The Kessler Syndrome

As Discussed by Donald J. Kessler

March 8, 2009
The “Kessler Syndrome” is an orbital debris term that has become popular outside the professional orbital debris community without ever having a strict definition. The purpose of this writing is to clarify the intended definition, to put the implications into perspective after 30 years of research by the international scientific community, and to discuss what it may mean to future space operations.

Historical Background

As far as I am aware, the term originated with a colleague, John Gabbard, who worked for NORAD.  NORAD maintained a catalogue of man-made objects in orbit, but did not maintain a breakup record of events in orbit. John unofficially kept a record of major satellite breakup events, which later proved very useful in understanding the sources of smaller orbital debris.  John is known for his description of these events with a graph we now call a “Gabbard Plot”.

When I met John in 1978, I had just published the Journal of Geophysical Research (JGR) paper, “Collision Frequency of Artificial Satellites:  The Creation of a Debris Belt”.  This paper predicted that around the year 2000 the population of catalogued debris in orbit around the Earth would become so dense that catalogued objects would begin breaking up as a result of random collisions with other catalogued objects and become an important source of future debris.  These finding were important for three reasons:

  1. At the time, it was generally assumed that there were very few objects in orbit that were too small to catalogue, although there was no definition as to what limiting size was in the catalogue.  The paper illustrated that even if this assumption were correct, future collisions between catalogued objects would produce a large amount of small debris fragments.  This small debris population would be more hazardous to other spacecraft than the natural meteoroid environment immediately after the first collision.
  2. Each collision would also produce several hundred objects large enough to catalogue, increasing the rate that future collision breakups would occur….resulting in an exponential growth in the collision rate and debris population.
  3. The only way to prevent this exponential growth was to reduce the number of rocket bodies and non-operational spacecraft left in orbit after their useful lifetime.

It was the second prediction that caught John Gabbard’s attention.  While talking to a reporter shortly after the publication of the JGR paper, John used the phrase “Kessler Syndrome” to summarize my prediction of a future cascading of collisions in orbit.  The reporter published the phrase.  Perhaps it was a 1982 Popular Science article that made the term more popular, since the Aviation and Space Writers Association gave the author, Jim Schefter, the 1982 National Journalism Award for the article.  However, regardless of the source, the label stuck, becoming part of the storyline in some science fiction, and a three-word summary describing orbital debris issues.

However, not all who have used the phrase have referred to it in the context of its original meaning.  It was never intended to mean that the cascading would occur over a period of time as short as days or months.  Nor was it a prediction that the current environment was above some critical threshold…although the concept of a critical threshold was an important possibility that was studied in detail more than 10 years later.  The “Kessler Syndrome” was meant to describe the phenomenon that random collisions between objects large enough to catalogue would produce a hazard to spacecraft from small debris that is greater than the natural meteoroid environment.  In addition, because the random collision frequency is non-linear with debris accumulation rates, the phenomenon will eventually become the most important long-term source of debris, unless the accumulation rate of larger, non-operational objects (e.g., non-operational payloads and upper stage rocket bodies) in Earth orbit were significantly reduced.  Based on past accumulation rates, the 1978 publication predicted that random collision would become an important debris source around the year 2000, with the rate of random collisions increasing rapidly after that, if the accumulation rate were not reduced to near zero.

Findings Since 1978

Combined with the discovery that 42% of the catalogued objects were the results of only 19 explosions in orbit of U.S. upper stage rockets and that NORAD was not tracking “all man-made objects” as generally believed, NASA took these findings and predictions seriously.  Beginning in October of 1979, I was given funds to begin research for data to more accurately define the current and future debris hazard, and understand techniques to limit the future growth in the debris population.  With these funds, we accomplished our objectives with a combination of modeling, measurements that sampled the environment, ground tests to simulate space collisions, and coordination with the space community to determine cost-effective techniques to minimize future growth of the debris population.

We sampled the small debris environment by developing and using ground telescopes and powerful, shorter wavelength radars.  We also analyzed recovered spacecraft surfaces for impacts using scanning electronic microscopes, which allowed us to determine the chemistry of the objects causing those impacts.   Together with the Air Force, we conducted hypervelocity ground simulation of collisions and examined ground explosion data to more accurately predict the amount of small debris generated.  We also developed much more elaborate computer models which we used to test our assumptions and ground data against the data we obtained by sampling the environment.  We used these computer models to test the effectiveness of various techniques to minimize future growth in the debris population. These efforts were lead by a team of scientists in what is now known as the NASA Orbital Debris Program Office. Other international governmental agencies participated in this research, forming an international organization now known as the Inter-Agency Space Debris Coordination Committee (IADC).   The following conclusions were reached as a result of this research:
  1. The hazard from the debris that was too small to catalogue had already exceeded the hazard from the natural meteoroid environment.  The sources of that debris included not only explosions, but paint flecks from spacecraft surfaces, exhaust from solid rocket upper stages, and leaks of coolant from nuclear reactors.
  2. Better data and more accurate modeling by NASA and the international community support the conclusion that the long-term threat to the environment is collision cascading, as predicted in 1978.
  3. Modeling results supported by data from USAF tests, as well as by a number of independent scientists, have concluded that the current debris environment is “unstable”, or above a critical threshold, such that any attempt to achieve a growth-free small debris environment by eliminating sources of past debris will likely fail because fragments from future collisions will be generated faster than atmospheric drag will remove them.
  4. Although the rate of growth in the catalogued population has been reduced as a result of new operational procedures that minimize the possibility of explosions in orbits and leaving non-operational upper stages and payload in orbit for periods longer than 25 years, the catalogued population continues to increase, but at a lower rate than it was increasing prior to the 1978 paper.
Significance of the “Kessler Syndrome” Today

On February 10, 2009 the Iridium 33 and Cosmos 2251 satellites collided with a velocity of ll.6 km/sec, at an altitude of 790 km.  The collision was catastrophic, likely producing hundreds of fragments large enough to catastrophically breakup other satellites, and tens of thousands of fragments large enough to damage other satellites.  This is the first clear example of what was predicted in 1978.  Although there have been three other random collisions between catalogued objects since 1991, none of those were catastrophic.

Although all existing data and analysis support the major conclusions presented in the 1978 JGR paper, there are minor differences.  The most obvious is the difference between the predicted growth rate in the catalogue population of 510 objects per year compared with the actual growth rate, which was less.  There were a number of conditions that contributed to the lower rate:  1. The success of the orbital debris program in establishing international agreements that reduced the number of accidental explosions in orbit. These explosions had been a major source of catalogued debris.  2. An abnormally high solar activity increased the upper atmospheric density and caused more satellites to reenter.   3.  The declining economy and eventual fall of the USSR significantly reduced the number of Soviet launches.  As a result of these conditions, the actual average growth rate over the last 50 years was about 300 objects per year.  This rate would have been lower, had it not been for the Chinese anti-satellite test in 2007, which produced over 2000 fragments large enough to catalogue.  A rate of 300 objects per year is close to the lower assumed rate in the 1978 JGR paper.  This average growth rate would predict the first collision between catalogued objects to have occurred around the year 2000, and it was assumed to be a catastrophic collision.

The lower growth rate of 320 objects per year in the 1978 paper predicted two collisions by 2009, both catastrophic.  Although the actual number of collisions is too few to be statistically meaningful, they may indicate that the actual collision rate could be higher than predicted, but fewer are catastrophic.  This higher collision rate would be consistent with the uncertainty in spacecraft area subject to collisions, as was noted in 1978.   In 1991 and 2000 publications, the collision area was shown to be about 2.5 times greater than adopted in 1978.   The 2000 publication also concluded that not all cataloged fragments were massive enough to cause a catastrophic collision…this would be especially true if the colliding fragment hit an antenna, stabilizer boom, or solar panel, or if the target were the empty tank of an upper stage. The presences of antennae, solar wings, and stabilizer booms were ignored in 1978, and obviously hitting one of these areas will only transfer a fraction of the impact energy to the entire spacecraft structure, reducing the likelihood of a catastrophic breakup.  Also an impact into the empty fuel tank of an upper rocket stage may not transfer all the impact energy to the rocket body structure….again not causing a catastrophic breakup.  We may have been lucky that only one of the four collisions since 1991 was catastrophic…or it may be that only one out of four of the collisions between catalogued objects will be catastrophic.  The 1978 prediction of collision frequency becomes more consistent with the actual collision frequency by simply assuming that the area used in 1978 is the average catastrophic collision area, which was the intent of the paper.  However, a more accurate understanding of both the non-catastrophic and catastrophic collision frequency is achieved by using data generated since 1978 in more accurate models currently used by the Orbital Debris Program Office.

Despite the absence of random catastrophic collisions, the predicted fluxes of smaller debris in 1990 and beyond in the JGR paper are not too different from what has been measured as a result of the orbital debris program.  Accidental explosions and a few intentional collisions almost certainly contributed to the similarity…. and possibly some non-catastrophic collisions involving an un-catalogued object also contributed.  However, the major contributors were a number of small debris sources that were discovered since 1978.  Even though these sources have produced a debris environment in the past that is about the same as predicted from collisions, past debris sources are fundamentally different from future random collisions between catalogued objects.   The past sources produce debris at a rate that is proportional to the number of objects in orbit, while the future frequency of collisions will produce debris at a rate that is proportional to the square of the number of objects in orbit.  For example, if one were to double the number of upper stages and payloads in orbit, each having a probability that they would explode, then the rate that debris is generated by explosions would also double.  However the rate that debris is generated by collisions between these objects would increase by a factor of four.

The 1978 prediction of a catastrophic collision between catalogued objects of 0.013 per year was based on a catalogue containing 3866 objects; today, the catalogue contains about 13,000 objects, or more than 3 times as many objects.  This gives a collision rate that is more than 10 times what it was just over 30 years ago, or 0.13 per year….which is the same as one catastrophic collision between cataloged objects every 8 years….with the time between collisions rapidly becoming shorter as the catalog continues to grow.  The larger fragments from either explosions or collisions will further accelerate the rate of collisions.

Most of the collisions in the 1978 paper were predicted to take place between 800 km and 1000 km altitude.  That is even truer today.  Not only is this region rapidly growing, certain altitudes contain a high concentrations of satellites, and the inclinations of their orbits are near polar, both conditions increasing the probability that they will collide, and do so with collision velocities that average more than 10 km/sec.

We are entering a new era of debris control….an era that will be dominated by a slowly increasing number of random catastrophic collisions.   These collisions will continue in the 800 km to 1000 km altitude regions, but will eventually spread to other regions.  The control of future debris requires, at a minimum, that we not leave future payloads and rocket bodies in orbit after their useful life and might require that we plan launches to return some objects already in orbit.

These control measures will significantly increase the cost of debris control measures; but if we do not do them, we will increase the cost of future space activities even more.  We might be tempted to put increasing amounts of shielding on all spacecraft to protect them and increase their life, or we might just accept shorter lifetimes for all spacecraft.  However, neither option is acceptable:  More shielding not only increases cost, but it also increases both the frequency of catastrophic collisions and the amount of debris generated when such a collision occurs.  Accepting a shorter lifetime also increases cost, because it means that satellites must be replaced more often….with the failed satellites again increasing the catastrophic collision rate and producing larger amounts of debris.

Aggressive space activities without adequate safeguards could significantly shorten the time between collisions and produce an intolerable hazard to future spacecraft.  Some of the most environmentally dangerous activities in space include large constellations such as those initially proposed by the Strategic Defense Initiative in the mid-1980s, large structures such as those considered in the late-1970s for building solar power stations in Earth orbit, and anti-satellite warfare using systems tested by the USSR, the U.S., and China over the past 30 years.  Such aggressive activities could set up a situation where a single satellite failure could lead to cascading failures of many satellites in a period of time much shorter than years.

As is true for many environmental problems, the control of the orbital debris environment may initially be expensive, but failure to control leads to disaster in the long-term. Catastrophic collisions between catalogued objects in low Earth orbit are now an important environmental issue that will dominate the debris hazard to future spacecraft.