNIGIS Master of Science in GIS

107

this study like road crossings, weather, natural disasters, etc. It is advisable to rethink the need of analyzing each sample for all the water parameters presented in this study, resources may be better spent if the only parameters being analyzed are the ones utilized for the calculation of the WQI, leaving possible extra funds for more sampling locations or more often sampling events. Furthermore, it is very important to work together with the State of Alaska, USGS and KWF to upgrade the river and stream network of the State of Alaska to the same standards found in the lower 48 (upgrade from NHD to NHD plus) in order to perform more reliable and replicable modelling.

UNIGIS Master of Science in GIS

108

CHAPTER 6: REFERENCES ADEC, Alaska Department of Environmental Conservation. (2008). Alaska Water Quality Criteria Manual for Toxic and Other Deleterious Organic and Inorganic Substances. Retrieved 06/25/2015, from https://dec.alaska.gov/water/wqsar/wqs/pdfs/Alaska%20Water%20Qualit y%20Criteria%20Manual%20for%20Toxic%20and%20Other%20Deleteri ous%20Organic%20and%20Inorganic%20Substances.pdf. ADEC, Alaska Department of Environmental Conservation. (2012). Water Quality Standards, 18 AAC 70. Retrieved 06/25/2015, from http://dec.alaska.gov/commish/regulations/pdfs/18%20AAC%2070.pdf. ADEC, Alaska Department of Environmental Conservation. (2015). Alaska Nonpoint Source Water Pollution Control Strategy Fiscal Years 20162020. Retrieved 06/25/2015, from http://dec.alaska.gov/water/wqsar/Docs/AlaskaNonpointSourceWaterPoll utionControlStrategy.pdf ADEC, Alaska Department of Environmental Conservation. (n.d). The Kenai River. Retrieved 06/25/2015, from http://www.adfg.alaska.gov/staticsf/Region2/pdfpubs/kenairiver.pdf ADFG, Alaska Department of Fish and Game. (n.d). Definition of Data Collection Terms pertinent to the eLandings System. Retrieved 06/27/2015, from https://elandings.alaska.gov/documentation/ar01s09.html Agency for Toxic Substances and Disease Registry. (2001). Toluene Fact Sheet: Division of Toxicology and Environmental Medicine ToxFAQs. Retrieved 06/24/2015, from http://www.atsdr.cdc.gov/toxfaqs/tfacts56.pdf Agency for Toxic Substances and Disease Registry. (2007a). Benzene Fact Sheet: Division of Toxicology and Environmental Medicine ToxFAQs. Retrieved 06/24/2015, from http://www.atsdr.cdc.gov/toxfaqs/tfacts3.pdf Agency for Toxic Substances and Disease Registry. (2007b). Ethylbenzene Fact Sheet: Division of Toxicology and Environmental Medicine ToxFAQs. Retrieved 06/24/2015, from http://www.atsdr.cdc.gov/toxfaqs/tfacts110.pdf Agency for Toxic Substances and Disease Registry. (2007c). Xylene Fact Sheet: Division of Toxicology and Environmental Medicine ToxFAQs. Retrieved 06/24/2015, from http://www.atsdr.cdc.gov/toxfaqs/tfacts71.pdf Ashton, W. (1998). Framework for Water Quality Monitoring of the Kenai River. Retrieved 06/25/2015, from

UNIGIS Master of Science in GIS

109

http://ibrarian.net/navon/paper/FRAMEWORK_FOR_WATER_QUALITY _MONITORING_OF_THE_KEN.pdf?paperid=1405389 Bach, P. M., Rauch, W., Mikkelsen, P. S., McCarthy, D. T., & Deletic, A. (2014). A critical review of integrated urban water modelling – Urban drainage and beyond. Environmental Modelling & Software, 54, 88-107. doi: http://dx.doi.org/10.1016/j.envsoft.2013.12.018 Bartier, P. M., & Keller, C. P. (1996). Multivariate interpolation to incorporate thematic surface data using inverse distance weighting (IDW). Computers & Geosciences, 22(7), 795-799. doi: http://dx.doi.org/10.1016/0098-3004(96)00021-0 Bartram, J., & Ballance, R. (1996). Water Quality Monitoring - A Practical Guide to the Design and Implementation of Freshwater Quality Studies and Monitoring Programmes. Published on behalf of United Nations Environment Programme and the World Health Organization Retrieved 06/25/2015, from http://www.who.int/water_sanitation_health/resourcesquality/waterqualm onitor.pdf. Bash, J., Berman, C., & Bolton, S. (2001). EFFECTS OF TURBIDITY AND SUSPENDED SOLIDS ON SALMONIDS Center for Streamside Studies. University of Washington and Washington State Department of Transportation. Bengraı̈ ne, K., & Marhaba, T. F. (2003). Using principal component analysis to monitor spatial and temporal changes in water quality. Journal of Hazardous Materials, 100(1–3), 179-195. doi: http://dx.doi.org/10.1016/S0304-3894(03)00104-3 Bonansea, M., Ledesma, C., & Rodriguez, M. C. (2016). Assessing the impact of land use and land cover on water quality in the watershed of a reservoir. Applied Ecology and Environmental Research 14(2), 447-456. doi: 10.15666/aeer/1402_447456 Brabets, T. P., Nelson, G. L., Dorava, J. M., & Milner, A. M. (1999). WaterQuality Assessment of the Cook Inlet Basin, Alaska—Environmental Setting. (99-4116). Anchorage: Retrieved 06/25/2015, from http://ak.water.usgs.gov/Publications/pdf.reps/wrir99.4025.pdf. Cadkin, J. (2002). Understanding Dynamic Segmentation, Working With Events in ArcGIS 8.2. Retrieved 06/24/2015, from http://www.esri.com/news/arcuser/1002/files/dynseg_2.pdf Chang, H. (2008). Spatial analysis of water quality trends in the Han River basin, South Korea. Water Research, 42(13), 3285-3304. doi: http://dx.doi.org/10.1016/j.watres.2008.04.006 Clark, S. E. (1970). Water Quality Management Research Needs for Alaska. Federal Water Quality Administration Department of the Interior. UNIGIS Master of Science in GIS

110

Retrieved 06/26/2015, from https://nepis.epa.gov/Exe/ZyNET.exe/9100ZNBW.txt?ZyActionD=ZyDoc ument&Client=EPA&Index=Prior%20to%201976&Docs=&Query=&Time =&EndTime=&SearchMethod=1&TocRestrict=n&Toc=&TocEntry=&QFiel d=&QFieldYear=&QFieldMonth=&QFieldDay=&UseQField=&IntQFieldOp =0&ExtQFieldOp=0&XmlQuery=&File=D%3A%5CZYFILES%5CINDEX% 20DATA%5C70THRU75%5CTXT%5C00000016%5C9100ZNBW.txt&Us er=ANONYMOUS&Password=anonymous&SortMethod=h%7C&MaximumDocuments=1&FuzzyDegree=0&ImageQuality=r75g8/r75g8/x 150y150g16/i425&Display=hpfr&DefSeekPage=x&SearchBack=ZyAction L&Back=ZyActionS&BackDesc=Results%20page&MaximumPages=1&Z yEntry=1 Councell, T. B., Duckenfield, K. U., Landa, E. R., & Callender, E. (2004). TireWear Particles as a Source of Zinc to the Environment. ENVIRONMENTAL SCIENCE & TECHNOLOGY, 38(15), 4206-4214. Ding, J., Jiang, Y., Fu, L., Liu, Q., Peng, Q., & Kang, M. (2015). Impacts of Land Use on Surface Water Quality in a Subtropical River Basin: A Case Study of the Dongjiang River Basin, Southeastern China. Water, 7(8), 4427. ESRI. (2010). What is linear referencing. ArcGIS Resources. Retrieved 06/24/2015, from http://resources.arcgis.com/en/help/main/10.1%20/index.html#/What_is_l inear_referencing/003900000001000000/ Fan X., Cui B., Zhao H., Zhang Z., Zhang H. (2010). Assessment of river water quality in Pearl River Delta using multivariate statistical techniques. Paper presented at the International Society for Environmental Information Sciences 2010 Annual Conference (ISEIS). Fan, F. M., Fleischmann, A. S., Collischonn, W., Ames, D. P., & Rigo, D. (2015). Large-scale analytical water quality model coupled with GIS for simulation of point sourced pollutant discharges. Environmental Modelling & Software, 64, 58-71. doi: http://dx.doi.org/10.1016/j.envsoft.2014.11.012 Flynn, K. M. (1999) The Alabama Watershed Demonstration Project: Water Quality and Geographic Information Systems. Alabama Cooperative Extension System ANR-1162. Alabama A&M and Auburn Universities. Retrieved 06/25/2015, from http://www.aces.edu/pubs/docs/A/ANR1162/ANR-1162.pdf Gassman, P. W., Reyes, M. R., Green, C. H., & Arnold, J. G. (2007). THE SOIL AND WATER ASSESSMENT TOOL: HISTORICAL DEVELOPMENT, APPLICATIONS, AND FUTURE RESEARCH DIRECTIONS. American Society of Agricultural and Biological Engineers, 50(4), 1211-1250. Geosphere Inc. (2006). Hydrocarbon Characterization for Use in the Hydrocarbon Risk Calculator and Example Characterizations of Selected UNIGIS Master of Science in GIS

111

Alaskan Fuels Technical Background Document and Recommendations: CH2MHILL. Glass, R. L. (1999). Water-Quality Assessment of the Cook Inlet Basin, Alaska—Summary of Data Through 1997. (99-4116). Retrieved 06/23/2015, from http://ak.water.usgs.gov/Publications/pdf.reps/wrir99.4116.pdf. Glass, R. L. (2001). Ground-Water Quality, Cook Inlet Basin, Alaska, 1999. (4208). Retrieved 06/23/2015, from http://pubs.usgs.gov/wri/wri01-4208/. Glass, R. L., & Frenzel, S. A. (2001). Distribution of Arsenic in Water and Streambed Sediments, Cook Inlet Basin, Alaska. Retrieved 06/23/2015, from http://pubs.usgs.gov/fs/fs-083-01/pdf/fs-083-01.pdf. Grunwald, S., & Qi, C. (2006). GIS-BASED WATER QUALITY MODELING IN THE SANDUSKY WATERSHED, OHIO, USA1. JAWRA Journal of the American Water Resources Association, 42(4), 957-973. doi: 10.1111/j.1752-1688.2006.tb04507.x Kadhem, A. J. (2013). Assessment of Water Quality in Tigris River-Iraq by Using GIS Mapping. Natural Resources, 4, 441-448. Khairy, W. M., Hannoura, A. P., Cothren, G. M., & McCorquodale, A. (2000). Application of information systems in modelling of nonpoint pollution. Paper presented at the 4th International Conference on Integrating GIS and Environmental Modeling (GIS/EM4): Problems, Prospects and Research Needs. , Banff, Alberta, Canada. Retrieved 06/23/2015, from http://www.srcosmos.gr/srcosmos/showpub.aspx?aa=5833 Knapp, G. (2012). Economic Impacts of Kenai Peninsula Borough Fish Industries [Powerpoint slides]. Retrieved 08/26/2016,from http://www2.borough.kenai.ak.us/AssemblyClerk/assembly/Info/2012/102 312/Econonics%20of%20the%20Seafood%20Industry%20Presentation. pdf KWF, Kenai Watershed Forum. (2001). Quality Assurance Project Plan, submittal to the Alaska Department of Environmental Conservation: Division of Air and Water Quality, 95 p. KWF, Kenai Watershed Forum. (2003). Non-point Source Pollution Monitoring – Hydrocarbons Fact Sheet, (NPS-FS-001), 4 p. Retrieved 06/24/2015, from http://web.acsalaska.net/~kenaiwatershed.forum/factsheet_hydrocarbon. pdf KWF, Kenai Watershed Forum. (2012). Turbidity Monitoring on the Lower Kenai River, 2008-2010. Retrieved 06/23/2015, from https://dec.alaska.gov/water/wnpspc/protection_restoration/KenaiRiverW Q/pdfs/KWF_KENAI_RIVER_TURBIDITY_REPORT.pdf

UNIGIS Master of Science in GIS

112

KWF, Kenai Watershed Forum. (2015). Water Research Program. Retrieved 06/24/2015, from http://kenaiwatershed.org/science-in-action/researchinformation/water-quality/ Kyle, R. E., & Brabets, T. P. (2001). Water temperature of streams in the Cook Inlet basin, Alaska, and implications of climate change. (2001-4109). Retrieved 06/25/2015, from http://pubs.er.usgs.gov/publication/wri014109. Litchfield, V. P., & Kyle, G. B. (1992). Kenai River Water Quality Investigation Completion Report. (123). Retrieved from 06/25/2015, http://www.adfg.alaska.gov/FedAidpdfs/FRED.123.pdf. Lu, G. Y., & Wong, D. W. (2008). An adaptive inverse-distance weighting spatial interpolation technique. Computers & Geosciences, 34(9), 1044-1055. doi: http://dx.doi.org/10.1016/j.cageo.2007.07.010 Mena, C. F., Walsh, S. J., Frizzelle, B. G., Xiaozheng, Y., & Malanson, G. P. (2011). Land Use Change on Household Farms in the Ecuadorian Amazon: Design and Implementation of an Agent-Based Model. Applied Geography (Sevenoaks, England), 31(1), 210–222. http://doi.org/10.1016/j.apgeog.2010.04.005 North Carolina Division of Water Quality Watershed Assessment Team. (2009). Development of a WQI for Watershed Assessments. Retrieved 06/25/2015, from https://ncdenr.s3.amazonaws.com/s3fspublic/Water%20Quality/Surface%20Water%20Protection/PDU/WAT/Pro jectFiles/WQIDevelopment_Task1and2andEEPcomments.pdf Oasis Environmental, Inc. (2004). Kenai River hydrocarbon assessment final report. Retrieved 06/25/2015, from https://dec.alaska.gov/water/wnpspc/pdfs/Kenai%20Report_FINAL_14Ja n04.pdf Ouyang, Y., Nkedi-Kizza, P., Wu, Q. T., Shinde, D., & Huang, C. H. (2006). Assessment of seasonal variations in surface water quality. Water Research, 40(20), 3800-3810. doi: http://dx.doi.org/10.1016/j.watres.2006.08.030 Palacios, K. (2000). The Potential of Dynamic Segmentation for Aquatic Ecosystem Management: Pacific Lamprey Decline in the Native Lands of the Confederated Tribes of Siletz Indians (Oregon, USA) (unpublished master’s thesis). Oregon State University, Corvallis, Oregon, USA. Retrieved 08/27/2016, from https://ir.library.oregonstate.edu/xmlui/handle/1957/29201?show=full Pelletier, G. J., Chapra, S. C., & Tao, H. (2006). QUAL2Kw – A framework for modeling water quality in streams and rivers using a genetic algorithm for calibration. Environmental Modelling & Software, 21(3), 419-425. doi: http://dx.doi.org/10.1016/j.envsoft.2005.07.002

UNIGIS Master of Science in GIS

113

Quinn, N. W. T., & Hanna, W. M. (2003). A decision support system for adaptive real-time management of seasonal wetlands in California. Environmental Modelling & Software, 18(6), 503-511. doi: http://dx.doi.org/10.1016/S1364-8152(03)00025-2 Rickwood, C., Carr G., (2007). Global Drinking Water Quality Index Development and Sensitivity Analysis Report. In U. N. E. P. G. E. M. S. G. W. Programme (Ed.): United Nations Environment Programme Global Environment Monitoring System (GEMS)/Water Programme. Schwarz, G. E., Hoos, A. B., Alexander, R. B., & Smith, R. A. (2006). The SPARROW Surface Water-Quality Model: Theory, Application and User Documentation. Simeonov, V., Stratis, J. A., Samara, C., Zachariadis, G., Voutsa, D., Anthemidis, A., . . . Kouimtzis, T. (2003). Assessment of the surface water quality in Northern Greece. Water Research, 37(17), 4119-4124. doi: http://dx.doi.org/10.1016/S0043-1354(03)00398-1 Srebotnjak, T., Carr, G., de Sherbinin, A., & Rickwood, C. (2012). A global Water Quality Index and hot-deck imputation of missing data. Ecological Indicators, 17, 108-119. doi: http://dx.doi.org/10.1016/j.ecolind.2011.04.023 Srivastava, G., & Kumar, P. (2013). Water Quality Index with Missing Parameters (p. 609) by , 2013, International Journal of Research in Engineering and Technology, 2 (4), p. 609. Copyright ÂŠ 20142016 eSAT Publishing House Pvt. Ltd. All rights reserved Strager M.P., Fletcher J. J., Strager J.M., Yuill C.B., Eli R.N., Petty J.T,, Lamont S.J. (2010). Watershed analysis with GIS: The watershed characterization and modeling system software application. Computers & Geosciences, 36, 970-976. Stoner, J. D. (1978). Water-Quality Indices for Specific Water Uses. Arlington, VA: Geological Survey Circular. Tsihrintzis, V., Hamid, R., & Fuentes, H. (1996). Use of Geographic Information Systems (GIS) in water resources: A review. Water Resources Management, 10(4), 251-277. doi: 10.1007/BF00508896 UNEP, United Nations Environmental Programme. (2008). VITAL WATER GRAPHICS An Overview of the State of the Worldâ&#x20AC;&#x2122;s Fresh and Marine Waters (Second ed.). United States Census Bureau. (2016). Kenai Peninsula, Alaska. Retrieved 02/03/2017, from http://www.census.gov/quickfacts/table/PST045215/02122 University of Alaska Center for Economic Development. (2016). Kenai Peninsula Borough Comprehensive Economic Development Strategy UNIGIS Master of Science in GIS

114

2016. Retrieved 02/03/2017, from https://www.uaa.alaska.edu/academics/business-enterpriseinstitute/center-for-economicdevelopment/reports/kenai_peninsula_borough_ceds_072216.pdf UNW, United Nations Water. (2012). Status Report on The Application of Integrated Approaches to Water Resources Management: UN-Water. USEPA, United States Environmental Protection Agency. (1976). Quality Criteria for Water Retrieved 06/25/2015, from http://water.epa.gov/scitech/swguidance/standards/criteria/current/upload /2009_01_13_criteria_redbook.pdf. USEPA, United States Environmental Protection Agency. (1980a). Ambient Water Quality Criteria for Chromium. Retrieved 06/25/2015, from http://water.epa.gov/scitech/swguidance/standards/upload/2001_10_12_ criteria_ambientwqc_chromium80.pdf. USEPA, United States Environmental Protection Agency. (1980b). Ambient Water Quality Criteria for Lead. Retrieved 06/24/2015, from http://water.epa.gov/scitech/swguidance/standards/criteria/current/index. cfm USEPA, United States Environmental Protection Agency. (1987). Ambient Water Quality Criteria for Zinc â&#x20AC;&#x201C;1987. Retrieved 06/25/2015, from http://nepis.epa.gov/Exe/ZyNET.exe/2000M5KN.txt?ZyActionD=ZyDocu ment&Client=EPA&Index=1986%20Thru%201990&Docs=&Query=&Tim e=&EndTime=&SearchMethod=1&TocRestrict=n&Toc=&TocEntry=&QFi eld=&QFieldYear=&QFieldMonth=&QFieldDay=&UseQField=&IntQField Op=0&ExtQFieldOp=0&XmlQuery=&File=D%3A%5CZYFILES%5CINDE X%20DATA%5C86THRU90%5CTXT%5C00000008%5C2000M5KN.txt& User=ANONYMOUS&Password=anonymous&SortMethod=h%7C&MaximumDocuments=1&FuzzyDegree=0&ImageQuality=r75g8/r75g8/x 150y150g16/i425&Display=p%7Cf&DefSeekPage=x&SearchBack=ZyAct ionL&Back=ZyActionS&BackDesc=Results%20page&MaximumPages=1 &ZyEntry=1. USEPA, United States Environmental Protection Agency. (2000). 2000 UPDATE OF AMBIENT WATER QUALITY CRITERIA FOR CADMIUM. Retrieved 06/25/2015, from http://water.epa.gov/scitech/swguidance/standards/criteria/aqlife/cadmiu m/upload/2000_08_17_criteria_cadmium_cadmium.pdf. USEPA, United States Environmental Protection Agency. (2003). Minor Clarification of National Primary Drinking Water Regulation for Arsenic; Final Rule (EPA 815-Z-03-002): U.S. Environmental Protection Agency. Retrieved 06/24/2015 from https://www.federalregister.gov/articles/2003/03/25/03-7048/minorclarificationof-national-primary-drinking-water-regulation-for-arsenic

UNIGIS Master of Science in GIS

115

USEPA, United States Environmental Protection Agency. (2007). AQUATIC LIFE AMBIENT FRESHWATER QUALITY CRITERIA - COPPER 2007 Revision. Retrieved 06/25/2015, from http://water.epa.gov/scitech/swguidance/standards/criteria/aqlife/copper/ upload/2009_04_27_criteria_copper_2007_criteria-full.pdf. USEPA, United States Environmental Protection Agency. (2013). The Importance of Water to the U.S. Economy. Retrieved 06/25/2015, from http://water.epa.gov/action/importanceofwater/upload/Importance-ofWater-Synthesis-Report.pdf. USEPA, United States Environmental Protection Agency. (2014). Water Quality Standards Handbook. Retrieved 06/25/2015, from http://www2.epa.gov/wqs-tech/water-quality-standards-handbook. USEPA, United States Environmental Protection Agency. (2015a). CopperFree Brake Initiative. Retrieved 08/06/2015, from http://water.epa.gov/polwaste/npdes/stormwater/copperfreebrakes.cfm USEPA, United States Environmental Protection Agency. (2015b). Water Quality Analysis Simulation Program (WASP). Retrieved 06/24/2015, from http://www2.epa.gov/exposure-assessment-models/water-qualityanalysis-simulation-program-wasp USEPA, United States Environmental Protection Agency. (2015c). What are Water Quality Standards? Retrieved 06/24/2015, from http://www2.epa.gov/standards-water-body-health/what-are-waterquality-standards USEPA, United States Environmental Protection Agency. (2016). Polluted runoff: Non point source pollution. Retrieved 02/03/2017, from https://www.epa.gov/nps/what-nonpoint-source USEPA, United States Environmental Protection Agency. (n.d). EPA National Library Catalog. Retrieved 02/03/2017, from https://cfpub.epa.gov/ols/catalog/advanced_bibliography.cfm?&FIELD1= SUBJECT&INPUT1=Tolerances&TYPE1=EXACT&LOGIC1=AND&COLL =&SORT_TYPE=MTIC&start_row=251 USFWS, United States Fish and Wildlife Service. (2010). Water Resources. Retrieved 06/24/2015, from https://www.fws.gov/alaska/water/about.htm USGS, United States Geological Survey. (2014). Frequently Asked Questions about the NHD & WBD Datasets. Retrieved 06/24/2015, from http://nhd.usgs.gov/Frequently+Asked+Questions+about+the+NHD+&+ WBD.htm USGS, United States Geological Survey. (2015). Hydrography, National Hydrography Dataset, Watershed Boundary Dataset. Retrieved 06/24/2016, from http://nhd.usgs.gov/

UNIGIS Master of Science in GIS

116

World Resources Institute. (n.d). Aqueduct Measuring and Maping Water Risk. Retrieved 10/12/2016 from http://www.wri.org/our-work/project/aqueduct WWAP, World Water Assessment Programme. 2012. The United Nations World Water Development Report 4: Managing Water under Uncertainty and Risk. Paris, UNESCO. Xu, H. S., Xu, Z. X., Wu, W., & Tang, F. F. (2012). Assessment and Spatiotemporal Variation Analysis of Water Quality in the Zhangweinan River Basin, China. Procedia Environmental Sciences, 13, 1641-1652. doi: http://dx.doi.org/10.1016/j.proenv.2012.01.157 Yan, C.-A., Zhang, W., Zhang, Z., Liu, Y., Deng, C., & Nie, N. (2015). Assessment of Water Quality and Identification of Polluted Risky Regions Based on Field Observations & GIS in the Honghe River Watershed, China. PLoS ONE, 10(3), e0119130. doi: 10.1371/journal.pone.0119130 Yang, X., & Jin, W. (2010). GIS-based spatial regression and prediction of water quality in river networks: A case study in Iowa. Journal of Environmental Management, 91(10), 1943-1951. doi: http://dx.doi.org/10.1016/j.jenvman.2010.04.011 Yu, S., Xu, Z., Wu, W., & Zuo, D. (2016). Effect of land use types on stream water quality under seasonal variation and topographic characteristics in the Wei River basin, China. Ecological Indicators, 60, 202-212. doi: http://dx.doi.org/10.1016/j.ecolind.2015.06.029 Zhou, Z., & Zhao, Y. (2011). Research on the Water Quality Monitoring System for Inland Lakes based on Remote Sensing. Procedia Environmental Sciences, 10, Part B, 1707-1711. doi: http://dx.doi.org/10.1016/j.proenv.2011.09.268 Zimmerman, D., Pavlik, C., Ruggles, A., & Armstrong, M. P. (1999). An Experimental Comparison of Ordinary and Universal Kriging and Inverse Distance Weighting. Mathematical Geology, 31(4), 375-390. doi: 10.1023/a:1007586507433

UNIGIS Master of Science in GIS

117