1

Forofficeuseonly
T1 ________________
T2 ________________
T3 ________________
T4 ________________

TeamControlNumber

53461

ProblemChosen

Forofficeuseonly
F1________________
F2________________
F3________________
F4________________

2016
MCM/ICM
SummarySheet
(Yourteam'ssummaryshouldbeincludedasthefirstpageofyourelectronicsubmission.)
Typeasummaryofyourresultsonthispage.Donotinclude
thenameofyourschool,advisor,orteammembersonthispage.

Space debris is a growing concern for space agencies. Orbital debris

causes
expensive damage to spacecraft in low earth orbit (LEO). It is the aim of this
model to forecast the population of existing space debris, the cost to repair
damage to the International Space Station (ISS) that is caused by space
debris, and also model three debris removal methods and their cost savings
to the ISS. This model uses a system of differential equations based on
various probabilistic parameters to determine each outcome. This type of
model was chosen because the nature of space debris collisions is variable
and difficult to accurately predict on such large time scales. The model
finds that there are economically attractive removal methods in the near
future. The cost of taking no action will only grow as space debris continues
to accumulate in LEO.

Team#5346

Page2of19

Non-Deterministic Modeling of Low Earth Orbital Debris

Population in Conjunction with Cost Analysis for the
International Space Station
Team 53461
February 1st, 2016

Introduction
The concern over space junk has been steadily increasing. It is estimated
that over 500,000 pieces of space debris exist and are being tracked. Space
junk can take many forms. It can be pieces as small as tiny flecks of paint up
to large debris like entire abandoned satellites. The particles move at an
1
average of over 5-7 kilometers per second in orbit
and are tracked by space
agencies if over a certain size threshold so that spacecraft can effectively
dodge it and avoid a dangerous and costly collision.
We seek to generate several probability models for space debris population
growth over a time scale of 50 years from the present. Ultimately we seek to
compare estimated costs of non-removal (ie: leaving space junk in space), as
well as the costs associated with several removal techniques and their
effectiveness. We would like to determine which method provides us with
the most effective and least costly approach.
For our models, we make several assumptions about the nature of the initial
problem we would like to solve. The first assumption is that the space junk
we consider is only removed through human intervention, only generated
through orbital collisions with other space debris and human interaction,
and generally uniformly distributed around low earth orbit. Through
NASAs orbital debris program office we found that at altitudes over
800km, the time for orbital decay is often measured in decades, and above
1000km the time scale for debris orbital decay is far outside of our time
2
window.
Additionally, NASA states that one third of orbital debris is
created through deliberate human interaction with space objects, and prior
to 2007, the principal source of debris was from explosions of old launch
2
vehicle upper stages left in orbit.
We assume that debris is uniformly

Team#5346

Page3of19

distributed around low earth orbit which is confirmed by NASA, which

2
states that most debris reside within 2,000km of the surface of earth.
Our second assumption was that the majority of new space junk is created
through orbital collisions or randomly distributed collisions caused by
humans such as the Fengyun Chinese Anti-Satellite Missile Test that
3
generated the FENGYUN 1C Debris.
As we can see in Figure (1), the largest
increases in space debris over the last 70 years have been from collisions
initiated by humans and from objects humans left in space.
Our third assumption was that a collision creates, at most, a certain number
of new space junk particles. The largest collision has created at most 5,000
new particles, as evidenced by Figure 1.

(Figure 1)

Our fourth assumption was that the particles in our model are all of
uniform size, or effectively an average of all possible particle sizes (paint
2
fleck to abandoned satellite).
We chose to use uniform particle size because
it lets us compute the result of an average collision of two pieces of space
junk. This is important because small objects can be equally as destructive
as large objects because particle velocity varies greatly in orbit.
The type of model chosen is a continuous system of probability based
differential equations. This was chosen because it afforded a flexibility in

Team#5346

Page4of19

the parameter and variable selection. Most specifically: records of every

piece of space debris, every orbital collision between space debris, and
every collision between space debris and spacecraft such as the
International Space Station (ISS) is not directly available or easily accessed.
However, estimates are available for both probabilities and the populations
associated with those events and parameters, so it made sense to build a
model around them. Additionally, these interactions and events are
certainly not deterministic in nature, they are stochastic processes, ie:
dependent upon chance. The main strength of this type of model is that it
gives us a general idea of the future growth of space debris. Again, because
the processes that grow space debris are seemingly random, creating a
deterministic model to forecast that growth is an overstatement of the
accuracy of the methods used there. The model presented here seeks to
give a probabilistic generalization of space junk growth over time and fits
with past data collection as seen in Figure 1.

Variables & Parameters

-Probability of natural orbital collision in space:
One of the primary concerns of space debris growth is the potential for a
cascade of object collisions resulting in exponential growth of current
debris, i.e Kessler Syndrome. While this concept seems to be sound in logic,
however; we find, in our model, that mathematically the Kessler Syndrome
is incredibly improbable. If we consider the space of low earth orbit (LEO)
to be broken up into evenly distributed bins of 1
the probability of two
objects occupying the same bin is

Team#5346

Page5of19

(Figure 2)

Extending this concept to all objects we obtain our expression for

This expression for

assumes uniform and random movement of all
particles in LEO Space. Although this is a crude approximation for the
probability of a collision in space; most space debris is created through
human intervention.
-Particles created per collision:
Two objects colliding in space can result in a myriad of different outcomes.
One of the hardest parameters to quantitatively predict is exactly how
much debris is created due to a collision. The Fengyun collisions resulted in
over 5,000 new particles distributed throughout space, we use this as our
upper bound for the creation of new particles due to a collision. We then
simply generate a random number,
, from a uniform distribution on the
open interval (0,1), then multiply
by the upper bound of potential
particle creation. This generates

Team#5346

Page6of19

our expression for the number of particles created given a collision occurs.
-Probability of collision due to humans:
The majority of additional space debris is contributed through human
intervention, as a result we include an additional parameter to account for
human accidents. Based on Figure 1 (NASA) the largest magnitude
increases in space debris occur because of intentional and unintentional
manmade collisions. We have approximated, based on this NASA
information, that there have been five such collisions between 1990 and
2010. This gives us our probability of a human interaction collision to be .25
each year. Using this parameter in our most basic model we obtain growth
of space debris that is qualitatively similar to the NASA record shown in
Figure 1. We generate another random number,
, from a uniform
distribution on the open interval (0,1). This gives us to our expression for
.

-
Cost of ISS repair per unit junk:
To approximate the cost associated to space debris growth we only consider
the cost of repairs to the International Space Station(ISS) because it
provides the most tangible expense to the United States government and
people. First, the probability that space debris collides with the ISS once
4
during a six month period has been estimated to be 1/42
at current levels of
space debris. We can then determine that the probability of a collision
during an entire year is 1/21. Additionally, we include a new term to
represent the proportional growth of space debris over time from initial
conditions because collisions become more likely or less likely as space
debris is generated or removed, respectively. We would like to quantify the
maximum amount of potential damage to the ISS that a collision with
debris could inflict. To do this we examined several integral parts of the ISS
but chose the Integrated Equipment Assembly (IEA) which houses the solar
array wing power supply systems that provides power to the space station as
the largest piece of equipment that would need to be entirely replaced. The
5
estimated weight of the IEA, based on Boeing data, is 17,000 pounds.
Additionally, we approximated the marginal cost of sending a pound of
equipment to the ISS from the overall launch cost of one of the most recent
resupply mission to the ISS: the SpaceX Falcon 9 whose total launch cost
6
was $62.1 million and was delivering 3,700 pounds of equipment.
From

Team#5346

Page7of19

there we derived our cost per unit pound of replacement or repair materials
and equipment. For Gamma We take another uniform random number,
,
on an open interval (0,1) multiplied by the maximum amount of damages to
account for and included our new term for debris growth at the end.

Potential Solutions
The ElectroDynamic Debris Eliminator (EDDE):
7
The EDDE is a large electrified ring that is about 3 km long.
A current flows
through the loop, powered by photovoltaic cells, which creates an
electromagnetic field inside of it that reduces the velocity of debris passing
through it so that the debris can fall to Earth. The EDDE moves using the
Earths magnetic field, so it has no need for propellant in space and
therefore has no fuel costs after the initial investment of about $5-10
7
million to build one.
It can bring down a large piece of debris such as an
abandoned satellite in about 10 days or bring down small debris much
7
faster.
-
Proportion of debris removed by EDDE
:
7
The EDDE spacecraft is estimated to remove one large object in 10 days.
It
is estimated that it could remove small objects 1,000 times faster due to
their reduced mass assuming the same density. Therefore, a single EDDE
could remove about 36,000 small objects in a year. The reduction in debris
by EDDE is limited by the surface area of the ring, so as the amount of
debris decreases the EDDE removes less debris per unit time since its
7
chance of meeting a particle decreases.
This is incorporated into the EDDE
model via removing a percentage of the existing space debris at that specific
time. With parameter
then 7.2% of the space debris each year
would be removed. For example it: in the first year it would remove 36,000
of the 500,000 small objects currently in space.

Team#5346

Page8of19

Space Debris Elimination System (SpaDE):

The SpaDE system fires a cylinder of gas from a balloon at an altitude of 80
km above the surface into LEO. This cylinder then slows debris that passes
through it, causing it to fall back to earth. Most of the cost of SpaDE lies in
8
the cost of energy.
This is difficult to predict. The main positive aspect of
SpaDE is that it does not require shooting anything into LEO, therefore the
risk of increasing space debris as well as the risk of being damaged by space
debris is zero.
-
Proportion of debris removed by SpaDE:
The volume of LEO, a cylinder, affected by a single burst of gas from
SpaDE was calculated with the surface area affected by SpaDE in LEO and
the heights of minimum LEO and maximum LEO. The proportion of
objects passing through this volume was estimated simply with the
proportion of total LEO that this cylinder occupies, we take this as an upper
bound to the effectiveness of the cloud. To do this we generate another
random number,
, on the uniform distribution (0,1) and multiply by the
maximum effectiveness of SpaDE. This gives us,
.

Laser Orbital Debris Removal (LODR):

The LODR is a laser broom. Laser brooms fire a laser at an object which
vaporizes part of the object, creating a gas cloud. This gas cloud slows the
object, again forcing it to drop out of orbit. Multiple lasers could remove
9
almost all dangerous small objects, ranging from 1 to 10 cm, within 2 years.
The LODR model utilizes a rate of firing the laser twice per year.
-
Rate of LODR laser broom debris removal:
9
The laser broom can eliminate 2 small pieces of junk every hour.
Unlike
the other methods of removal, it eliminates them at a constant rate
independent of the total quantity of junk. This yields,
.

Reducing the chance of collisions:

Since space is a natural resource shared by every nation, we could agree to
an international moratorium on actions that are known to create large
amounts of debris like blowing up a satellite with a missile. An alternative to
this strict moratorium is that all tests could occur in a specified junkyard
orbit which other spacecraft could easily avoid. We could also make sure

Team#5346

Page9of19

space agencies can track the location of all satellites currently in space, to
ensure that they are not on course to collide with each other. This would
help to avoid accidents like the Iridium-33 and Kosmos-2251 satellite
collision in 2009. This would allow space agencies to maneuver their
spacecraft out of harms way and avoid the creation of thousands of new
pieces of space junk.

Debris Removal Models

Principle model with no removal

(Figure 3)

Here we see how space debris and ISS repair costs will increase over the
next 50 years without any removal methods.

Team#5346

Page10of19

Principle model with EDDE removal

(Figure 4)

Figure 4 shows the results of EDDE removal of space debris, as well as the
the growth of removed space debris, and ISS repair costs.
Principle model with SpaDe removal

Team#5346

Page11of19

(Figure 5)

Figure 5 shows us the effects of the SpaDE removal system on space debris.
A much slower removal of space debris from LEO is noticed, and no
noticeable decline in ISS repair costs from the principle model without
removal.
Principle model with LODR removal

Team#5346

Page12of19

(Figure 6)

Figure 6 shows us how space debris changes as a result of LODR removal

treatment, as well as ISS cost repair analysis predictions.

Discussion
Results
This models test of robustness was a qualitative comparison with
previous NASA data (seen in Figure 1) with initial conditions replicating
those estimated for year 1965 as follows: time interval of 0-40, human
intervention collision probability of .25, and initial space debris population
of 1,000. Figure 1 shows that the model tracks very closely to the known
data from years 1965 to 2010. Based on this accuracy, our model can
confidently predict short term future space debris populations. Based on
the graphs, the EDDE system is the most effective based on our models,
followed by LODR, and SpaDE. Both EDDE and SpaDE remove a fixed
proportion of existing space debris. This is an effective strategy and results
in an effective clean up of LEO. The LODR is more effective than the
EDDE. While the LODR removed debris the fastest and was the most
effective removal strategy based on our model it is highly unadvisable to
implement this in the near future due to its initial cost ambiguity and
potential operational costs. The SpaDE treatment was not effective enough
to consider implementing, even when parameterized to represent firing the
air pulse once an hour. Additionally, its costs were ambiguous due to its
energy consumption. With that much energy required, it may rely on
energy sources that are not renewable or environmentally friendly. The
difference in ISS cost repairs between the EDDE and LODR methods were
nearly indistinguishable: between $500,000 and $1 million. Only the SpaDE

Team#5346

Page13of19

method resulted in no change in ISS repair costs from the principle model
without removal (around $3 Million Dollars).
Limitations of the Model
While our model is accurate in many ways, such as predicting the
current amount of space junk based on the amount in 1965, it has some
limitations as well. The technologies for removal referenced are still
developmental, so their respective costs and effectiveness may change as
they are improved. Our 0.25 chance of a serious collision each year, while
historically accurate at predicting the amount of space junk, is also based on
a small sample size of collisions. Because this actual NASA dataset was
unavailable, a regression analysis was not possible. It is never possible to
exactly predict the future so this model relies on the probabilities that were
laid out for current conditions, however the parameters were calculated
with reasonable considerations for changes in the space debris population.
Future Research
Current cost estimates of the removal methods could change
significantly as the cost of new technologies and production methods will
decrease. Additionally, better parameterization would provide more reliable
methods of analyzing the future of space debris. A final note is that other
potential removal methods exist which were not analyzed in this model,
although this model could be used to do so in the future.

Executive Summary
We love movies about space. We have since our childhoods. Growing up,
we can recall watching movies based in reality like From the Earth to the
Moon (1998), a docudrama mini-series retelling the gripping drama of
space exploration and NASAs Apollo missions, and Apollo 13 (1995), the
Hollywood box office drama depicting an astronaut crews struggle to
return to Earth from space after an accident aboard their spacecraft. We
also remember movies less based in fact, and instead in a galaxy far far
away. In the present, our new space movies are in high definition, the
computer generated imaging is lightyears beyond what we could have
imagined, and we can seamlessly binge watch them on our 4K T.V or
streaming from our laptops connected to the internet wirelessly.
One new movie in particular has been catching a lot of attention because of
its initial premise. Gravity (2013) is a disaster-thriller in which astronauts

Team#5346

Page14of19

are effectively marooned in space while performing extravehicular activity,

the cause: unexpected orbital debris. NASA defines orbital debris as junk
that is circling Earth, and can have many names such as: space junk or
space debris. While the accuracy of Hollywoods Gravity space debris
storm is up for debate, what they did get right was that space debris is a
cause for growing concern. Space junk can take many forms. It can be
pieces as small as tiny flecks of paint up to large debris like entire
abandoned satellites. The particles move at an average of over 22,000 miles
per hour in orbit and are tracked by space agencies if over a certain size
threshold so that spacecraft can effectively dodge it and avoid a dangerous
and costly collision.
As space junk grows, the danger to spacecraft in orbit increases significantly.
Impacts from space junk can be quite costly to the space program. Not only
does it cost money to send new parts to space if an impact damaged
spacecraft, it costs time and money to move our spacecraft out of the way of
oncoming debris. Additionally, the disruption to astronauts aboard the craft
is draining, especially if they are required to enter an escape capsule like the
Soyuz for an expected conjunction, which is an event in which debris has
a probability of passing closely to the craft. We need effective solutions to
this problem. Without removal of the space junk, we estimate that the total
cost in repairs to the ISS is approximately $3.5 million.
In our models we considered several potential solutions. The most
successful solutions were the Laser Orbital Debris Removal (LODR) method
and the ElectroDynamic Debris Eliminator (EDDE). The LODR method
effectively vaporizes pieces of space debris and causes them to fall into the
Earths atmosphere. The initial costs of the LODR method are unknown
and it costs a few thousand dollars to a million dollars to shoot something
down. It obviously is still in development and the total costs are not known,
however it removed all space debris after 50 years according to our model.
The key point here is that an estimate of the initial cost is still unknown,
which makes this the least viable short term option.
The EDDE method is effectively a large electrified ring in space that works
much like an electric bug zapper. It has estimated initial costs from $5-10
million and very low operational costs due to how it navigates space via
earths magnetic field and makes corrections to its orbit without propellant.
This method also removed nearly all space debris within 50 years according

Team#5346

Page15of19

to our model. Because of its low operational costs, and solid initial cost
estimates we believe that the EDDE is the most viable option of the
solutions we considered.
With a space debris removal method implemented that removes all space
debris within 50 years, we expect that the ISS will save approximately $2.5
million in repair costs. This still creates a budget gap of $2.5 million to $7.5
million. This is not a death sentence for the EDDE though. Our model only
estimates costs to the ISS, one of numerous important spacecraft in orbit
around the earth. If our model for the cost to the ISS is indicative of the
relative cost to a large, important space installation then we should expect to
see equally proportional cost savings to other spacecraft in LEO like the
Hubble telescope. We can expect to recoup these costs not just in reduction
of damage to current spacecraft but in reduction of damage to future
spacecraft.

Team#5346

Page16of19

Works Cited
1

Dunbar, Brian. NASA. Accessed January 31, 2016.

http://www.nasa.gov/audience/forstudents/5-8/features/nasa-knows/w
hat-is-orbital-debris-58.html.

"NASA Orbital Debris FAQs." NASA Orbital Debris FAQs. Accessed January
31, 2016.
http://orbitaldebris.jsc.nasa.gov/faqs.html
.

"NORAD Two-Line Element SetsCurrent Data." CelesTrak: Current

NORAD Two-Line Element Sets. Accessed January 31, 2016.
http://celestrak.com/NORAD/elements/
.

United States of America. National Aeronautics and Space Administration.

Office of Inspector General.
NASA.Gov.S
eptember 18, 2014. Accessed
January 31, 2016.
https://oig.nasa.gov/audits/reports/FY14/IG-14-031.pdf
.

United States of America. Boeing. IDS Business Support. I

nternational
Space Station Electric Power System(EPS).
6

NASA. Accessed January 31, 2016.

https://www.nasa.gov/content/liftoff-of-spacex-resupply-mission-to-t
he-international-space-station
.

Pearson, Jerome, Joseph Carroll, and Eugene Levin.

EDDE Spacecraft
Development for Active LEO Debris Removal.Technical paper. IAF, 2014.

Gregory, Daniel, JF Mergen, and Aaron Ridley.

Space Debris

Team#5346

Page17of19

Elimination(SpaDE) Phase I Final Report.Technical paper. University of

Michican.
9

Claude R. Phipps Et. Al, Claude R. "Removing Orbital Debris With Pulsed
Lasers." In
American Institute for Physics.Proceedings.
http://photonicassociates.com/documents/LODR.pdf.

Team#5346

Page18of19

Code Appendix:

Team#5346

Page19of19