Term Project
Spring Semester 1998
CE 394K.3 GIS in Water Resources
The University of Texas at Austin
Using GIS to model Water Quality Data in the San Gabriel Watershed
by Michael D. Sufnarski
The objectives of this term project are to:
Information sources include:
Expected results include:
The quality of water in our environment is clearly a matter of increased public concern. We can pick up just about any newspaper or magazine and find an article about a polluted stream, river, or lake that is affecting the water we use for drinking, bathing, or recreation. Traditionally, the responsibility of monitoring and reporting the health of our streams and lakes has rested with state sponsored professional monitoring agencies. However, budget constraints in recent years has limited their ability to keep pace with expanded monitoring needs. Most professional agencies are now being used for "targeted monitoring" in which they only monitor water quality in state mandated areas of concern. In an effort to ensure a more complete coverage, many states are now implementing innovative strategies that use public volunteer monitoring groups to complement professional monitoring agencies. One of the model volunteer monitoring programs in the country, called Texas Watch, is located right here in Texas.
It was my personal interest to learn more about Texas Watch that led me to the topic of this term project: "Modeling Water Quality Data in the San Gabriel Watershed". As a first semester graduate student in the Environmental and Water Resources Program, this course was my first introduction to the field of Geographic Information Systems (GIS). By completing this term project I have gained valuable experience in the operations of ArcView and Arc/Info, the internet, html formatting, and in the overall field of water resources by interacting with several state organizations. It is my intent that by reading this report you will gain an appreciation of how both volunteer and professional monitoring data can be integrated into a GIS environment to assess water quality for a particular watershed. I have done my best to write this report so that a novice user of ArcView will be able to recreate or expand upon my efforts.
The watershed that I selected to model the volunteer and professional data is the San Gabriel Watershed. This watershed was selected primarily because of the availability of volunteer monitoring data from TNRCC. It is a part of the Brazos River Basin and is located in south central Texas just north of Austin. It covers an area of 1351.14 square miles that includes portions of Williamson, Milam, and Burnet counties. The 1990 census population was 137,675. The primary land use for the watershed is farmland at 47%. The major waterways in the watershed include the San Gabriel River and Georgetown and Granger Lakes. The climate for the watershed can be characterized by long hot summers and moderate winters. The mean annual temperature is 68 degrees fahrenheit. The mean daily temperature is 97 degrees fahrenheit in August (warmest month) to about 36 degrees fahrenheit in January (coldest month). The mean annual precipitation is about 32.3 inches; varying from about 15 inches in the upper part to 35 inches in the lower part of the watershed. The heaviest rainfall occurs between April and June. Click on the following image to view the watershed:
Texas Watch was established in 1991 in response to public concerns about a series of fish kills in the Pecos River. Sponsored by the Texas Natural Resources Conservation Commission , this organization is based on the premise that water quality issues are linked with air, biological, and human resource issues, and that protection of environmental resources requires cooperative participation of all citizens. It currenly consists of a network of nearly 4000 trained volunteers that work together to monitor the health of Texas lakes, rivers, streams, wetlands, bays, and estuaries. The program offers guidance and training to those citizens with water quality concerns so that they will be able to collect useful water quality data that can be used to support environmental management decisions. The three principal goals of Texas Watch are to:
Texas Watch participants represent a diverse group that includes professional scientists, students, teachers, scout troops, senior citizens, and community organizations. It is interesting to note that anyone with an interest in the environment can join. There are various levels of participation based on the type of training and certification. The fours levels of participation include:
In order to ensure accurate data, Texas Watch has a very strict quality assurance program that follows measures approved by the U.S. Environmental Protection Agency (EPA). The program covers data measurement, sampling and calibration techniques, quality control, and reporting techniques that results in data comparable to professional monitoring agencies.
The variables monitored by Texas Watch volunteers include a series of core tests and field observations. All measurements are recorded on the official monitoring form shown below. Click on the image below to see an expanded view:
The following is a narrative description of the variables recorded on the form:
Conductivity
Conductivity is a measure of the ability of water to pass an electrical current. Since dissolved substances in water dissociate into ions with the ability to conduct current, there is a direct relationship between conductivity and the presence of dissolved solids.
The quantity of dissolved solids depends mainly on the solubility of the rocks and soils that the water contacts. Water that flows through limestone (typical of Texas) tends to have high levels of dissolved ions since limestone readily dissolves into calcium, carbonate, and sulfate ions.
Discharges into water systems can also affect the conductivity. For example, a leaking sewer system near a stream could raise the conductivity because of the presence of chloride, phosphate, or nitrate ions. At the same time, an oil spill could lower conductivity since oil is a poor conductor of current.
Conductivity is measured in micro mhos per centimeter or microsiemens per centimeter. Distilled water has a conductivity for 0.5 to 3 micro mhos per centimeter. Studies indicate that Texas streams with healthy fish populations have a range between 150 and 500 micro mhos per centimeter.
Dissolved Oxygen
Oxygen is essential for aquatic life. In fact, without oxygen, a stream or lake could not support fish or plants. For this reason, many experts consider dissolved oxygen to be the most important indicator of water quality.
Oxygen is easily dissolved in water. It is so soluble that water can contain a greater percentage of oxygen than the atmosphere. Because of this phenomenon, water naturally diffuses from the air into the water. Oxygen is also produced by aquatic plants and algae as a by-product of photosynthesis. Once in the water, oxygen diffuses very slowly and distribution depends on the movement of aerated water by turbulence caused by the wind or by water flowing over rocks. Water temperature affects the capacity of the water to retain dissolved oxygen. Cold water can hold more oxygen than warm water.
Some of the factors that can lead to decreased levels of dissolved oxygen include increased algae growths that consume oxygen during respiration, effluent from wastewater treatment plants that contains high levels of biochemical oxygen demand (BOD), and detergents from leaking sanitary sewers or car washing.
The amount of oxygen required to sustain aquatic life varies from species to species. Generally, dissolved oxygen levels of 5.0 to 6.0 mg/L are required to support life. Levels below 3 mg/L are stressful to most aquatic organisms and levels below 2 mg/L will not support fish.
pH
pH is a measure of acidity in water. Small changes in pH can affect aquatic life indirectly by changing the chemistry of the water. An example is that toxic metals trapped in sediment may precipitate as a result of small changes in pH. These precipitated metals may be toxic to fish.
A range of pH 6.5 to pH 8.2 is optimal for most aquatic organisms.
Algae Cover
Algae are photosynthetic plants that are an important living component of water. Algae oxygenates the water through photosynthesis, converts inorganic material to organic matter, and serves as an important food for aquatic life. At the same time, it consumes oxygen during respiration and during the decomposition of organic matter. The most important factor for algae growth is its supply of nutrients.
Excessive algae growth (stimulated by an increased supply of nutrients), known as an algae bloom, is an indicator of cultural eutrophication. Cultural eutrophication is enrichment of the water from human activities that may include agriculture and residential developments. Algae blooms are a cause for concern since large blooms can lower dissolved oxygen concentrations and result in fish kills. Algae blooms can also give the water an unpleasant odor or color and form scums on the surface that can impact recreational activities. The two nutrients most responsible for algae blooms are phosphorous (in the form of phosphate) and nitrogen (in the form of nitrates or ammonia).
Phosphorus is an essential nutrient needed for the growth of plants and the metabolic functions of plants and animals. It normally enters water through inadequately treated sewage released from wastewater treatment plants or through non point sources like illegal sewer connections to storm sewers and animal wastes. Acceptable levels of phosphate range from 0.1 to 1.0 mg/L.
Nitrogen is another essential nutrient needed by plants and animals for protein building. It normally enters water through decomposition of dead plants and animals and through excretion from live animals. It can also enter the water through wastewater treatment water effluent, faulty septic systems, and from runoff from lawn and agriculture fertilizers. Acceptable levels of nitrate nitrogen are from 1 to 10 mg/L and ammonia nitrogen is generally found at levels less than 1.0 mg/L. Ammonia nitrogen can be toxic to fish from 0.2 to 4.8 mg/L depending on the pH and temperature of the water.
Water Color
A simple observation of water color is used as a common physical indicator of water quality. The following are some examples:
Odor
Common odors and their causes include:
Water Clarity or Turbidity
Turbidity is a measure of how much the concentration of suspended solids decrease the passage of light through the water. The most familiar cause of turbidity is suspended sediment due to soil erosion or turbulence; however, high turbidity may also be caused by microscopic plankton in the water. In addition to causing turbidity, suspended sediment can carry nutrients and pesticides throughout the water. Suspended particles near the surface can absorb heat from the sun and raise the temperature of the water.
High turbidity is associated with runoff from disturbed and eroded soil or blooms of plankton due to large quantities of nutrients. These nutrients may come from runoff from fertilizers containing nitrogen used in agriculture.
Texas Watch Data Management
The key to effective water quality management is the ability to recognize changes in water quality and take appropriate actions to minimize detrimental effects. Completed monitoring forms are sent to Texas Watch after each sample so water quality problems can be identified immediately. The database coordinator at TNRCC is responsible for entering the data into the Certified Water Quality Managers Database (CWQM). Manipulation of the CWQM database enables comparison and analysis of water quality measurements; however, there is currenly no link to a GIS environment where the results can be viewed in relation to a map. The GIS section is currently in the process of creating a GIS version similar to the project in this report.
The data from the CWQM database is also be entered into the TNRCC's Texas Regulatory and Compliance System (TRACS) database that is used for state water assessments. Volunteer data is "tagged" in TRACS to differentiate it from data collected by professional monitors.
Data from the CWQM database is currently not available from the internet (although it will be in the very near future). However, it is available by contacting the database manager (Patricia Davis) at TNRCC (Telephone 512-239-4739). You can request all of the available data or only the data for a specified watershed or river. For this project, I requested all of the available Texas Watch Data for the San Gabriel Watershed from 1992 to 1997. The data that I received was in delimited text format. It contained measurements for 10 stations and included over 313 records. The locations (latitude and longitude) of each of the monitoring sites was included as a separate file.
Pat Davis was extremely helpful during my visit to TNRCC. In addition to the data, I received a complete briefing about Texas Watch and numerous pamphlets and books about the program. I was also introduced to other water quality programs that TNRCC sponsors. I would highly encourage anyone with an interest in water quality to stop by and visit TNRCC.
As expected, the monitoring sites for the Texas Watch program did not extend throughout the entire watershed. In order to create a more complete coverage, I had to search for professional monitoring data. Finding this data was not as easy as I expected. Although there is an abundance of flow data available over the internet (USGS, Corps of Engineers, TNRCC), I had difficulty finding water quality data for the San Gabriel Watershed. I believe one of the reasons for this is that the watershed has relatively good water quality and is not the focus of any major studies at this time. During my search, I discovered many intersting sites on water quality. One of those sites is EPA Surf Your Watershed which can be used to find information about any watershed in the U.S..
I eventually discovered the EPA's STORET database. STORET was established in 1964 and is the EPA's central database for both surface and ground water. This database is managed by the Assessment and Watershed Protection Division of the EPA's Office of Water in Washington, DC. The database is broken down into 11 different areas: REACH, Industrial Facilities Discharge, Drinking Water, Gage, Biological Data, Daily Flow, Water Quality, Parameter, City, County, and Fish Kills. The section that applies to this project is the Water Quality System (WQS). WQS contains a collection of professional measurements arranged by monitoring sites.
Direct access to STORET is limited to those individuals or organizations who subscribe to an EPA user account. However, the data is accessible to the general public by contacting STORET User Assistance. Using the toll-free number (1-800-424-9067), I contacted the STORET staff and requested water quality data for the San Gabriel Watershed. The first step was to obtain an inventory of all the parameters for the San Gabriel Watershed that were in the database (a complete listing of parameters is available on the STORET parameter website). The STORET staff initiated a database search using the Hydrologic Unit Code (HUC) for the San Gabriel Watershed. HUC boundaries are a subdivision of the US made by the USGS that depict major and minor river basins. There are 2, 4, 6, and 8 digit boundaries where the larger the number, the smaller the area. The 8-digit HUC code for the San Gabriel Watershed is 12070205. Within 24 hours I received a text file (via email) showing all of the monitored parameters for the San Gabriel Watershed. I also received a telephone call at home to confirm receipt of the files and to ask if I had any questions. Click on the following image to see the format of the inventory:
After reviewing the inventory, I determined that I wanted the records from the database that most closely matched those monitored by Texas Watch (D.O. - code 00300, Conductivity - code 00095, pH - code 00403, Water Temp - code 00010, NH3-N - code 00610, NO3-N - code 00615, and Orthophosphates - code 00671). However, in order to extract the data, I had to learn the STORET database "language". There are four steps to follow when extracting the data:
Step 1: Select the Location
A location can be identified in one of four ways:
For this search I used the HUC code for the San Gabriel Watershed (12070205)
Step 2: Selection of Data
The required data must be identified in one of four ways (or any combination):
For this extraction I used a search by parameter code for the years 1992 to 1997.
Step 3: Extraction Format
The output extracted from the database can either be in the form of a list of sampled values or a statistical summary. Since I was using the data to conduct my own analysis in ArcView, I requested an extraction of sampled values. The options for the output format included: paper, diskette, attached email file, or ftp. Since the data file for the San Gabriel Watershed was relatively small, I requested output via an attached email file in a delimited text format (for use with Excel and ArcView).
After sending the request, I received the extracted file within 24 hours. Again, I received a call at home to confirm receipt and to ask if I had any questions. I found the STORET staff to be extremely helpful and prompt during the entire process. The file contained measurements for 14 stations and included over 464 records. However, the data was not in the most user-friendly format. Click on the following image to see the format of the data:
The files created in the process of building this model quickly exceed what can be stored on a single 1.44MB diskette. As a result I purchased my first zip disk which contained more than enough space (100MB) for the entire project. I would highly recommend the purchase of a zip disk for anyone taking this course.
1) Create Basemap for Watershed (Boundary, Streams, and Lakes).
There are many different ways to create a basemap for the San Gabriel Watershed. Basically, I pasted together portions of various themes that we used in class or that I downloaded from the internet and created smaller shapefiles of the specific area of interest. These are all techniques that we used in GIS Exercise 4: Building a base Map and a Point Shapefile.
Watershed Boundary:
The San Gabriel Watershed boundary theme (Sangab.shp) was developed by using the Table Query Builder tool in ArcView. I queried the 8-digit HUC of the watershed from the HUC coverage of Texas (huctx.e00) that was created by Dr. Maidment. He created this coverage from the US Geological Survey's HUC coverage by querying the HUC zones for Texas (11,12, and 13) in a process similar to the one described below. The huctx.e00 coverage is available in the class directory under Civil2/LRC/Class/Maidment/giswr/basemap/. HUC coverages for the entire US can be downloaded from the USGS website. Since the huctx.e00 file is in Arc/Info format (identified by .e00 extension), I had to use the ArcView Import 71 utility to import the coverage into ArcView. The first step in the query process was to add the huctx polygon theme to the view. After making it the active theme and calling up the associated table, I used Theme/Properties and the 
icon to create the following query:
By clicking on OK, the view went from displaying the entire huctx polygon to displaying just the polygon for the San Gabriel Watershed. The completed query was converted to an ArcView shapefile using Theme/Convert to Shapefile. The resulting theme, named sangab.shd, is much smaller since it only contains the attributes associated with the San Gabriel Watershed. Click on the following image to see what was accomplished by the query:
NOTE: The projection of this theme (sangab.shp) is Texas State Mapping System (TSMS) Albers Equal Area. To learn more about map projections read the Map Projection Overview prepared by the Department of Geography.
Stream and Rivers:
Using the same technique as for the watershed boundary, the Table Query Builder tool was used to select the US Environmental Protection Agency's River Reach File 1 (RF1) that corresponded to the 8-digit HUC for the San Gabriel Watershed. Again, a file called rf1tx.e00 was created by Dr. Maidment that contains only the Rf1 reaches for Texas. The Rf1 river reach files are a series on national hydrological databases that identify and interconnect stream segments of the surface water drainage system of the entire US. To learn more about Rf1 files see the EPA website. The final step of creating this theme was to use the Theme/Convert to Shapefile command to create Rfgab.shp which only contained the attributes for the San Gabriel Watershed. Click on the following image to see Rfgab.shp with the rivers named using the Theme/Auto Label command.
NOTE: The projection of this theme is also TSMS Albers Equal Area
Lakes: The Rf1 coverage does not depict lakes. To get a lake coverage for Texas, I used the internet to download the coverage from the Texas Natural Resources Information System (TNRIS) web page. This file, lakes.e00.gz, was in a compressed format and had to be unzipped using WINZIP. As a novice computer user, this was my first time to unzip a file but I found it very easy. After extracting the file, I used the Import 71 utility to bring it into ArcView. After displaying the coverage and reading the readme file, I noticed two problems with this theme. The first was that the theme was in a Lambert Conformal Conic projection which, although similar, is not compatible with the TSMS Albers projection of the other two themes. The second problem was that the lake theme did not contain Georgetown Lake which, as noted, is a major body of water in the San Gabriel Watershed.
Converting Granger Lake Theme to TSMS Albers Projection using Arc/Info:
Since ArcView cannot conduct this transformation, I had to use Arc/Info. Before doing the conversion, I queried the lake theme to create an ArcView shapefile containing only the attributes for Granger lake. The next step of this process was to ftp the three files (granger.shp, granger.shx, and granger.dbf) from my windows NT account (personal folder) to my Unix account. For those of you who are not familiar with using ftp (I never used ftp prior to this course), it is not that difficult. The following is a easy way to complete the transfer. While in the main screen for windows, click on the start button in the lower left hand corner and click on run. The following window will appear in which you must enter cmd.
At the next screen, follow the steps as outlined in the following screen capture to complete the ftp. It is essential to specify binary format for the transfer to work.
The files transferred to the Unix account are ArcView shapefiles which are not understood by Arc/Info and must be converted to Arc/Info coverage files using the following commands:
Arc: shapearc granger granger1
Arc: clean granger1 granger2 # # poly
cleaning /HOME2/SUFNARMD/GRANGER
Sorting...
Intersecting...
Assembling Polygons...
Rebuilding AAT...
Arc: build granger2 poly
Building polygons...
Re-building AAT...
Arc: items granger2.pat
NOTE: To learn more about this process see GIS Exercise 3: Map Projections.
Before doing the new projection, I had to create a projection file using the text editor. This text file tells Arc/Info the parameters of the existing coverage (input) and the parameters of the new projection (output). The file that I created to transform the coverage from a TSMS lambert projection to a TSMS albers projection (albers.prj) is shown below:
input
projection lambert
units meters
datum NAD83
spheroid GRS1980
parameters
27 25 0.000
34 44 0.000
-100 0 0.000
31 10 0.000
1000000.0
1000000.0
output
projection albers
units meters
datum NAD83
spheroid GRS1980
parameters
27 25 00
34 55 00
-100 00 00
31 10 00
1000000.0
1000000.0
end
The following command was used to conduct the transformation:
Arc: project cover granger2 granger4 albers.prj
After completing the transformation, the file was ftp'd back to my NT account and was added to the view. Since they were both in the TSMS parameters, there was only a small difference. Click on the following image to see the lake before (Tx_Lake) and after (Granger4) the projection transformation:
Obtaining/Projecting Coverage of Georgetown Lake
After contacting the GIS section at TNRIS (where I obtained the lake coverage), I learned that only about 60% of the lakes in Texas are currently included in the coverage. The lakes not included on the coverage either have not been surveyed or have not yet been added to the lakes.e00 coverage. I was referred to the reservoir survey section of the Texas Water Development Board (TWDB) who is responsible for conducting the surveys. I learned that a survey of Georgetown Lake was complete and could be downloaded from the website in Arc/Info export format (.e00).
Since this coverage was in State Plane NAD83, it had to be converted to TSMS Albers Equal Area using Arc/Info. The only difference in the process from the Granger Lake projection was the parameters of the projection file. The projection file created in the text editor for this conversion (spalbers.prj) is shown below:
input
projection STATEPLANE
datum NAD83
units feet
parameters
output
projection albers
units meters
datum NAD83
spheroid GRS1980
parameters
27 25 00
34 55 00
-100 00 00
31 10 00
1000000.0
1000000.0
end
As with the Granger Lake projection, the new theme was then added to the view in ArcView. The following image shows how the Georgetown Lake looked before and after the projection:
2) Creating and Projecting Point Shapefile for Monitoring Sites
NOTE: The following describes how I did this for the Texas Watch site. The process was exactly the same for the STORET sites.
The locations of monitoring sites for both the Texas Watch and STORET data that I received were in geographic degrees, minutes and seconds for longitude and latitude (taken from USGS maps). Since GIS works in decimal degrees, the coordinates had to be converted before they could be placed on the basemap. The following formula was used to do the conversion:
Decimal Degrees = Degree + minutes/60 + seconds/3600 (West longitude is negative in decimal degrees)
The easiest way to do the conversions was on an Excel spreadsheet. The spreadsheet must be saved as a text (tab delimited) file to be able to import it into ArcView in a later step. Click on the following image to see an example of the spreadsheet I used to for the conversions of the Texas Watch monitoring sites:
Import Table into ArcView: Using the 
icon in the project window and Add, the delimited text file for the Texas Watch monitoring sites was added to the view.
Generate and Project Point Shapefile for Monitoring Sites: At this point, the basemap of the San Gabriel Watershed is in an Albers Equal Area Projection while the monitoring sites are in geographic coordinates (decimal degrees). The conversion from the geographic coordinates to an Albers Equal Area Projection and the creation of a point shapefile can be done in one of two ways. One way is to create a projection file in ArcInfo but the easier way is to use an avenue script in ArcView (written by a GIS student at the University of Texas). I chose to use the avenue script (defgages.ave) which converts the datum from NAD27 (datum used for USGS maps for which latitude and longitude was originally extracted) to NAD83. The only modification that I made to the script was to change the title of one of the output fields (alberstat# to Station ID). This modification was completed using the text editor in ArcView. An excellent resource that contains many other prepared scripts is the Environmental Systems Research Institute (ESRI) website.
The TSMS projection parameters used in the defgages.ave script are:
central meridian: -100
lower standard parallel: 27.4167
upper standard parallel: 34.9167
reference latitude: 31.1667
false easting: 1000000
false northing: 1000000
spheroid: #spheroid_grs80
After compiling
and running 
the script, a new point shapefile with an associated attribute table was created. At this point there is a relationship that allows the user to click on the point attribute table and see the monitoring site on the basemap highlighted in yellow. Click on the following image to see this relationship:
3) Adding Attributes to the Point Shapefile
The purpose of this step is to add another field (Station Description) to the point attribute table. To accomplish this, I had to make the attributes of txsites.shp table active and use the Table/Start Editing command. After clicking on Edit/Add Field a new dialog box appears in which I typed:
name Station Description
type String (allows you to enter either numbers or letters)
width 75
After the new field appears, the 
icon is used to add the names of the respective monitoring sites into the Station Description field. After completing the editing, the Table/Stop Editing command saves the updated table. Click on the following image to see what was accomplished by this process:
4) Formatting Texas Watch and STORET Data for ArcView
Texas Watch Data: As stated previously, the Texas Watch data for the San Gabriel Watershed was obtained from the Texas Watch database coordinator at TNRCC in the form of an Excel spreadsheet. Overall, I only had to make minor cosmetic changes to the spreadsheet. In order to link this data with the point attribute table, I had to ensure that the point attribute table (.PAT) and data table shared a common field. Station ID was the name of the field I used. This field contained the 5 digit number assigned to the monitoring site. The revised data file was saved as text (tab delimited) so it could be imported into ArcView when linking the tables.
STORET Data: Formatting the STORET data was not as easy. The delimited text file that I received via email from STORET had to be opened using the text import wizard in Excel. This is a 3 step process. The first step was to change the type of file to formatted text from delimited text. This allowed me to manually format the columns by fixed field widths. The second step was to create column breaks by dragging and creating or deleting break lines. As you can see in the following screen capture, I had to manipulate the break lines to get rid of the commas and spaces.
In the third step I chose "general" for column data and previewed the table. The resulting Excel table contained the data neatly delineated into columns.
At this point, I had to format the column headings by changing the parameter codes to words (i.e. 00300 as depicted in table above had to be changed to Dissolved Oxygen, etc). As with the Texas Watch data, I had to ensure that there was a common field between the point attribute table and the data table. Again, I chose to change the field for monitoring site numbers to Station ID. After the data was in the format I wanted, I saved it as text (tab delimited) so it could be imported into ArcView.
5) Linking Point Attribute Table to Monitoring Data
This is the point when the water quality data for each monitoring site is linked to its respective point attribute table. At the end of this step, there is a one to many relationship between each monitoring site and the recorded measurements in the data table.
The first step was to get both the point attribute table and the data table into the same view. To do this, I used the add button in the 
section of the project window to add the delimited text data file to the view. Once both tables were in the current view, I was ready to link the two tables. This was accomplished by first highlighting the Station ID field of the data file (txwatch.txt) and then making attributes of txsites.shp active and highlighting its Station ID field; the two tables were then linked by using the Link command under the tables toolbar. I used the link command as opposed to the join command since there was a one to many relationship. Click on the following image to see the "link" relationship:
The completed model is a basemap of the San Gabriel Watershed that includes rivers, lakes, and locations for both volunteer and professional monitoring sites. The most import part of this GIS model is that the monitoring sites on the map are linked to data tables containing water quality measurements. Click on the following image to see the completed model and the associated tables:
ArcView is a powerful tool that allows the user of this model to display, query, summarize, and organize the data in order to conduct an analysis of the watershed. The following are some examples of products produced by using this model.
DISSOLVED OXYGEN vs TEMPERATURE: The overall dissolved oxygen levels in the watershed can be characterized by the following chart which shows the classic relationship between water temperature and dissolved oxygen. The lowest DO levels occur during the summer months when water temperatures are the warmest. This is when the water possesses its least capacity to hold DO. Water at 31 deg C (typical of Texas water in August) can only hold half as much oxygen as water at 3 deg C (typical of Texas water in January). Although the chart shows that DO levels drop significantly, they still remains above the 3 mg/l which is accepatable to most aquatic species.
AVERAGE NH3-N CONCENTRATION: The Field/Statistics command was used to determine the average NH3-N concentration for STORET Monitoring Site 12107 in the North Fork of the San Gabriel River. The average concentration for that site (from 1988 to 1994) was 0.04 mg/l. This is below the 0.2 to 4.8 mg/l level that can be toxic to fish. The method for conducting this analysis is quite easy. Since data records are linked to the monitoring sites, the associated records were highlighted in yellow after clicking on Station 12107 in the Attributes of Storsites.shp table. After clicking on the field name (NH3-N mg/l) in the data table and using the Field/Statistics command, ArcView produced a statistics summary for all of the highlighted records in the NH3-N field. Records included in the summary can be added or deleted by highlighting the respecitive records in the data table. To get a summary of a different field, the user just has to highlight the name of that field and run the Field/Statistics command again. Click on the following image to see the result:
CONDUCTIVITY: The following chart for Texas Watch Station 13496 shows that conductivity levels from 1992 to 1996 have remained above the 150 to 500 micro mhos per centimeter (typical of streams with healthy fish populations).
pH LEVELS: The following is an example of a layout produced in ArcView which depicts a zoomed-in view of the eastern portion of the watershed along with the highlighted point attribute table and a chart of pH levels for that station. The charts shows that although the water entering Granger Lake tends to be on the alkaline side, it remains within the acceptable range.
River Reach Information: The 
icon can be used to identify information about any feature on the map by clicking on that feature with the respective theme active. The following image shows the information identified by clicking on Russel Creek with the Rfgab.shp active. The information identified includes the name of the river, HUC codes, Rf1 ID number, and other information associated with that location.
The development and use of this model indicates that there are no significant water quality concerns in the San Gabriel Watershed. An analysis of the available data indicates that dissolved oxygen, pH, NH3-N, NO3-N, and conductivity are all within acceptable ranges. These findings agree with the assessment of the watershed made by TNRCC and as a result, the watershed is currently not "targeted" or the focus of any major watershed studies. However, in order to maintain the health of this watershed, it is essential that Texas Watch volunteers continue to monitor and immediately identify any changes in water quality. The importance of volunteer monitoring organizations will continue to expand as state resources continue to shrink.
A GIS model, similar to this one, is an excellent tool for any volunteer monitoring program to analyze and display their results. It could be used to produce interesting layouts and graphs for the public, volunteer monitors, or school children involved with monitoring. GIS is a powerful tool that effectively links data to a spatial environment and enables the user to solve problems by uncovering and analyzing trends. This report only brushed the surface of the capabilities of ArcView and ArcInfo. For future work, I would recommend adding a coverage of wastewaster treatment facility discharge points and a coverage of land use to see if they have any local effects on water characteristics.
By completing this term project I have gained valuable experience in the operations of ArcView and ArcInfo, the internet, html formatting, and in the overall field of water resources by interacting with several state organizations (TNRCC, TWDB, EPA). I found the internet to be full of interesting websites dealing with water quality and the environment. The real challenge of this project was finding and formatting the data. As with any model, the results are only as good as the input data.
(1) Volunteer Environmental Monitoring in Texas: 1995 Texas Watch Report, (October 1995), Texas Natural Resources Conservation Commission.
(2) Texas Watch Volunteer Environmental Monitoring Manual, (1997), Texas Natural Resources Conservation Commission.
(3) Kaough, C. W., "Watershed Water Quality Data and GIS", CE394K Term Project (Spring 97), http://www.ce.utexas.edu/prof/maidment/tmpaper/spring97/kaoughcw/termproj.htm.
(4) Texas Natural Resources Conservation Commission (TNRCC) Internet Site: http://www.tnrcc.state.tx.us.
(5) Texas Water Development Board (TWDB) Internet Site: http://www.tndb.state.tx.us.
(6) US Environmental Protection Agency Internet Site: http://www.epa.gov.
(7) ArcView GIS Users Manual, Environmental Systems Research Institute.
(8) CE 394K Exercise 4: Building a Basemap, http://www.ce.utexas.edu/prof/maidment/giswr98/ex498/base_map.htm
(9) CE 394K Exercise 3: Map Projections, http://www.ce.utexas.edu/prof/maidment/giswr98/ex398/mapproj.htm
