Restoration of Northwood V Pipe

Using Step-Pools and Debris Removal

Final Report
January 2018 to April 2018

Prepared by:
Michigan Restoration Collaborative

Lucy Aley
Lindsey Hayter
Michael Mata
Bradley Sweet

Date Prepared
16 April 2018

Prepared for
Aline Cotel
Rob Sulewski
Christian Casper
Executive Summary

Our restoration project focuses on a damaged pipe in Millers Creek at the corner of Hubbard
Road and Huron Parkway. The pipe discharges high velocity water from the impermeable area
around Northwood V during storm events. The area near the damaged pipe has been subjected to
years of erosion. This has caused the formation of concave banks near the mouth of the pipe.
Additionally, the pipe has several leaks from the bottom, causing the earth underneath the pipe to
erode so that a large portion of pipe protrudes from the earth unsupported.

Our criteria for the restoration project are reducing erosion in the area, maintaining a natural
aesthetic, and ensuring our design can withstand a 100-year storm.

To prepare our design we researched previous restoration projects, took measurements at the
problem site, and created a model to test our preliminary design. We used literature regarding
step-pools in the Midwest to guide our design features and dimensions. Once we chose the
problem site, we took transect measurements and analyzed the sediment at the site. Using these
measurements, as well as the literature, we built a 3D 1:20 scale model and tested it on a flow
table.

Our design features a step-pool system consisting of four step-pools leading from the mouth of
the pipe into a widened channel to Millers Creek. The step-pools will be comprised of heavy,
limestone boulders that are anchored to the bank. The banks will be regraded, and a channel will
be cut to direct the flow from the pools into the main channel of Millers Creek.

The step-pools will be spaced such that the kinetic energy of the water of the pipe is minimized
as it travels into the creek, resulting in a slower flow velocity and less erosion. The regraded
banks will be stabilized and prevent further erosion while improving the natural aesthetic. In
order to maximize the pool’s durability such that it can withstand the 100-year storm conditions,
limestone boulders were specifically chosen. When anchored into the bank, they offer resistance
to erosion and appear natural. These factors all combine to form a durable step-pool system that
efficiently directs discharged stormwater into Millers Creek.
TABLE OF CONTENTS

Executive Summary
1 Introduction .………..…………………………………………………………………………...1
2 Design Criteria ……………………………………………………………………………..3
2.1 Reduce Erosion ……………………………………………………………………..3
2.2 Withstand a 100-Year Storm ……………………………………………………..3
2.3 Improve Aesthetic……………………………………………………………………..4
3 Description of Product and Functions…………………………………………………………...4
3.1 Step-Pool Design ……………………………………………………………………..4
3.2 Step-Pool Function ……………………………………………………………..7
3.3 Remediation of Surrounding Area ……………………………………………..7
3.4 Channel Widening ……………………………………………………………..7
4 Testing Procedures and Results ……………………………………………………………..8
4.1 Step-Pool Model and Testing ……………………………………………………..8
4.2 Limitations and Error for Step-Pool Model Testing ……………………………10
4.3 Sediment Analysis of Project Site……………………………………………………10
5 Criteria ……………………………………………………………………………………11
5.1 Reduce Erosion ……………………………………………………………………11
5.2 Design for a 100-Year Storm ……………………………………………………12
5.3 Improve Aesthetic……………………………………………………………………12
5.4 Green Engineering Principles ……………………………………………………12
5.4.1 Non Hazardous Materials …………………………………………....12
5.4.2 Waste Prevention …………………………………………………....13
5.4.3 Durability …………………………………………………………....13
5.4.4 Design For Necessary Capacity …………………………………………....13
5.4.5 Lack of Material Diversity …………………………………………....13
5.4.6 Integration of Available Energy and Material ……………………………13
6 Additional Benefits ……………………………………………………………………………14
7 Budget ……………………………………………………………………………………14
8 Possible Improvements ……………………………………………………………………14
8.1 Rock Cascade ……………………………………………………………………15
8.2 Pool Curvature ……………………………………………………………………15
9 Conclusion ……………………………………………………………………………………15
9.1 Criteria Met ……………………………………………………………………15

Appendix A: Northwood V Calculations

1 Introduction

Our chosen restoration site is at the corner of Hubbard Road and Huron Parkway (Figure 1).

Figure 1. Map of Millers Creek

Specifically, we are focusing on the broken pipe that carries away stormwater from the
Northwood V housing complex located east of the restoration site (Figure 2).

Figure 2. Plan View of Problem Site

1
The main problem at our chosen location is the high velocity of water discharged from the
damaged pipe. Water that is meant to be discharged into Millers Creek instead leaks into the
ground that supports the pipe causing erosion of the earth that surrounds and supports the pipe
(Figure 3).

Figure 3. Project Area Side View From the Eastern Bank

Figure 3 shows a plan view of the area surrounding the damaged pipe, where the height of the
unsupported section of the pipe from the ground is 3.2 meters, the length of the carved out
culvert is 7.9 meters, and the width of the culvert from bank to bank is 5.9 meters. The erosion
over time has caused the banks surrounding the damaged pipe to form steep cliffs. Any water
that is discharged from the mouth of the pipe is moving at such a high velocity that sediment
between the pipe and Millers Creek has eroded, causing a small island comprised of the sediment
to form near the bank of Millers Creek. Dead trees have fallen into the pit and need to be cleared
to make way for bank stabilization and step-pool implementation. If the pipe is not repaired and
the banks are not stabilized soon, further damage could occur to the pipe and the banks
surrounding the pipe. The erosion is also approaching Hubbard Road and will cause damage to
the road that will be dangerous and expensive to repair if the area is not remediated soon.  
 
We have designed a step-pool system to handle the discharge of water flowing from Northwood 
V to minimize the erosion in the area. Our step-pool design will consist of four step-pools 
beginning at the pipe and leading down to the main channel of Millers Creek. 
 
In this report we begin by discussing our design criteria and detailed features and functions of 
our step-pool design and go on to explain the testing of a scale model of the design in the flow 

2
table; then we explain criteria that is met, extra benefits, budget, possible design improvements, 
and conclusion. 
 
2 Design Criteria 
 
The criteria that we set out to meet were to reduce erosion, design for a 100-year storm, and to 
improve the aesthetics of the area. 
 
2.1 ​Reduce Erosion 
 
First and foremost, we want our design to reduce the erosion within the problem site. Since the 
erosion has caused the pipe to break, the banks to collapse, and habitats to be destroyed, we want 
to try to eliminate erosion as much as possible. Additionally, if erosion is not mitigated, the 
collapsing bank may reach Hubbard Road and cause further damage and safety hazards.  
 
2.2 ​Design for a 100-Year Storm 
 
We want our design to be able to withstand high velocities of water during both yearly and 100
year storms, and we can calculate the maximum expected velocity by looking at the source of the
water coming from the pipe: the Northwood V housing complex. Using Google Maps, we have
measured the Northwood V housing complex dimensions as approximately​ 112,500 square 
meters (Figure 4). 
 

 
Figure 4. Map of Northwood V Housing and Restoration Site

3
With housing and parking lots, we estimated that about 70% of the area (78,750m​2​)​ ​is 
impermeable (Google Maps). Using data from the largest storm in 2017 as a lower bound and a 
100 year storm as an upper bound, we have calculated velocities flowing through the pipe range 
from 0.69 to 4.30 m/s. 
 
2.3 Improve Aesthetic 
 
Our last design goal is to maintain a natural aesthetic of the site so that it fits in with the 
surrounding environment even though it is man made.  

3 Description of Design and Functions

Our solution to the erosion and runoff problems is focused on a step-pool system consisting of
four step-pools made of limestone boulders that will cascade from the pipe down to the bed of
Millers Creek. Before the step-pools are built; however, the damaged region needs to be
remediated by cutting the pipe and regrading the banks. A channel will also be cut to direct the
flow from the step-pools to the main channel of Millers Creek. The upcoming sections will cover
our step-pool design, step-pool function, remediation of the surrounding area, and channel
widening.

3.1 Step-Pool Design

Step pools are a series of small, level retention ponds that direct water over a layer of rocks and
down to the next pool when the retention pond begins to overflow. Our design consists of four
pools that start at the top of the broken pipe and lead into the main channel of Millers Creek.
Figure 5 shows the dimensions of each step-pool.

4
  
Figure 5. Step-Pools Design and Dimensions 
 
Each pool is 2m long and 2.4m wide, taking up a total length of 8m. The height of the first two 
pools are half the height (0.5m) of the bottom two pools (1m) taking up a total height of 3m. 
Additionally, the depth of each pool is 0.45m. These measurements fit the problem site. The 
step-pool dimensions were decided using literature as a guide and then tested with a polystyrene 
scale model which is explained in section 3.1 Step-Pool Model and Testing.

Figure 6 shows a plan view of the step-pools. Each step-pool is filled with imported gravel with a 
d​50​ size around 2.5 cm. We did a sediment analysis on a sample from the site and found that the 
sediment present is too small and does not meet the requirements for the 100-year storm velocity 
of water flowing through the pools. 

5
  
Figure 6. Plan View of Step-Pools and Boulder Size 
 
The walls of each step-pool are constructed out of 3 flat limestone boulders that are generally the 
same size: 0.9m diameter across, 0.4m wide, and 1.1m tall for the top 2 pools and 1.6m tall for 
the bottom pools. Figures 5 and 6 show the position of the boulders forming the pool. The two 
boulders on the sides of the pool are angled away from the bank at a 110° angle to form a 
concave pool. This measurement was based on past steppools from literature. ​The boulders make
up the height of the step-pools and are also embedded 0.6m in the bank. ​Figure 7 shows a side 
view of the step-pools as well as how the boulders are embedded and their approximate heights. 
 

 
Figure 7. Cross-Sectional View of Approximate Boulder Heights 

6
3.2 Step-Pool Function 
 
In our step-pool design water flows out of the pipe, down into a pool, and over a limestone
barrier to move down the slope. This way, turbulence is created, and kinetic energy of the water
is dissipated. The resulting velocity is less and is unable to erode the bank as it would were it to
move directly down the bank. To keep the boulders in place during high velocity flow, the
boulders are embedded in the bank. These boulders act as an anchor in the bank and reduce
possible erosion. Likewise, the gravel at the bottom of the pools is there to mitigate erosion. The
pools have a depth of 0.45m which gives them the ability to act as a habitat for macrobenthos.

3.3 Remediation of Surrounding Area

Before the step-pools are put in place, the broken pipe will be cut, vegetation cleared, and the
bank regraded (Figure 8). The pipe will be cut back to the bank and removed so that there will be
no unsupported pipe.

Figure 8. Remediation of Surrounding Area

This way, the water leaking from the pipe will no longer erode sediment below it. Additionally,
the water flowing from the pipe will flow into the first step-pool. The step-pools will be built
over a gradual slope, so the bank needs to be regraded, stabilized, and revegetated according to
our calculated slope of 0.4.

3.4 Channel Widening

The kinetic energy coming from the pipe of our problem site will be reduced by widening the
connecting channel to Millers Creek. The water discharged through the problem site is currently
redirected to a small channel 1.95 meters wide that is connected to the main channel of Millers
7
Creek. The channel has undercut banks due to erosion from high velocity flooding events. Beside
the channel there is a large circular collection of sediment from the eroded banks of the site with
a diameter of roughly 10.9 meters. This channel will be widened by removing a section of the
sediment build-up, ultimately making the channel around 7m wide. The wider channel will help
slow down the flow of the water and reduce the erosion on the channel’s bank. A plan view of
the widened channel can be seen in Figure 9.

Figure 9. Plan View with Suggested Channel Widening

4 Testing Procedures and Results

We ran two tests for our overall design. We built a model of our step-pool and tested it in the
University of Michigan hydraulics lab. We also conducted a sediment analysis to see if we could
reuse any sediment needed for the step-pool.

4.1 Step-Pool Model and Testing

In order to run quantitative tests on our step-pool design, we constructed a polystyrene scale
model (Figure 10, Figure 11). We scaled down our design to construct our model (1:20). We
chose polystyrene because it is waterproof and relatively cheap.

8
 Figure 10.​ ​Side Plan of Step​-​Pool Model

Figure 11. Front Facing View of Step-Pool Model

After our model was constructed we ran quantitative tests in the University of Michigan
hydraulics lab. The objective of our tests was to make sure the velocity of the water coming out
the last step-pool was slower than the water coming into the first step-pool. We put our step-pool
model on a water table, and used a hose as our water source to imitate the Northwood V pipe. To
measure the velocity of the water coming out the hose, we took a small piece of paper and timed
how long it took the paper to travel a certain distance in the running water. Then we took that
same piece of paper, and placed it at the last step, and timed how long it took to travel a through
the last step-pool.

9
We found that the water velocity did slow down significantly after it reached the bottom of the
step-pool. The measured velocity coming out the hose was around 0.17 m/s, and the measured
velocity of the water coming out of the last pool was around 0.08 m/s. Our design reduced the
velocity of water by more than half of the original speed. Using the 1:20 scale of our model, we
estimate our design will be able to reduce a velocity around 3.4 m/s, 20 times the maximum
velocity of the hose.

4.2 Limitations and Error for Step-Pool Model Testing

There were limitations and possible sources of error from our step-pool testing. Our model was
made of polystyrene, while our actual design consists of gravel and limestone. The two different
materials will have different viscous and friction forces with water which could cause error to
our velocity measurements. Also, our velocity measurements were limited to video recording
how long it took a piece of paper in the water to travel through the last step-pool. We had limited
time in the lab, so our procedures were only repeated two times.

4.3 Sediment Analysis of Project Site

To save resources, we tested a sediment sample to see if we could use the present sediment in the
problem site as the bottom of the step-pool. The beds of each step-pool require a sediment size
with a d​50​ of around 2.4 cm. We collected a sediment sample at the project location and ran a
sieve analysis (Figure 12).

We found the d​50​ particle size to be 0.31 mm, which is too small to handle the velocity of a
100-year storm in the step-pools. This led us to determine that we would need to buy sediment
for the step-pools.

Figure 12. Sieve Analysis of Project Site:

d50 Found to be Smaller Than Required Sediment Size
10
5 Criteria

The main goals of our design are to reduce erosion and repair the pipe with a durable solution
that can withstand a 100-year storm and to improve the aesthetics of the area. We met our goals
of reduced erosion and improving the aesthetics, however, our model was not adequate to test for
the conditions of a 100-year storm.

5.1 Reduce Erosion

We met the main criterion of the restoration project of reduction of erosion around the pipe area.
The eroded pit is expanding away from Millers Creek, and will reach Hubbard Road without
intervention (Figure 13).

Figure 13. Erosion Proximity to Hubbard Road

The step-pool design reduces the velocity and dissipates the kinetic energy of the water from the
pipe. The reduced speed and energy limit the water from causing more severe erosion. We will
also return much of the eroded earth to the pit when we widen the side channel directing the
water to the creek. This will stabilize the bank to prevent further erosion.

11
5.2 Design for a 100-Year Storm

We were not able to simulate 100-year storm conditions in the hydraulics lab with our
polystyrene model step-pool system, thus we are unable to conclude whether or not it can
withstand a 100-year storm. The velocity of the water discharged from the test hose was
approximately 3.4 m/s, which is 0.9 m/s less than the theoretical 100-year storm test velocity
(4.3m/s calculated in Appendix A). However, our model was able to reduce the velocity of the
water by a factor of two, proving it successful at dissipating the kinetic energy of high velocity
water. The high velocity discharge we were able to achieve during test conditions is more akin to
that of a 50-year storm. Although we are confident that our test model would have returned
similar results for 100-year storm conditions, our findings are inconclusive as to whether or not
our design can withstand 100-year storm conditions.

5.3 Improve Aesthetic

We plan to implement all-natural materials in the construction of our step-pool system that will
fit in with the natural aesthetic of the surrounding area. Our step-pool system will consist solely
of gravel, sediment from the island situated at our restoration site, and limestone boulders. By
using these natural materials, we will be able to effectively dissipate the kinetic energy of the
discharged water while maintaining a natural look to our design.

5.4 Green Engineering Principles

In our design, we incorporated the Green Engineering Principles of using non hazardous
materials, preventing future waste, designing for durability and necessary capacity, and
integrating available material and energy.

5.4.1 Non Hazardous Materials

Principle 1 of Green Engineering states, “Designers need to strive to ensure that all material and
energy inputs and outputs are as inherently nonhazardous (sic) as possible,” (Anastas, 96), and
this principle was satisfied by our design. The main material that we will be using is limestone. It
is a naturally-occuring substance that will not have negative effects on the environment. The
other material that we will be using is sediment to fill in the pit. We will use sediment that is
similar to the sediment already in the area, most likely sediment that we remove when we widen
the channel.

12
5.4.2 Waste Prevention

Principle 2 of Green Engineering states, “It is better to prevent waste than to treat or clean up
waste after it is formed,” (Anastas, 96), and this principle was satisfied by our design. Initially,
we will do some clean up by removing the pipe and relocating some debris from the channel, but
overall we are preventing waste. Left unremediated, the eroded area would continue to expand,
eventually reaching and undercutting the Huron Parkway. By implementing a restoration project
now, we are avoiding a much lengthier and more costly restoration that would also include road
repairs, as well as keeping the road safe for the heavy traffic that goes down the parkway on a
daily basis.

5.4.3 Durability

Principle 7 of green engineering states, “Targeted durability, not immortality, should be a design
goal,” (Anastas, 96), and this principle was satisfied by our design. Our design uses a 100-year
storm event as the maximum amount of water that it would be able to handle, a standard that all
new buildings in Ann Arbor must be built to handle. The limestone boulders will erode with time
and eventually need to be replaced within 100 years.

5.4.4 Design for Necessary Capacity

Principle 7 of Green Engineering states, “Design for unnecessary capacity or capability (e.g.,
“one size fits all”) solutions should be considered a design flaw,” (Anastas, 96), and this
principle was satisfied by our design. Our step-pool system was conceptualized using design
parameters specific to our restoration site, such as the size of the affected area in terms of height,
width, and depth. By customizing our step-pool system to operate in conditions specific to the
restoration site, we have maximized its efficiency.

5.4.5 Integration of Available Energy and Material

Principle 10 of Green Engineering states, “Design of products, processes, and systems must
include integration and interconnectivity with available energy and materials flows,” (Anastas,
96), and this principle was satisfied by our design. Our design uses the pipe from Northwood V
that is already in place. Removing the entire pipe and replacing it would be a much bigger
project that would include tearing up roads and more earth, and then installing a new pipe. We
will remove the damaged section of the pipe, but leave the pipe that is still embedded in earth in

13
place. Our step-pool system was designed using the velocity of water coming from the
cross-sectional area of the pipe, so the step-pools will be integrated with the pipe that is in place.

6 Additional Benefits

Our step-pool design will act as a habitat for macrobenthos. Since the pools have a depth of
0.45m, stormwater will collect in the pools and offer a habitat for macrobenthos along the banks
and limestones.

7 Budget

The raw materials used in our step-pool system will cost approximately $2160-$3180. Since our
step-pool system is only comprised of limestone boulders and sediment from the area
surrounding the restoration site, our main costs will be due to purchasing the boulders and
renting the construction equipment and labor necessary to build our step-pool system. We did not
choose to focus on stabilizing the banks; however, bank stabilization techniques and labor would
add to the costs. Our estimated costs are based on using 12 limestone blocks, 1 ton of gravel for
each of the 4 pools, and sediment that we remove from widening the channel (Table 1).

Materials Estimated Cost

Limestone (12 boulders) $2000-3000

Gravel (4 tons) $160-180

Sediment No additional cost

Total $2160-3180
Table 1. Budget for Materials

We were not able to reach any restoration contractors or the city of Ann Arbor for estimates on
construction costs.

8 Possible Design Improvements

There are several design modifications that we could apply and test to our existing model that
could improve the overall function of our design. Two that would have the fewest design
trade-offs would be the implementation of a rock cascade and a meandering path for the
step-pool system.

14
8.1 Rock Cascade

A rock cascade below the mouth of Northwood V pipe leading into our step-pool could be a
possible modification. This rock cascade could slow down water coming out the pipe before it
even enters the step-pool, allowing the design to handle faster velocities of water. This
improvement would make our design more effective because it would further dissipate the
kinetic energy of the water discharged from the pipe, lessening the possibility of erosion.
However, it would require many more calculations and lead to a more complicated
implementation. A cascade may also increase our material diversity, which would be a trade-off
with Green Engineering Principle 9.

8.2 Pool Curvature

Another possible design improvement we could make would be adding curvature to our
step-pools. Our current design has a straight path for incoming water to travel down, but if we
could find the right curvature where it would direct all the water into the side channel, leading
into Millers Creek. This modification could possibly reduce future erosion by directing all the
water into an existing channel. This modification also gives our solution a more natural look by
creating a step-pool system that meanders much like a natural body of water. If the curvature is
miscalculated, however, it could lead to an increase in erosion instead of a decrease.

9 Conclusion

Our step-pool system for the broken pipe restoration site at the intersection of Hubbard Road and
Huron Parkway effectively dissipates the kinetic energy of the storm water discharged from the
pipe. Our design, which was designed to withstand 100-year storm conditions but has not yet
reached that goal, consists of four step-pools, with the two step-pools closest to the mouth of the
pipe being half the height of the lower two step-pools.

9.1 Criteria Met

We designed our step-pool system such that most of if not all of our design criteria would be met
by our final design. By creating a step-pool system with varying step-heights, we can effectively
direct all storm water, even in the most severe conditions, into Millers Creek. Although our
testing did not confirm whether or not our step-pool system could not withstand a 100-year
storm, it was able to withstand 50-year storm conditions as demonstrated in section 4.2
Limitations and Error for Step-Pool Model Testing. Furthermore, our system has proven to
reduce the kinetic energy of discharged water by a factor of 2, proving that it can also reduce
erosion in the restoration site. Our system will also be comprised of natural materials, including

15
limestone boulders, gravel, and local sediment, enabling our design to have a natural and
aesthetically pleasing look.

16
Appendix A: Northwood V Calculations

Northwood V Runoff Calculations

Approximate Northwood V Area: 375m x 300m= 112,500m​2
Estimate 70% of the area to be impermeable: 78,750m​2

Precipitation
Most rainfall in 2017 over 24 hr, Ann Arbor: 1.62in = 0.041148m
100-Year storm rainfall in 24 hr: 5.11in =0.129m

Volume of rainfall
Yearly storm: 0.041148m * 78,750m​2​ = 3,240m​3
100-Year storm: 0.129*78,750=10158m​3

Table A-1. Storm Velocity Data

Storm Timeframe Discharge Velocity ½ Velocity ⅔ Velocity Full
(hr) (m​3​/s) Pipe (m/s) Pipe (m/s) Pipe (m/s)

Yearly 4 0.22 2.76 2.07 1.38

Yearly 8 0.11 1.38 1.03 0.69

100-Year 4 0.71 8.68 6.51 4.34

100-Year 8 0.35 4.34 3.26 2.17

Formulas for Calculations:

Discharge= volume rain(m​3​)​/timeframe(s)
Velocity= discharge (m​3​/s)/area of pipe (m​2​) *Area of pipe depends on ½, ⅔, or full pipe
*Note we chose the full pipe in a 100 year storm over 4 hours as our upper bound and a full pipe
of a yearly storm at 8 hours as our lower bound.

A-1
 Works Cited

Allen, H. H., & Leech, J. R. (1997). ​Bioengineering for Streambank Erosion Control​. US Army 
Corps of Engineers.

Anastas, P. T., & Zimmerman, J. B. (2003). Design Through the 12 Principles of Green
Engineering. ​American Chemical Society​, 94-101. 
 
Chinn Anne. (2008). Linking Theory and Practice for Restoration of Step-Pool Streams.
​Environmental Management. 
 
Climate Ann Arbor - Michigan and Weather averages Ann Arbor. (n.d.). U.S. Climate Data. 
Retrieved February 17, 2018, from  
https://www.usclimatedata.com/climate/ann-arbor/michigan/united-states/usmi0028/2017 
/3 
 
Cornell University. (n.d.). ​Step-Pool. ​[Diagram]. Retrieved February 21, 2018 from  
http://www.clrp.cornell.edu/nuggets_and_nibbles/articles/2016/stream.html 
 
Cotel, Aline (7 February 2018). Sediment Transport: A Brief Introduction. Lecture at University  
of Michigan, Ann Arbor. 
 
Flores, H., McMonigle, D., & Underwood, K. (2012). Regenerative Step-Pool Storm 
Conveyance (SPSC). Anne Arundel County Maryland. Retrieved from 
https://umich.instructure.com/courses/198276/files/folder/Restoration 
literature?preview=7480883 
 
Halpert, Julie. (June 2015). Storm Over the U-M. The Ann Arbor Observer. Retrieved February  
17, 2018, from 
http://annarborobserver.com/articles/storm_over_the_u-m_full_article.html 
 
National Engineering Handbook. (2007). ​Stone Sizing Criteria​. Retrieved from  
https://umich.instructure.com/courses/198276/files/folder/Restoration  
literature?preview=7480858 
 
Northwood V Housing. (n.d.). Google Maps. Retrieved February 17, 2018, from 
https://www.google.com/maps/place/Northwood+V+Housing/@42.2935654,-83.704051 
6,1224m/data=!3m1!1e3!4m8!1m2!2m1!1sgoogle+maps+northwood+5!3m4!1s0x883ca
e 
98957e1615:0x457128cc4c54ff7f!8m2!3d42.2935655!4d-83.6996745