The Space Reviewin association with SpaceNews

An Illustration of a recent version of thw DMSP weather satellite, from a family susceptible to in-orbit breakups. (credit: US Ai Force)

Satellite breakups and related events: a quick analysis

Bookmark and Share

This is a brief look at some satellites that have generated pieces of debris in orbit, with the goal of possibly illuminating why they have shed pieces as well as which ones might shed pieces in the future. The data that goes into this analysis is very widely scattered and is hard to correlate, and thus errors tend to creep in. As much of this data as possible comes from official reports or the official US Air Force satellite catalog instead of from secondary sources such as news reports. Please forgive any mistakes in this article and feel free to suggest corrections or identify additional sources of information.

The central question of this article is this: have some of these satellites shed debris, and will others do so in the future? Might we speculate about which ones might shed pieces in the next few years?

These satellites are vulnerable to orbital debris: impacts could cripple them. We use these orbits for many commercial and government satellites, and any additional orbital debris in these orbits could affect a lot of satellites.

This discussion looks at three series of spacecraft in orbit, performing similar missions that use similar designs and fly in similar orbits. They are the Defense Meteorological Satellite Program (DMSP), the National Oceanographic and Atmospheric Administration (NOAA) Polar Orbiting Environmental Satellites, and the Landsat series. This discussion mentions a few more satellites, but will concentrate on a manageable population for now, mainly the DMSP and NOAA satellites, since they both have a history of shedding a few pieces—or sometimes a lot of pieces. The Landsat series will be discussed as needed, since those satellites have not been know to shed debris.

All of these satellites operate in sun-synchronous orbits, which allow them to pass over regions at the same local time, resulting in the same lighting conditions that permit comparisons from day to day. The usefulness of this type of orbit means that there are many satellites in such orbits. The satellites also tend to be large, complex vehicles with solar panels, antennae, and so on. These satellites are vulnerable to orbital debris: impacts could cripple them. We use these orbits for many commercial and government satellites, and any additional orbital debris in these orbits could affect a lot of satellites. Many spacecraft in these orbits have shed pieces and perhaps many more will in the future.

The first DMSP was launched on May 23, 1962, and they came in many models. The last DMSP launched was on April 3, 2014, with one more in storage (The US spent millions on this spacecraft and needs the data that it could provide desperately, but due to bureaucratic inertia and the inability to make a decision, this satellite will probably be moved to a museum and not used. The U.S. will have to live without the data it could provide for several years, until some replacement satellite series can be designed, built, and flown.)

The DMSP spacecraft have come in many models but this discussion will focus on the “5D-1, 5D-2, and 5D-3” models. According to a fascinating history of the DMSP, put out by the National Reconnaissance Office1, the 5D models really should have been called the 6, but they decided that a big model change might give the program too much visibility and might cause the elected leadership to ask for changes! The satellites are listed below by their “F” numbers (numbers that have been reused and could cause confusion unless we only discuss the 5D series). So these begin with F1 and go to F20, but to minimize confusion the table also includes by their satellite number. The comments column of the table provides the launch date and also the number of pieces that have been cataloged from that launch; these are pieces that have been shed by the primary payload.

Table 1: DMSP 5D-1, 5D-2, 5D-3 Spacecraft

F Number Satellite Number Model Launched/Comments
F1 9415 5D-1 Launch Sep. 11, 1976. 6 pieces, some well after launch
F2 10033 5D-1 Launch June 5, 1977. 4 pieces, 2 well after launch
F3 10820 5D-1 Launch May 1, 1978. 2 pieces added well after launch
F4 11389 5D-1 Launch June 6, 1979. 1 piece added after launch
F5 N/A 5D-1 Launch failure July 14, 1980
F6 13736 5D-2 Dec. 21, 1982, mission ended Aug. 24, 1987, some pieces separated on or before Sep. 20, 2015
F7 14506 5D-2 Nov. 18, 1983, mission ended Oct. 17, 1987, 7 pieces cataloged - 3 well after the initial ones
F8 18123 5D-2 June 20, 1987, decommission Aug. 13, 1991. 7 pieces cataloged - 3 well after the initial ones
F9 18822 5D-2 Feb 3, 1988, decommission Feb. 24, 1992. 11 pieces cataloged - 6 well after the others
F10 20978 5D-2 Dec.. 1, 1990, 2 pieces cataloged well after the others
F11 21798 5D-2 Nov. 28, 1991, decommission Aug. 30, 2000. April15, 2004 breakup declared
F12 23233 5D-2 Aug. 29, 1994, decommission Apr il28, 1997, breakup Oct. 24, 2016
F13 23533 5D-2 March 24, 1995, breakup Feb. 3, 2015. Passivated after breakup.
F14 24753 5D-2 April 4, 1997
F15 25991 5D-3 Dec. 12, 1999
F16 28976 5D-3 Oct. 18, 2003
F17 29522 5D-3 Nov. 4, 2006
F18 35951 5D-3 Oct. 18, 2009
F19 39630 5D-3 April 3, 2014. Communication malfunction Feb. 11, 2016
F20 N/A 5D-3 In Storage

Four 5D-1 satellites made it into orbit and all have shed a few pieces, some well after launch. But they are not as “productive” as the 5D-2 satellites have been: the 5D-2 had a suspect Battery Charge Assembly (BCA) that was redesigned for the 5D-3. Two of these 5D-2 debris generation events have been recent and have been categorized as breakups, but even the older ones have had pieces cataloged long after they were launched. DMSP F6 has had 11 pieces cataloged. The satellite was launched on December 21, 1982 and completed its mission on August 24,1987. Probably after mid-September of 2015 it shed six pieces, though it does not appear in the NASA list of satellites2 that broke up. F7 had seven pieces cataloged, three well after launch. F8 had seven cataloged, three well after launch. F9 had 11 pieces cataloged, six well after launch. So some component on those satellites failed.

From the reporting of the F11 and F13 spacecraft, there could be at least two mechanisms for generation of pieces: hydrazine ignition and battery overcharging.

When we see that pieces have been cataloged from these spacecraft we want to understand the process that created them. A good start is to see how long after launch a piece separated. But that is very hard to determine. It is interesting to see the number of objects for each mission, their satellite numbers, and the dates that the objects were added to the satellite catalog. A close look at the satellite catalog makes an analyst wonder if those pieces separated shortly before they were cataloged or maybe they separated and were tracked for a while before they were cataloged. As an analyst, I often waited until satellites separated enough that they could be uniquely identified by the computer system, and then cataloged them. Certainly many of them were not associated with the launch, and thus must have separated after the satellite had been in orbit for years. I have cataloged many space objects and can imagine that the orbital analysts of the time could have easily tracked objects for perhaps weeks before cataloging them. This practice is referred to in NASA’s Orbital Debris Quarterly News, Volume 19, Issue 2, so it appears to still be done.3

Additionally, I have exchanged email with the Air Force’s Joint Space Operations Center (JSpOC) and they gave me dates that objects were added to the satellite catalog that seem to indicate that the satellite number would be far different than it is. When a satellite is cataloged it is usually given the next open sequential number, so you can generally estimate the date that it was added by looking at the previous and next launches. Launch dates are easy to verify. When I was an orbital analyst it was strictly forbidden to reserve a satellite number except under very special circumstances, to avoid having a satellite number that was never used. So very likely we cannot assume that a piece separated shortly before it was cataloged. The satellite number is possibly not a good indicator of when the piece separated.

The decay dates are also interesting. I have carefully watched CubeSats, which are known to often be low density, and they reenter the atmosphere an altitude of about 160 kilometers. Many of these pieces of debris were marked as decayed when their perigee was nearly 400 kilometers. This could indicate that the orbits reported on these objects were not precise.

F6 through 10 appear to qualify as “anomalous events” since they do not appear in the latest NASA list of breakups,2 but that has not been updated since 2008. These satellites seem to have had several episodes of piece shedding—they do not apparently “break up” in a single event but possibly have several events. The Cosmic Background Explorer spacecraft has had similar behavior (see below). Still, we want to be very careful about what information we accept when trying to guess causes for piece separation.

Interestingly, the 5D-3 satellites have not shed pieces, although, of course, they are a lot newer.

F11 shed pieces on April 15, 2004, possibly from a hydrazine ignition, since it had previously been passivated and the batteries had been discharged.5 The F12 satellite has just had one piece from it cataloged—perhaps some part blew off and vented pressure?5 Or perhaps it will shed pieces in the future like F6 through F10. F13 (a semi-retired spacecraft at the time that was sending down telemetry) shed pieces on February 3, 2015 and it had not been passivated.6 Telemetry did indicate a spike in temperature in the electrical subsystem. F14 is a 5D-2 that has not (yet?) shed pieces, although it has only been in orbit since March 4, 1999. It has not yet been passivated. Both F11 and F13 broke up after about 15 years in orbit, so F14 may break up sometime in 2017. From the reporting of the F11 and F13 spacecraft, there could be at least two mechanisms for generation of pieces: hydrazine ignition and battery overcharging.

Interestingly, the 5D-3 satellites have not shed pieces, although, of course, they are a lot newer. Only F15 has been in space long enough to have the potential to shed pieces. If the battery charger assembly was the major problem, these satellites may not shed debris.

NOAA Satellites

The first NOAA polar-orbiting satellite was launched in December 1970. These satellites came in several models, with latest being an Advanced TIROS-N (ATN). (TIROS stands for Television InfraRed Observation Satellite.) NOAA 6 and 7 have shed pieces, but only a few, so NASA refers to these as “anomalous events.” Some pieces are immediately subsequent to the payload, which indicates that they were pieces that were a normal part of the launch process. Many launches have pieces that are designed to separate when the payload and upper stage enter orbit, and those objects normally get satellite numbers very close to the payload’s number. But many pieces have satellite numbers that indicate that they were cataloged often years after the launch. They certainly have not been around since the launch.

The table below lists the NOAA POES spacecraft. The early ones appear to have not shed much debris. One satellite—NOAA B, satellite number 11819—failed to enter the correct orbit and did not stay in space long.

Table 2: NOAA Spacecraft Of Interest

NOAA # Satellite Number Model Launched/Comments
1 4793 TIROS-M Dec. 11, 1970. 1 piece
2 6235 TIROS-M Oct. 15, 1972, no pieces
3 6920 TIROS-M Nov. 6, 1973, no pieces
4 7529 TIROS-M Nov. 15, 1974, no pieces
5 9057 TIROS-M July 29, 1976, no pieces
6 11416 TIROS-N Junw 1979, 3 pieces, at 2 different times? See NASA report page 453
B 11819 TIROS-N May 3, 1980. Failed to enter correct orbit See NASA report page 451 two events years apart?
7 12553 TIROS-N June 23, 1981, 8 pieces, at 3 different times? see NASA report page 457
8 13923 ATN Launch March 30, 1983, breakup Dec. 30, 1985 with 9 new pieces
9 15427 ATN Dec. 12, 1984
10 16969 ATN Sep. 17, 1986, 6 pieces, several events
11 19531 ATN Sep. 24, 1988, 4 pieces, 2 well after launch
12 21263 TIROS-N May 14, 1991, 6 pieces, maybe three events?
13 22739 ATN Aug. 9, 1993, shut down later that year
14 23455 ATN Dec. 30, 1994, 3 pieces, launch and 1 event
15 25338 ATN May 13, 1998
16 26536 ATN Launch Sep. 21, 2000, retired June 2014, shed pieces on Nov. 25, 2015
17 27453 ATN June 24, 2002. Retired.
18 28654 ATN May 20, 2005
19 33591 ATN Feb. 6, 2009

Of particular interest is NOAA 8, which was launched on March 30, 1983. An event happened nine months later, when this satellite suffered a malfunction that caused the battery to overcharge, leading to pieces being shed.1 This could indicate that battery-charging problems also could affect NOAA satellites. NOAA 10, 11, 12, and 14 have each shed a few pieces as well. NOAA 16 was launched in September 2000 and had a “critical anomaly” in June 2014, after which it was retired. It shed pieces on November 25, 2015.7


The first Landsat was launched on July 23, 1972. and these spacecraft have not shed pieces, though some of their upper stages have exploded and generated large numbers of pieces. The most interesting thing about the Landsat series is that they are similar to the other spacecraft and have had plenty of time to generate debris—but have not.

Prospects for the future

When satellites have broken up, some pieces go into higher orbits and some go into lower orbits. The altitude affects their rate of precession, and so the pieces will drift out of their initial orbital plane at different rates and potentially endanger satellites in other orbital planes.

We and our friends the Russians have launched many upper stages into orbit that have broken up, scattering many thousands of pieces into critical orbits. The stages used to launch the Landsat series have contributed greatly to the debris population.

So we are left with a list of large, complex satellites that could potentially continue to scatter thousands of pieces of debris into a very important orbit. It is very likely that DMSP F-14 will shed a large number of pieces at some time in the next few years. From historical records, DMSP F-11, F-12, and F-13 may yet generate more debris, but probably not much. It remains to be seen if the 5D-3 satellites will shed pieces, possibly with the F11 failure mechanism.

Also, NOAA 13 and 15 may shed pieces in the future, probably with the earlier ones doing this first. If all ATN satellites eventually shed pieces, as might be the case, even NOAA 16 through 19 may create debris in the future.

Series spacecraft are not the only ones that break up. The Cosmic Background Explorer satellite launched on November 18, 1989, and in (possibly) January and March of 1993 shed pieces, followed by an August 1993 shedding event. The instruments were turned off on December 23, 1993, and the satellite was decommissioned in January 1994. So it shed pieces in 1993 in possibly several events while it apparently continued to work. The WorldView-2 imaging satellite shed pieces on July 18, 2016, even though it also continued to work.7 It had been in space for 6.7 years and appears to have not suffered a collision.2 After the event it continued to be operational. Look for a future article about satellites like these that shed pieces.

The subject of a future article will have to be fragmenting upper stages, such as the Delta upper stage. We and our friends the Russians have launched many upper stages into orbit that have broken up, scattering many thousands of pieces into critical orbits. The stages used to launch the Landsat series have contributed greatly to the debris population.


As a retired Air Force officer and space guy, the DMSP satellites have appealed to me is that they were an early “blue suit” operation. The satellites were launched by Air Force crews (in large part) from Vandenberg and were controlled from launch by Air Force crews at Offutt Air Force Base, Nebraska. Air Force officers provided on-site engineering support; the contractor certainly provided important sustaining engineering but, unlike many satellites, Air Force personnel had tremendous responsibility for the satellites. And as the first of this type of environmental satellite they have a longer history, with more anomalous events like piece shedding.

The author would like to acknowledge a former colleague, the late John Gabbard, for sparking an interest in breakups, and would like to salute the late Major Larry Ballard, USAF, who has been an inspiration for years.


  1. A History Of The Military Polar Orbiting Meteorological Satellite Program, R. Cargill Hall, Office of the Historian, National Reconnaissance Office, Sep 2001
  2. History Of On-Orbit Satellite Fragmentations, 14th Edition, June 2008. NASA/TM–2008–214779
  3. NASA Orbital Debris Quarterly News Vol 19, Issue 2, Page 1
  4. “Another USAF weather satellite just broke up in orbit” SpaceNews October 2, 2016
  5. “7 More USAF Weather Satellites at Risk of Explosion” SpaceNews July 20, 2015
  6. “NOAA Weather Satellite Breaks Up In Orbit” SpaceNews Nov 27, 2015
  7. NASA Orbital Debris Quarterly News Vol 20, Issue 4