Transcript Analyzing geospatial variations in water quality caused by acid mine

Dr. Paul T. Ryberg, Clarion University, Clarion, PA
This project analyzes the geospatial
variations in water quality (pH, iron and
sulfate) caused by acid mine drainage from
strip coal mining within a local watershed
(Toby Creek) in northwestern Pennsylvania
Applied GIS course
Taught by Geography faculty member in our department
Geology faculty as advisors for student geoscience projects
Students taking the Applied GIS course must have first
successfully completed the basic GIS undergraduate
course.
They must have the skills to construct and query
databases, and work with a variety of spatial data and
imagery
Hydrogeology Project for an
Undergraduate Applied GIS Course
This project is used instead of a final exam in an upper level
undergraduate course in Applied GIS. A student may propose
their own project, or choose one from a list from the instructor.
A brief proposal must be approved by the instructor before the
project can begin. Over several weeks, students will have
regular meetings with the instructor to verify progress and sort
out any difficulties. Students will construct a working, queryable database, use appropriate imagery, and use it to analyze a
problem, understand cause and effect, and show changes with
time. A final report must be submitted with all supporting
documentation in digital form. Students also give a PP
presentation in one of the last class meetings
Concept Goals
Students should be able to define the scope
and extent of a specific project
Students should be able to construct a
working, query-able database appropriate to
the project
Students should be able to produce a clear,
detailed written summary of the project
Students should be able to construct clear,
detailed cross-sections and maps appropriate
to the project
Students should be able to present a clear
and logical summary of the project to others
Higher Order Thinking Goals
Students should be able to analyze the relationships of the
geospatial data to recognize changes in time and space
Students should be able to develop strategies for problem
solving (i.e. remediation)
Students should be able to use geospatial data to make future
predictions in time and space
Database
In order to observe the effects of AMD on Toby Creek’s
watershed, data was collected at multiple points along
Toby Creek and its tributaries. The collection was done by
a professor and students from Clarion University during
the summers of 2000 through 2009. Water samples were
analyzed for the following parameters: Temperature,
Conductivity, Eh, pH, total Iron and Sulfate
concentration. The samples were collected at the same
designated locations each year, which allows analysis of
change in water quality over time and place. Tables were
created in Microsoft Excel to organize the data that was
then incorporated into a database in ArcCatalog 9.3
Methods
A database was created in ArcCatalog that holds layers and feature classes of
multiple types of data. The coordinate system that was used in this project was
the North American Datum of 1983 UTM Zone 17N. The county boundaries of
Pennsylvania were obtained from the Pennsylvania Spatial Data Access
(PASDA) website and were used for a map projection to locate Toby Creek
watershed. A feature class was created to contain the vector data of point layers
and stream layers. The outline polygon of the drainage basin was from Clarion
University’s M drive data sets and layers. Raster datasets of different stream
buffers were created using different tools in ArcMap 9.3. The streams that were
within the drainage basin boundaries were extracted using the Extract analysis
tool. Spatial Analyst Tools such as the IDW interpolation were used as ways to
interpolate the stream data into a specific number of classes. Multiple methods
were attempted to find an accurate and correct way to display and interpret the
data. The IDW (Inverse Distance Weighted) interpolation method estimates
the raster cell values based on the water quality data by averaging the values of
data within a neighborhood of cell values.
pH variations in stream tributaries
Models were developed to classify overall stream quality to prioritize the
remediation of the basin tributaries
IDW interpolation Method of the stream’s pH values for monitoring
water quality represented in 3 classes (left) and 32 classes (right)
Total dissolved iron variation
Total Iron concentrations are interpreted in three classes. Concentrations of less
than or equal to 9 ppm of Fe are considered low, between 10 and 25 ppm are
moderate, and greater than 25 ppm are considered high
Land use classification
Variations in stream chemistry are closely
related to land uses (clearly visible on
various RS imagery) which have changed
locally over time
An unsupervised land cover classification
map was created using ERDAS as a way to
show the strip mined areas in relation to the
stream and the collection sites