Anne Sexton
Intro to GIS
May 9, 2012
Final Paper
Title:
Weathering the Storm: Analyzing stormwater runoff from the Tufts University Medford campus
Introduction:
Under natural conditions, precipitation either runs off, evapotranspirates or infiltrates into the
soils, eventually reaching groundwater table. In urbanized areas, much of the natural land
surfaces have been replaced by impervious surfaces, resulting in less infiltration and more water
running along the surface and depositing into a nearby water body. As the runoff travels along
the surface, it picks up pollutants, like chemicals, oils, sediment, phosphorus, and nitrogen. More
impervious surface allows for greater accumulation of pollutants, making runoff in urbanized
areas highly contaminated.
The traditional, commonly used stormwater management has been to quickly collect and
transport stormwater off-site to a treatment facility through a system of gutters and pipes. A new
wave of techniques have become more popular called low impact development (LID), which
consists of multiple small-scale stormwater controls aimed at treating and infiltrating stormwater
on-site to help reduce volume of flow, quantity of pollutants, and restore natural hydrology.
There are a wide range of LID practices, including bio-retention, cisterns and rain barrels,
permeable pavement, vegetated swales, and green roofs.
This project stems from a past project conducted by two groups of Tufts students aimed at
improving the management of campus stormwater by utilizing more LID methods. The two
reports published provide an in-depth analysis of volume of stormwater, concentration of
pollutants, and removal rates of suggested LID technologies. This project gives an overview of
the campus drainage network in its current state, allowing for placement of an LID practice in a
areas of high volume of flow accumulation.
Data Source and Projection
Office of Geographic Information (MassGIS), Commonwealth of Massachusetts, Information
Technology Division, NAD 1983 State Plane Massachusetts Mainland FIPS, 2001.
DEM
I used 5-meter elevation data from MassGIS to create a digital elevation model (DEM) of the
campus and surrounding area. The accuracy of the map results from data type, mine being
floating point. Errors in were found in my DEM called sinks; areas surrounded by higher
elevation values and are areas of internal drainage. I used the Sink tool to identify these areas,
and corrected them by using the Fill tool and reprocessing the DEM map.
Slope
The slope of the area is reported in degrees, and it is calculated using the Slope tool and DEM
information. The slope of each cell is calculated as the maximum change in value from that cell
to its neighbor. The lower the slope value, the flatter the terrain; conversely the higher the slope
value, the steeper the terrain. The flattest slopes are shown in green, and the steepest in red.
Flow Direction
This map shows the direction of flow for every cell using cardinal directions. Each direction,
north, northeast, east, southeast, south, southwest, west, and northwest, are shown in different
colors. The direction is calculated from the direction of the steepest point in each cell, which is
why this is produced using the filled DEM information. This shows the trend of direction of
water flow for the campus area, which is useful to find the ideal location for LID stormwater
practices that rely on surface flow.
Flow Accumulation
Flow accumulation shows the total number of cells that flow into each cell by weight using the
values calculated in Flow Direction. Some areas with very high concentrated flow may be
natural stream channels or artificial channels produced by altering land use. This map highlights
the areas of high flow accumulation on campus and outline possible problem areas. These areas
may suffer from erosion and flooding.
Drainage Network
The results of this process are seen in the drainage network map, calculated using the Stream to
Feature tool and Flow Accumulation raster, and then overlaid onto an imagery of the area. I used
the Raster Calculator to create the conditional statement that cells with a flow accumulation
greater than 50 would be labeled as 1, and colored separately from the rest of the cells. With this
map, the cells with 50 or more accumulated flow are shown in yellow. Many of these areas are
impervious surfaces, or areas with no infiltration like roads, parking lots, and sidewalks.
Results
Impending regulation changes for the Tufts campus will soon encourage a reduction in volume
of stormwater runoff and impervious surfaces. Using this map, I chose two areas of high flow
accumulation that could benefit from the implementation of LID stormwater practices and
marked them on all maps. The first area is the open space and road in front of Latin Road
Dormitory, and the second is the Aidekman Parking Lot 1. I included photos of both locations
along with suggestions for LID practices to install.
Reflections
I had trouble with my initial data collected from the National Map viewer of the USGS database.
It was a little heartbreaking to go through all the steps to find the drainage network and then
realize something looked very wrong with all of my maps. Luckily, I took screen shots and had
my problem solved by contacting my professor. Unfortunately, that problem was unforeseeable,
but it might help future students thinking of pursuing a future very local project.
References
Di Luzio, M., Arnold, J. G. and Srinivasan, R. (2005), Effect of GIS data quality on small
watershed stream flow and sediment simulations. Hydrol. Process. 19: 629–650.
doi: 10.1002/hyp.5612
The authors of this paper were focused on simulating the total runoff and sedimentation of a
watershed in Mississippi using GIS, and other software. For them, using the DEM map was
critical to define their watershed and subwatershed boundaries. I had the choice between using
the DEM model and the TIN model, and this article affirmed my decision to use the Digital
Elevation Model once I found my elevation data.
Maidment, David, R. (1996), GIS and Hydrologic Modeling. Presented at The Third
International Conference on GIS and Environmental Modeling. Santa Fe, New Mexico.
This presentation shows the importance of both GIS tools for spatial analysis with hydrologic
systems. He states that GIS can allow for the increase in definition of spatial subunits, allowing
for greater knowledge of the movement of water. He does not use his methods in a way that
could be guidance for this project, but he does speak of the significance of using GIS when
working with water, namely because hydrologic modeling assumes uniform spatial properties.
B.H.P. (Ben) Maathuis, Ir. C.M.M. (Chris) Mannaerts, N. (Nasreen) Islan Khan. (December,
1999). Evaluating Urban Stormwater Drainage Using GIS and RS techniques: A case study in
Dhaka, Bangladesh. GeoCarto International. 14:4: 21-32.
A study based in Bangladesh that uses GIS to map the flow of urban stormwater drainage. This is
a nice example of what, with a more extensive knowledge of GIS, our project could turn out to
look like. They use mainly raster data to map the flow of water from the urban center, as well as
predicting future flow based on urban growth predictions. It is mainly a technical article showing
the calculations used, but the inputs they used are insightful (such as rainfall intensity).