Nine Mile Run Stream Assessment Emily Broich

1
Nine Mile Run Stream AssessmentEmily Broich
Michael MuderUniversity of Pittsburgh,
Department of Geology and Planetary Science
Nine Mile Run Stream AssessmentEmily Broich
Michael MuderUniversity of Pittsburgh,
Department of Geology and Planetary Science
Abstract Human alterations to the Nine Mile Run
watershed in Pittsburgh, have resulted in deeply
incised channels and changes in the longitudinal
profile. These alterations include dumping of
molten slag from local steel foundries and
concrete channelization of much of the stream.
In addition, much of the streams headwaters were
diverted underground to allow for dense urban
development in the neighborhood of Regent Square
and the boroughs of Wilkinsburg and Edgewood.. A
GIS-based analysis of Nine Mile Run was conducted
in winter of 2007. Recent topographic maps,
aerial photography, and digital elevation models
were compared to historical topographic maps.
Three different investigations were conducted
changes to the streams courses changes in the
streams longitudinal profile and changes to the
streams cross-section. These comparisons were
made before and after the dumping of the slag and
the implementation of the underground concrete
box culvert. Background Steel production in
Pittsburgh began in 1875. Local steel factories
included the Edgar Thompson Works, Homestead
Works, and Jones and Laughlin Steel. A by-product
of the steel making process is slag, formed when
iron ore is melted in a blast furnace with
limestone and coke. The iron ore is melted, and
the slag contains phosphorous and sulfate
impurities, in addition to oxides of aluminum,
calcium, and magnesium, all removed from liquid
iron. For every ton of steel produced,
approximately a quarter ton of slag is produced.
Slag was dumped for approximately 70 years during
the presence of the steel industry in Pittsburgh.
It was dumped over an area encompassing 238
acres, extending from the Squirrel Hill Tunnel of
Interstate 376 to the confluence of Nine Mile Run
with the Monongahela River. In September, 1922,
Duquesne Slag Products Co. of Pittsburgh
purchased 94 acres of land in Nine Mile Run and
began dumping slag that same year. The company
dominated the valley for 70 years to dispose of
slag from the three steel mills in Pittsburgh,
even constructing rail lines to facilitate
disposal.




Source
http//slaggarden.cfa.cmu.edu/
In addition to the slag, fill from the
construction of the right of way for Interstate
376, as well as fill from the Squirrel Hill
Tunnel construction, was dumped in Nine Mile Run.
Slag ceased to be dumped in Nine Mile Run in 1972
after approximately 17 million cubic yards of
waste had been deposited.




Source http//slaggarden.cfa.cmu.edu/
In 1931, Nine Mile Run was diverted through a
concrete box culvert for a portion of the
streams length. In 1995, slag pile was purchased
by the Pittsburgh Urban Redevelopment Authority
for 3.8 million. A major stream restoration was
completed in July 2006 by the US Army Corps of
Engineers. The restoration included stream
channel reconfiguration, wetland reconstruction,
native wildlife habitat enhancement, and native
tree, shrub, and wildflower plantings.
Study Reach The study encompasses a 1.8 mile
stretch of Nine Mile Run, from its confluence
with the Monongahela River to an outfall from a
large underground concrete culvert near Braddock
Avenue, and all tributaries and floodplains
within the study reach. Methodology The asse
ssment of Nine Mile Run was divided into three
sections 1) Analysis of changes in stream
location throughout history 2) Preparation of a
historic and modern longitudinal profile for the
main stream channel 3) Preparation of multiple
stream valley cross-sections, both historical and
present, in order to portray the incorporation of
an immense volume of slag waste into the
watershed. In the analysis, a number of
historical maps were acquired and studied.
The oldest map of Nine Mile Run was the 1872 sur
vey by the G.M. Hopkins Company. This map is a
hand-colored lithograph depicting the 22nd Ward
of Pittsburgh, and is available on the David
Rumsey Historical Map Collection
(www.davidrumsey.com). This map lacked a
georeference, or a set of coordinates to relate
it to its actual position on the earths surface.
This problem was resolved by using intersections
of major roads on the historic map that likely
havent changed in over 125 years. Examples of
such roads include Forbes Avenue, Fifth Avenue,
Shady Avenue, and Penn Avenue. Using ArcMap 9.2,
control points were created that link these
intersections on the historic map to the same
locations on a modern map. The modern map used
was the most recent US Geological Survey map of
the Pittsburgh East Quadrangle, revised in 1996.
The second oldest map is a topographic map
produced in 1907 by the US Geological Survey.
This map is one of a series of fifteen-minute
quadrangle maps that were surveyed between 1903
and 1904. This map was acquired from the
Pennsylvania Spatial Data Access website, and was
already georeferenced. The third oldest map in
chronologic order was created in 1927. This map
is a topographic map prepared by the City of
Pittsburgh Department of Planning. The master
image contains 171 rectangular plates of equal
size. The maps within the study reach were
acquired on
CD-ROM thanks to Ed Galloway of the University of
Pittsburghs Digital Research Library. The plates
did not have a georeference. However, X, Y
coordinates were printed at the four corners of
each plate in degrees, minutes, seconds. This
allowed for direct input of the coordinates of
each map. The most recent map used in the
analysis was the US Geological Survey topographic
map of 1996. Unlike the other map sets, this map
reflects changes to the watershed after the
dumping of slag and channelization of the stream.
This map was downloaded from the PASDA website,
and was already georeferenced.
How has the stream changed? For the first stage
of analysis, the path of Nine Mile Run on each
map was traced. This polyline was saved as a
shapefile. The numerous lines were combined to
reflect observable changes in the streams path
over the course of time. Below is the resulting
map. The second map provides a more detailed view
of the slag dumping ground.


Basemap USGS 1996 Pittsburgh East 24k Topo
After the slag dumping, the anthropogenic effect
s in the valley are quite evident. The pink line,
corresponding to the 1876 stream, and the green
line, corresponding to the 1907 stream, extend
further northwest from the current stream course.
Slag dumping started in 1922, and the stream path
shifted southeast shift and meandered less. There
was still enough width in the valley bottom for a
small pond to exist, even though slag started
being disposed of five years earlier. However,
the 2000 stream path (prior to restoration)
reflected the straightest stream path and the
least amount of meandering. This change in the
stream resulted from the dumping of slag on
either side of the valley. The 1907 stream path
deviates from the other stream paths where they
empty into the Monongahela River. This change
might be explained either by the construction of
a sea wall after the 1907 survey, or an unknown
coordinate/georeference offset or error. The
headwaters of Nine Mile Run become shorter and
less complex over time due to stream burial and
storm drainage effects. The location of the
outfall of this major culvert is near the
junction of Braddock Avenue and Interstate 376.
It can be seen on the map where the dark blue
(2000) stream ends, but the previous stream
courses in green (1907) and light blue (1927)
continue. How has the stream changed? The seco
nd stage of the analysis called for the
comparison of Nine Mile Runs longitudinal
profiles, both past and present. The 1927
Pittsburgh Department of Planning map and the
most recent digital elevation model of 1996 were
used in this study. The elevation contour lines
were used directly from the 1927 map. The
distance between each elevation increment was
complied using the distance measurement tool of
ArcMap. Elevation was plotted against the
distance from the mouth of the stream where it
meets the Monongahela River. The raster
resolution of the 1996 map was not sufficient for
this process. Instead, a digital elevation model
of the study area was downloaded from the PASDA
website, and was converted to elevation contour
lines using the Spatial Analyst extension of
ArcMap. The two longitudinal profiles were
plotted on the same graph to compare changes in
stream slope. The following chart provides dire
ct comparison for Nine Mile Runs longitudinal
profile as a result of anthropogenic effects in
the watershed.
Before the incor
poration of slag into the valley, the slope of
Nine Mile Runs longitudinal profile was
relatively consistent from its headwaters to the
Monongahela River. After the slag was dumped, the
slope became non-uniform. As a result of the
slag, the lower portion of the stream rose in
elevation. This likely affected the slope
characteristics upstream, resulting in sediment
evacuation.. This steep slope occurs near the
effluent of the underground culvert. It is also
likely that bank and bed erosion from high
discharge events racing out of the underground
tunnel resulted in deep incision in this area.
How much slag was dumped? In the third stage of
the analysis, cross sections of the stream
valley were created using the 1927 topographic
map and the 1996 topographic map. This process
was similar to that of the longitudinal profiles.
Four reference lines were created across the
valley, perpendicular to the stream. They were
strategically placed in areas that would give the
best representation of the change in topography
from the slag. These lines were saved as
shapefiles with coordinates so that they could be
referred to in the future. The lines were labeled
1 through 4. Each end of the line was assigned a
letter to indicate measurement direction. The A
node was the northwest end of each reference
line, and the B node was the southeast corner of
each line. All cross-section measurements took
place from A to B. The following graphic shows
the four cross-section reference lines over the
1996 USGS topographic map.
The contour elevations of the map were
plotted against the cumulative distance from
point A to point B to obtain the cross sections.
Four graphs were produced, with two separate
cross sections contained within for comparison of
topography changes in the Nine Mile Run valley.

The graphs clearly depict dramatic topography
changes from the incorporation of slag and tunnel
fill in the Nine Mile Run valley. In Cross
Section 1, the slag plateau seems to have pushed
the stream location slightly to the east. The
southeastern slope of Cross Section 2 appears to
be the only slope with no net change. Cross
Section 3 demonstrates that slag was not only
dumped along the northwestern bank of the stream
but also along the southeast bank, further
upstream. Cross Sections 3 and 4 show the most
amount of fill in volume. 4 displays a very flat
plateau feature. The southeastern end of this
cross section has experienced erosion between
1927 and 1996, the only negative net change over
time. The streams appear to be very incised in
the 1996 measurements, with the most extreme case
being Cross Section 3, where slag was dumped on
either side of the stream. The cross sections
provide a unique and powerful visualization of
the topography changes in Nine Mile Run.
Chemical Analysis of Nine Mile Run
Upcoming Work In the
month of April, we will be traveling to the
stream to install bank pins and scour chains.
Bank pins will allow us to study the magnitude of
stream bank erosion and deposition resulting from
high discharge events. Similarly, scour chains
tell the researcher of the amount and magnitude
of stream bed scour and deposition during high
discharge events. These instruments will be
installed and measured at a regular interval
(i.e. once a week). Erosion/scour data will be
recorded and plotted against precipitation
records and USGS hydraulic flow data to better
understand stream behavior during high intensity
flow. Acknowledgements We would first and fo
remost like to offer our thanks to Dr. Daniel
Bain of the University of Pittsburgh Department
of Geology for guiding us through the research
process. We would also like to thank Ed Galloway
of Pitts Digital Research Library for providing
us for the beautiful 1927 topographic maps. We
also offer our thanks to Mark Collins of the
Environmental Studies program at the University
of Pittsburgh, William Harbert of Pitts Geology
Department and GIS program, Marijke Hecht and
Jeff Bergman of the Nine Mile Run Watershed
Association, and Mary Kostalos of Chatham
University.
Dissolved Oxygen and Temperature. Warmer water
holds less dissolved oxygen and this inverse
relationship holds true at the Commercial St.
site. As seen in the plotted data, as water
temperature increases, amount of dissolved oxygen
decreases.
Total inorganic nitrogen content from NO3, NO2,
and NH3 plotted both upstream and downstream of
restoration efforts. After restoration in 2005, a
sharp spike in nitrogen below the outfall site is
simultaneous with a decrease in nitrogen
downstream at Commercial St. Is this decrease in
nitrogen a result of restoration working to
remove nitrogen from the system?
Below Outfall Dissolved Oxygen and Phosphate. The
data exhibits an inverse relationship between
dissolved oxygen content and phosphate. As more
phosphate is entering the system, this could be
contributing to excessive aquatic plant growth
and decay. A decreased amount of dissolved oxygen
results. This process of eutrophication is taking
place at three culvert discharge sites, before
the restored area of the stream.