Space debris mitigation in geosynchronous orbit

Available online at www.sciencedirect.com

www.elsevier.com/locate/asr

Advances in Space Research 41 (2008) 1091–1099

Space debris mitigation in geosynchronous orbit

L. Anselmo *, C. Pardini

Space Flight Dynamics Laboratory, Istituto di Scienza e Tecnologie dell’Informazione (ISTI), Consiglio Nazionale delle Ricerche (CNR),

Via G. Moruzzi 1, 56124 Pisa, Italy

Received 18 July 2006; received in revised form 7 December 2006; accepted 12 December 2006

Abstract

In order to preserve the geosynchronous region, the Inter-Agency Space Debris Coordination Committee (IADC) proposed andendorsed a re-orbiting strategy for spacecraft at the end-of-life: they should be disposed above the synchronous altitude and passivated,to reduce the risk of inadvertent explosions. The recommended perigee altitude of the disposal orbit took into account all relevant per-turbations and was a function of the expected perturbing acceleration induced by solar radiation pressure. It was intended to prevent anyfurther interference with a properly defined geostationary protected region.

This paper addresses four main aspects related to space debris mitigation in geosynchronous orbit, by reviewing the rationale andexpected effectiveness of spacecraft end-of-life disposal. First, the role played by the initial eccentricity vector on the trajectory evolutionof disposed satellites. Second, the collision risk posed by debris clouds and the importance of passivation to prevent energetic breakups.Third, the impact of the operational limitations characteristic of aging spacecraft (e.g. reliability of residual propellant estimates, maneu-ver constraints and subsystems performance) on the definition of practicable disposal strategies. Last, the potential problem representedby low energy, non-explosive, fragmentations leading to the release of debris with high area-to-mass ratio. Based on the modeling resultsobtained, some possible mitigation solutions are discussed, including possible enhancements or revisions of the IADC recommendation.� 2006 COSPAR. Published by Elsevier Ltd. All rights reserved.

Keywords: Orbital debris mitigation; Geosynchronous orbits; End-of-life re-orbiting; IADC formula; Eccentricity vector; Geostationary protected region

1. Introduction

Since the dawn of the space age, nearly circular geosyn-chronous orbits have been recognized as an essentialresource for satellite applications. Today, a large fractionof space launches (37.4% since 2001) is bound to the geo-stationary ring, and 346 out of 819 active operational sat-ellites, as of June 19, 2006, corresponding to the 42.3% ofthe total, carry out their mission in this important regionof space (estimations based on the satellite database pro-duced and updated by Grimwood (2006) for the Unionof Concerned Scientists). Considering eccentricities smallerthan 0.1, mean motions between 0.9 and 1.1 revolutions persidereal day, and inclinations lower than 20�, 1089 known

0273-1177/$34 � 2006 COSPAR. Published by Elsevier Ltd. All rights reserv

doi:10.1016/j.asr.2006.12.018

* Corresponding author. Tel.: +39 050 315 2952; fax: +39 050 315 2040.E-mail addresses: [email protected] (L. Anselmo),

[email protected] (C. Pardini).

objects have met such orbital constraints at the beginningof 2006 (Hernandez and Jehn, 2006; Jehn and Klinkrad,2006).

Following the rapid increase in the number of aban-doned spacecraft and apogee kick stages, since the late1970s there was growing concern in the technical commu-nity that this unique orbital regime might become over-crowded. It then became clear that also spacecraft andupper stage breakups contribute to the geosynchronousdebris environment. In order to preserve the synchronousregion, professional associations, international bodies,satellite operators and space agencies developed specificrecommendations and national guidelines. Finally, in1997 the Inter-Agency Space Debris Coordination Com-mittee (IADC) proposed and endorsed a re-orbiting strate-gy for spacecraft at the end-of-life.

The recommended perigee altitude of the disposal orbittook into account all relevant perturbations, and was a

ed.

1092 L. Anselmo, C. Pardini / Advances in Space Research 41 (2008) 1091–1099

function of the expected perturbing acceleration induced bysolar radiation pressure (IADC, 1997). Together with anend-of-life passivation requirement, to reduce the risk ofinadvertent explosions (IADC, 2000), this perigee altitudewas intended to prevent any further interference with aproperly defined geostationary protected region, consistingof a toroid centered on the geostationary orbit, extending200 km above and below this altitude and ±15� in declina-tion (IADC, 2002). While normal operations are usuallyconducted in the so-called geostationary ring, within75 km of the geostationary altitude and ±0.1� in declina-tion, the protected region was extended in altitude(±200 km), to create a maneuver corridor for spacecraftrelocation (plus a margin), and in declination (±15�), totake into account the natural orbit evolution of satelliteswithout inclination control.

According to the IADC recommendation, considered atpresent for adoption also by the International Telecommu-nication Union (ITU) and by the International Organiza-tion for Standardization (ISO), the perigee of the disposalorbit should be higher than the geostationary altitude byan amount DH (km) given by

DH ¼ 235þ Cr � 1000 A=M ð1Þ

where A is the average cross-sectional area (m2) of thesatellite, M is the satellite mass (kg), and Cr is a radiationpressure coefficient, typically between 1 and 2, which spec-ifies the amount of solar radiation transmitted, absorbedand reflected by the spacecraft (IADC, 1997, 2000, 2002).

2. The relevance of the disposal eccentricity

In order to avoid any further interference with the geo-stationary protected region, Eq. (1) – specifically the termdealing with solar radiation pressure – was developed withthe underlying hypothesis of a nearly circular disposalorbit. However, this legacy was lost in the end-of-lifere-orbiting requirement written in the IADC Space Debris

Mitigation Guidelines (IADC, 2002), because severalanalysts had become convinced that disposal eccentricitiesdifferent from zero might in the long-term have ensuredlower spatial densities and collision probabilities in thesuper-synchronous graveyard region (IADC, 2000).

However, subsequent analyses of the stability of super-synchronous graveyard orbits, for example by Martinet al. (2002), but especially by Lewis et al. (2004), con-firmed the validity of the hypothesis underlying Eq. (1).In other words, the success of the re-orbiting defined byEq. (1) in avoiding any further crossing of the geostation-ary protected region depended upon attaining a sufficientlysmall disposal eccentricity (60.005).

Other studies, on the other hand, pointed out the poten-tial of eccentricity vector control. For instance, by pointingthe disposal orbit perigee towards the sun and targeting the‘‘natural’’ eccentricity induced by solar radiation pressure,it would be possible to preserve the protected region witha slightly smaller altitude increase than that found with

Eq. (1) (Delong and Fremeaux, 2005). It was also foundthat even eccentricities as large as 0.05 did not cause anyfurther interference with the geostationary protectedregion, provided a proper orientation of the eccentricityvector was achieved after re-orbiting (Gopinath andGaneshan, 2005).

In order to clarify the matter, an extensive analysis wascarried out at ISTI/CNR in Pisa, Italy, which investigatedin depth the influence of the eccentricity vector magnitudeand direction, in addition to other relevant initial condi-tions and parameters. This activity was an extension ofan international study promoted by IADC, to furtherexplore the evolution and stability of super-synchronousdisposal orbits (Anselmo, 2006).

The orbital propagations were performed with a modi-fied version of a long-term orbit predictor using the methodof the variation of parameters in the formulation of theequations of motion (Kwok, 1986). The perturbations tak-en into account were: the earth’s geopotential harmonics,up to the 8th order and degree, the third body attractionof the moon and the sun, and the solar radiation pressure,including eclipses. To expedite the computations, the termsincluding the mean anomaly (fast variable) were removedin the earth’s and third-body potentials before numericalintegration (singly averaged method). On the other hand,when resonances occurred between the orbital period andthe earth rotation, the terms containing the mean anomalyin the tesseral harmonics of the geopotential were retained,thus giving rise to resonant effects. To average the potentialdue to solar radiation pressure, a standard 8th orderGaussian quadrature method was used. The resulting aver-aged equations were integrated numerically using a multi-step, variable step-size and variable order method. Thecomputation time was maintained under control, withoutcompromising the accuracy, due to the elimination of thefast variable and the possibility of using large step sizes(days).

The simulations, spanning 100 years, focused on the re-orbiting of spacecraft with A/M = 0.02 m2/kg and Cr = 1.2(representative average values for geostationary satellites).They considered three disposal altitudes (DH, DH · 2 andDH · 3), eccentricities up to 0.3, varying perigee angles(a) with respect to the right ascension of the sun, varyingvalues of the right ascension of the ascending node plusthe argument of perigee (X + x) and different values ofthe initial right ascension of the ascending node of themoon (�13� 6 XM 6 13�). The main conclusions of theanalysis were the following (Anselmo, 2006):

� Eq. (1) can prevent the long-term crossing of the geosta-tionary protected region only by constraining the initialeccentricity vector.� An unconstrained argument of perigee translates into a

maximum acceptable eccentricity of �0.005.� Eccentricities as large as 0.3 (or slightly more) are

acceptable with an appropriate perigee pointing drivenby luni-solar perturbations (X + x @ 90� or 270�).

2010 2020 2030 2040 2050 2060 2070 2080 2090 2100200

220

240

260

280

300

320

340

Year

Alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

Perigee towards the sunPerigee at 45˚ from the sunPerigee at 90˚ from the sun

Protected Region

Fig. 1. Perigee altitude evolution, over 100 years, after a simulatedre-orbiting according to Eq. (1) of a spacecraft with A/M = 0.02 m2/kgand Cr = 1.2. The initial mean eccentricity of the disposal orbit (0.005) isable to avoid the long-term interference with the geostationary protectedregion, irrespective of the perigee orientation. An initial sun-pointingperigee, however, would obtain the same result with a re-orbiting altitude(213 km) lower by about 18% compared to the value given by Eq. (1)(259 km).

2010 2020 2030 2040 2050 2060 2070 2080 2090 2100200

300

400

500

600

700

800

900

1000

Year

Per

igee

alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

All perturbationsNo sun & moonNo sunNo moonNo sun, moon, radiation pressure Only J

2

Fig. 2. Impact of different orbit perturbations on the evolution, over 100years, of the perigee altitude of a satellite re-orbited according to Eq. (1).The simulated spacecraft is characterized by A/M = 0.02 m2/kg andCr = 1.2. The initial disposal eccentricity is 0.1 and the perigee is orientedsuch that X + x = 270�.

L. Anselmo, C. Pardini / Advances in Space Research 41 (2008) 1091–1099 1093

� The details of the evolution depend on initial state vec-tor, right ascension of the sun, right ascension of themoon ascending node XM and CrA/M.� Increasing the re-orbiting altitude (DH) is not a viable

alternative to eccentricity vector control.� For small eccentricities, the choice of a sun-pointing

perigee may result in a smaller altitude increase, withrespect to Eq. (1), which is able to preserve the geosta-tionary protected region with a velocity variation(DV), or propellant saving, of �15%.� For eccentricities >0.005, optimal solutions, in terms of

maximum long-term separation from the geostationaryprotected region, can generally be obtained by combin-ing the proper choice of X + x (�90� or �270�) with asun-pointing perigee and XM � 0�. Unfortunately, forthe assigned values of X + x, a sun-pointing perigee ispossible only in certain periods of the year (aroundJune, for X + x � 90�, and around December, forX + x � 270�), while the value of XM, varying with aperiod of 18.6 years, is basically a function of the yearof disposal.

These results, in all or in part, were substantially con-firmed by other independent investigations (Yaoxiang,2006; Gopinath and Ganeshan, 2006; Lewis et al., 2006).

Therefore, Eq. (1) can in fact guarantee its intended goalonly by constraining the disposal orbit eccentricity. If thedisposal perigee is left unconstrained, the graveyard orbitmust be nearly circular, with an eccentricity typically small-er than 0.005. However, for small eccentricities, compara-ble to the natural values induced by solar radiationpressure, a sun-pointing perigee may result in a slightlymore efficient re-orbiting strategy, compared to Eq. (1),which is able to avoid any further interference with the geo-stationary protected region (Fig. 1).

On the other hand, for eccentricities larger than the natu-ral value induced by solar radiation pressure, the eccentricityvector evolution is driven by luni-solar perturbations, asshown in Figs. 2 and 3. In these cases, an unconstrained dis-posal eccentricity vector may produce large variations in theperigee altitude and the consequent violation of the protect-ed region (see, for example, Figs. 4 and 5). However, due tothe form of the second order luni-solar perturbation term(see, for example, Yaoxiang (2006)), at X + x @ 90� ± 15�or 270� ± 15� the perigee altitude evolution can guaranteethe long-term preservation of the geostationary protectedregion, even with initial disposal eccentricities as large as0.3 (Figs. 6–9). The details depend, of course, on the initialconditions, such as the eccentricity, the inclination, the sea-son (i.e. the right ascension of the sun; see Fig. 10), and theright ascension of the moon ascending node (Fig. 11). Butthe conclusion remains the same: even significant (and unre-alistic – from a propellant budget point of view) re-orbitingeccentricities may be acceptable, in terms of preserving thegeostationary protected region, provided that the disposalorbit perigee is oriented so as to minimize the altitude varia-tions induced by the luni-solar perturbations.

The results outlined in this section add a new dimensionto the operational flexibility available to geostationarysatellite operators in order to comply with the spirit ofthe IADC mitigation guidelines. Eq. (1) is in fact a goodguide and in most cases the eccentricities of re-orbitedspacecraft are lower than 0.005. However, the inaccuraciesof the residual propellant estimates, the passivationrequirement, operational constraints, the health of thesatellite, the status of the propulsion system and so on

2010 2020 2030 2040 2050 2060 2070 2080 2090 21000.084

0.086

0.088

0.09

0.092

0.094

0.096

0.098

0.1

0.102

Year

Ecc

entr

icit

y

All perturbationsNo sun & moonNo sunNo moonNo sun, moon, radiation pressureOnly J

2

Fig. 3. Impact of different orbit perturbations on the evolution, over 100years, of the eccentricity of a satellite re-orbited according to Eq. (1). Thesimulated spacecraft is characterized by A/M = 0.02 m2/kg and Cr = 1.2.The initial disposal eccentricity is 0.1 and the perigee is oriented such thatX + x = 270�.

2010 2020 2030 2040 2050 2060 2070 2080 2090 2100-100

0

100

200

300

400

500

600

Year

Alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

Ω + ω = 282˚Ω + ω = 327˚Ω + ω = 12˚Ω + ω = 57˚

Protected Region

Fig. 4. Perigee altitude evolution, over 100 years, after a simulatedre-orbiting according to Eq. (1) of a spacecraft with A/M = 0.02 m2/kgand Cr = 1.2. The initial mean eccentricity of the disposal orbit is 0.05.The outcome is strongly dependent on the initial value of X + x: if it issufficiently close to 270� (282� in this example), the perigee will remainabove the geostationary protected region, while in the other cases theinterference cannot be avoided.

2010 2020 2030 2040 2050 2060 2070 2080 2090 2100-100

0

100

200

300

400

500

600

Year

Alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

Ω + ω = 102˚Ω + ω = 147˚Ω + ω = 192˚Ω + ω = 237˚

Protected Region

Fig. 5. The same as Fig. 4, with different initial values of X + x,completing a 360� scan at steps of 45�. Again, the outcome is stronglydependent on the initial value of X + x: if it is sufficiently close to 90�(102� in this example), the perigee will remain above the geostationaryprotected region, while in the other cases the interference cannot beavoided.

2010 2020 2030 2040 2050 2060 2070 2080 2090 2100-600

-400

-200

0

200

400

600

800

1000

Year

Alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

Ω + ω = 0˚Ω + ω = 90˚Ω + ω = 180˚Ω + ω = 270˚

Protected Region

Fig. 6. Perigee altitude evolution, over 100 years, after a simulatedre-orbiting according to Eq. (1) of a spacecraft with A/M = 0.02 m2/kgand Cr = 1.2. The initial mean eccentricity of the disposal orbit is 0.1. Ifthe initial value of X + x is close to 90� or 270�, the perigee will remainabove the geostationary protected region, otherwise a long-term interfer-ence cannot be avoided.

1094 L. Anselmo, C. Pardini / Advances in Space Research 41 (2008) 1091–1099

may prevent, sometimes, the achievement of nearly circulardisposal orbits (Delong and Fremeaux, 2005; Alby, 2006;Fremeaux, 2006). This makes some of the alternative solu-tions presented here very appealing.

3. The importance of passivation

In the IADC Space Debris Mitigation Guidelines, theend-of-life re-orbiting defined by Eq. (1) was associatedwith propulsion system passivation (IADC, 2002). This

requirement stems from the need to avoid accidental break-ups, whose resulting fragments might interfere with thegeostationary protected region.

In order to evaluate the relative importance of passiv-ation with respect to satellite re-orbiting, a detailed investi-gation was carried out at ISTI/CNR with a new modelingapproach (Anselmo and Pardini, 2000; Pardini andAnselmo, 2001; Anselmo and Pardini, 2002). In particular,the short and long-term effects of spacecraft explosions, asa function of the end-of-life re-orbit altitude (up to

2010 2020 2030 2040 2050 2060 2070 2080 2090 2100200

300

400

500

600

700

800

900

1000

Year

Alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

Ω + ω = 90˚Ω + ω = 102˚Ω + ω = 270˚Ω + ω = 282˚

Protected Region

Fig. 7. Perigee altitude evolution, over 100 years, after a simulated re-orbiting according to Eq. (1) of a spacecraft with A/M = 0.02 m2/kg andCr = 1.2. The initial mean eccentricity of the disposal orbit is 0.1. Eveninitial values of X + x not too far from 90� or 270� (102� and 282�,respectively, in this example), are able to avert any long-term interferencewith the geostationary protected region.

2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

500

1000

1500

2000

2500

Year

Alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

Ω + ω = 282˚Ω + ω = 102˚

Protected Region

Fig. 8. Perigee altitude evolution, over 100 years, after a simulated re-orbiting according to Eq. (1) of a spacecraft with A/M = 0.02 m2/kg andCr = 1.2. The initial mean eccentricity of the disposal orbit is 0.2. Even inthis case, initial values of X + x not too far from 90� or 270� (102� and282�, respectively, in this example), are able to avert any long-terminterference with the geostationary protected region.

2010 2020 2030 2040 2050 2060 2070 2080 2090 21000

500

1000

1500

2000

2500

3000

3500

4000

4500

Year

Alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

Ω + ω = 282˚Ω + ω = 102˚

Protected Region

Fig. 9. Perigee altitude evolution, over 100 years, after a simulated re-orbiting according to Eq. (1) of a spacecraft with A/M = 0.02 m2/kg andCr = 1.2. The initial mean eccentricity of the disposal orbit is 0.3. Evenwith such a large eccentricity, initial values of X + x not too far from 90�or 270� (102� and 282�, respectively, in this example), are able to avert anylong-term interference with the geostationary protected region.

0 10 20 30 40 50 60 70 80 90 100150

200

250

300

350

400

450

500

550

Time [years]

Alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

Jan 1 (α = 0˚)Apr 1 (α = 271˚)May 1 (α = 243˚)Jul 1 (α = 182˚)Oct 1 (α = 94˚)

Ω + ω = 282˚

Protected Region

Fig. 10. Even choosing initial values of X + x not too far from 90� or 270�(282�, in this example), the details of the perigee altitude evolution dependon the initial conditions, for instance on the angle a between the disposalorbit perigee and the right ascension of the sun. Having fixed X + x, thevalue of a is a function of the disposal day of the year (season). For asimulated re-orbiting in 2005, according to Eq. (1), of a spacecraftcharacterized by A/M = 0.02 m2/kg, Cr = 1.2 and initial meaneccentricity = 0.05, the optimal solution, in terms of maximum long-termseparation from the geostationary protected region, is obtained with asun-pointing perigee (a = 0). The conditions that minimize the impact onthe eccentricity evolution of both luni-solar perturbations (through thecontrol of X + x) and solar radiation pressure (through a sun-pointingperigee) occur around June, for X + x � 90�, and December, forX + x � 270�.

L. Anselmo, C. Pardini / Advances in Space Research 41 (2008) 1091–1099 1095

2000 km) above the geostationary orbit, were analyzed interms of their additional contribution to the debris densityand flux in the geostationary ring.

Taking for good the predictions of the current environ-mental models, the results obtained showed that a relative-ly small number of explosions in geostationary orbit (<10)would be sufficient, in the long-term, to double the current-ly estimated collision risk in the geostationary operationalring with objects larger than 1 and 10 cm. Even the adop-tion of the end-of-life re-orbiting defined by Eq. (1) would

not be safe enough without propulsion system passivation.In fact, a number of breakups in between 10 and 20 wouldbe sufficient to double the average flux of debris larger than1 and 10 cm, thus matching the effect of the currently

10 20 30 40 50 60 70 80 90 100100

200

300

400

500

600

700

800

900

1000

Time [years]

Per

igee

alt

itu

de

abo

ve g

eost

atio

nar

y o

rbit

[km

]

ΩM

= 5˚Ω

M = 0˚

ΩM

=-13˚Ω

M = 13˚

Protected Region

Fig. 11. Having fixed X + x, and implicitly a, with the choice of the day ofthe year, the details of the perigee altitude evolution depend also on thevalue of the right ascension of the moon ascending node (XM) at the timeof disposal (�year). With a period of 18.6 years, XM assumes values inbetween, approximately, ±13�. For a simulated re-orbiting, according toEq. (1), of a spacecraft characterized by A/M = 0.02 m2/kg, Cr = 1.2,initial mean eccentricity = 0.1 and X + x = 270�, the optimal solution, interms of maximum long-term separation from the geostationary protectedregion, is obtained by super-synchronous disposals carried out in yearswith values of XM close to 0�.

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Mea

n E

ccen

tric

ity

1 month2 months3 months4 months6 months1 year

1096 L. Anselmo, C. Pardini / Advances in Space Research 41 (2008) 1091–1099

estimated background in the geostationary ring. Only witha re-orbiting altitude of at least 2000 km might the numberof explosions needed to double the debris flux in the geosta-tionary ring increase significantly (�100).

Therefore, the passivation requirement is very impor-tant. The widespread adoption of the end-of-life re-orbitingof geostationary satellites, according to Eq. (1) and the dis-cussion in Section 2, would only meet the mitigation targetimplicit in the IADC Space Debris Mitigation Guidelines ifpropulsion system passivation were extensively and suc-cessfully carried out, not only for spacecraft, but for upperstages remaining in the geosynchronous region as well. Ifthere is operational conflict between re-orbiting and passiv-ation, clear priority should be given to the latter. Forinstance, in certain cases an incompatibility may arisebetween disposal eccentricity minimization and full passiv-ation (Fremeaux, 2006). This may be due to uncertaintiesin the residual propellant estimates, to the effects of tanksdepressurization, and to operational constraints. Passiv-ation, meaning future breakup prevention, should clearlybe preferred in such cases, since it is statistically more effec-tive as a long-term mitigation measure.

0 10 20 30 40 500

0.1

0.2

Area-to-Mass Ratio [m2/kg]

Fig. 12. Short-term eccentricity evolution of objects released in geosta-tionary orbit as a function of area-to-mass ratio. The details may varyslightly depending on the initial conditions.

4. The release of debris with high area-to-mass ratio

Even properly disposed and passivated geostationarysatellites may release debris through low energy processessuch as surface degradation and material deterioration.The resulting fragments are generally characterized byquite high area-to-mass ratios (A/M) compared to those

of typical spacecraft and upper stages. The release of highA/M objects in the geostationary ring or in the graveyardorbits is not directly addressed in the IADC Space Debris

Mitigation Guidelines. Nevertheless, recent optical observa-tions, carried out with ESA’s 1-m telescope in Tenerife,have identified a population of faint uncatalogued objects,with mean motions of about 1 revolution per day and orbi-tal eccentricities as high as 0.55 (Schildknecht et al., 2005).The obvious explanation for this surprising discovery wasthe action of solar radiation pressure on bodies of suffi-ciently high area-to-mass ratios (see Liou and Weaver(2004), for example). In such cases, in fact, the direct solarradiation pressure may significantly affect the eccentricity,with small effects on the orbit total energy and, therefore,on the semimajor axis or mean motion.

Recently both the short and long-term orbital evolutionof geosynchronous objects with high area-to-mass ratioswas studied in detail (Anselmo and Pardini, 2005; Liouand Weaver, 2005; Pardini and Anselmo, 2006). As shownby the comprehensive simulations carried out at ISTI/CNR, a very rapid growth in eccentricity (Fig. 12) andinclination (Fig. 13) is expected, strongly correlated withA/M: the greater the latter, the larger the effect. Objectsabove a certain A/M threshold (40–45 m2/kg, dependingon the initial conditions) develop an eccentricity so large,and a perigee so low, that a decay from orbit in less thansix months is induced. The objects below this A/M thresh-old are not subject to this fate. In fact, after approximatelysix months their eccentricity inverts the trend, reaching anew minimum just about one year after release. Thisshort-term cycle typically repeats itself for many years,the exact duration depending on the initial release condi-tions (Anselmo and Pardini, 2005; Pardini and Anselmo,2006). The semimajor axis, on the other hand, remainsclose to the geosynchronous one until orbital decay, while

0 10 20 30 40 500

5

10

15

20

25

30

Area-to-Mass Ratio [m2/kg]

Mea

n In

clin

atio

n [

deg

]

1 month2 months3 months4 months5 months1 year

Fig. 13. Short-term inclination evolution of objects released in geosta-tionary orbit as a function of area-to-mass ratio. The details may varyslightly depending on the initial conditions.

L. Anselmo, C. Pardini / Advances in Space Research 41 (2008) 1091–1099 1097

the inclination and the ascending node experience a long-term perturbation in which an increase in the area-to-massratio has, as its consequence, a faster orbit pole precessionand a wider amplitude of the plane motion (Fig. 14).

The results obtained indicate that objects with an A/Min between, approximately, 10 and 25 m2/kg, dependingon the release initial conditions and elapsed time, mightexplain the recently discovered debris population withmean motions of about one revolution per day and orbitaleccentricities as high as 0.55. But they also clearly demon-strate that high A/M objects released with a negligible DV

from aging spacecraft abandoned in super-synchronousdisposal orbits may very rapidly cross the geostationaryprotected region, interfering with this important volume

0 10 20 30 40 500

5

10

15

20

25

30

35

Time [years]

Mea

n In

clin

atio

n [

deg

]

A/M = 1 m2/kgA/M = 5 m2/kgA/M = 10 m2/kgA/M = 15 m2/kg

Fig. 14. Long-term inclination evolution of objects released in geosta-tionary orbit with area-to-mass ratios of 1, 5, 10, and 15 m2/kg. Thedetails may vary slightly depending on the initial conditions.

of space for periods from months to several decades,depending on the initial conditions and area-to-mass ratio.

The potential collision risk posed by this new class ofobjects clearly depends on their exact source mechanismand production rate, which are both still unknown. Cur-rently, to avert energetic breakups, end-of-life re-orbitingis recommended along with propulsion system passivation.However, in the future, it may be needed the explicitintroduction of a new mitigation requirement, intendedto avoid the low energy release of debris above a givenmass and A/M threshold.

5. Conclusions

During the last decade an awareness of long-term miti-gation matured amongst the operators of geostationarysatellites, leading to a growing number of spacecraft beingremoved from the operational ring at the end-of-life. In2005, more than 50% of the disposals were carried out inagreement with the IADC recommendation, defined byEq. (1) (Jehn and Klinkrad, 2006). In addition, the formulawas presented and discussed at the United Nations, incor-porated into the IADC Space Debris Mitigation Guidelines(2002) and into the European Code of Conduct for Space

Debris Mitigation (2004), and considered for adoption bythe International Telecommunication Union (ITU) andthe International Organization for Standardization (ISO).

Recent observational, modeling and operational activi-ties have shown, however, that there is still room for clar-ification and improvement. Detailed analyses havedemonstrated that Eq. (1) must be coupled with require-ments on eccentricity vector management and control,and that in certain cases there are slightly more convenientoptions for end-of-life spacecraft disposal. In addition, there-orbiting operational practice highlighted the importanceof uncertainty in the residual propellant estimates, the pos-sible conflict between maximum eccentricity tolerance andfull passivation, the prospect of attitude loss and/or alti-tude decrease induced by tank depletion, and other engi-neering and navigational constraints (Alby, 2006).

Based on these arguments, a possible solution would beto abandon Eq. (1), and focus on the higher level require-ment, i.e. the long-term preservation of the geostationaryprotected region. This approach would guarantee the max-imum flexibility to satellite operators and the re-orbitingplanning would be optimized as a function of satellite oper-ational constraints, health and control strategy (e.g. sun-pointing mode), propulsion system, accuracy of residualpropellant estimate, passivation requirement, initial condi-tions, right ascension of the sun, right ascension of themoon ascending node, and expected solar radiationperturbation.

However, Eq. (1) is by now well known and popular,providing an easy to understand a nearly optimalready-made solution for quasi-circular disposal orbits. Inthe space debris mitigation community there is, therefore,no consensus about the prospect of getting rid of it

1098 L. Anselmo, C. Pardini / Advances in Space Research 41 (2008) 1091–1099

completely. Consequently, Eq. (1) will probably be main-tained for the time being, supplemented by additionalrequirements and/or clauses dealing with eccentricity vec-tor management.

With regard to the end-of-life passivation of propulsionsystems, its importance for the future preservation of thegeostationary region can never be stressed enough. Theimpact velocity in the operational geostationary ring ofexplosion fragments generated in quasi-circular disposalorbits, up to 2000 km higher, is in between 0.1 and0.9 km/s (Anselmo and Pardini, 2002), i.e. much lowerthan in low earth orbit. Nevertheless, energetic breakupsmust absolutely be avoided in the protected region and inthe super-synchronous graveyard orbits, because the frag-ments are not easily swept away by the perturbations, sincethey tend to accumulate and permanently interfere with thegeostationary space. In statistical terms, passivation is themost important long-term mitigation measure. If there isany operational conflict with other requirements, such asdisposal eccentricity minimization, passivation should beclearly favored.

The results of recent optical observations and modelingactivities demonstrate, however, that energetic breakupsare not the only source of concern. The low energy release,in disposal orbit, of objects with high area-to-mass ratiosmay, in fact, very rapidly affect the geostationary protectedregion, partially neutralizing the benefits of end-of-life re-orbiting and propulsion system passivation. Even thoughthe origin and the extent of the phenomenon are still notclear, this potential problem should be carefully evaluatedand addressed in the future, possibly leading to the formu-lation of specific mitigation requirements.

In conclusion, during the last decade significant progresshad been made to reduce the growth of orbital debris in thegeostationary region and to preserve this important volumeof circumterrestrial space for future generations and appli-cations. However, a continuous flow of new discoveriesand insights demands a step by step approach to keepupdated, improving and effective the strategies elaboratedfor space debris mitigation.

References

Alby, F. 1st European workshop on geostationary satellites end of life, in:Proceedings of the 24th IADC Plenary Meeting, JAXA, TsukubaSpace Center, Tsukuba, Japan, CD-ROM, April 10–14, 2006.

Anselmo, L. GEO disposal orbit eccentricity, in: Proceedings of the 24thIADC Plenary Meeting, JAXA, Tsukuba Space Center, Tsukuba,Japan, CD-ROM, April 10–14, 2006.

Anselmo, L., Pardini, C. The effects of spacecraft and upper stagebreakups on the geostationary ring. The Journal of the AstronauticalSciences 48, 1–23, 2000.

Anselmo, L., Pardini, C. Collision risk mitigation in geostationary orbit.Space Debris 2, 67–82, 2002.

Anselmo, L., Pardini, C. Orbital evolution of geosynchronous objects withhigh area-to-mass ratios, in: Danesy, D. (Ed.), Proceedings of theFourth European Conference on Space Debris, ESA SP-587. ESAPublications Division, Noordwijk, The Netherlands, pp. 279–284,2005.

Delong, N., Fremeaux, C. Eccentricity management for geostationarysatellites during end of life operations, in: Danesy, D. (Ed.), Proceed-ings of the Fourth European Conference on Space Debris, ESA SP-587. ESA Publications Division, Noordwijk, The Netherlands, pp.297–302, 2005.

European Space Debris Mitigation and Safety Working Group(EDMSWG). European Code of Conduct for Space Debris Mitiga-tion, Issue 1.0, European Network of Centres – Space Debris,Toulouse, France, June 28, 2004.

Fremeaux, C. Decommissioned satellites orbit evolution, in: Proceedingsof the 24th IADC Plenary Meeting, JAXA, Tsukuba Space Center,Tsukuba, Japan, CD-ROM, April 10–14, 2006.

Gopinath, N.S., Ganeshan, A.S. Long term evolution of objects inGSO-disposal orbit, in: Danesy, D. (Ed.), Proceedings of theFourth European Conference on Space Debris, ESA SP-587. ESAPublications Division, Noordwijk, The Netherlands, pp. 291–296,2005.

Gopinath, N.S., Ganeshan, A.S. GEO disposal orbit eccentricity study, in:Proceedings of the 24th IADC Plenary Meeting, JAXA, TsukubaSpace Center, Tsukuba, Japan, CD-ROM, April 10–14, 2006.

Grimwood, T. The UCS Satellite Database, Union of ConcernedScientists (UCS), Cambridge, Massachusetts, USA, http://www.ucsu-sa.org/satellite_database, June 19, 2006.

Hernandez, C., Jehn, R. Classification of Geosynchronous Objects, Issue8, European Space Operations Centre, ESA, Darmstadt, Germany,February 2006.

Inter-Agency Space Debris Coordination Committee (IADC). WorkingGroup 4 Minutes, in: Proceedings of the 15th IADC Plenary Meeting,IADC-97-03, NASA, Johnson Space Center, Houston, Texas, USA,document M-38, December 9–12, 1997.

Inter-Agency Space Debris Coordination Committee (IADC). SpaceDebris Issues in the Geostationary Orbit and the GeostationaryTransfer Orbits, IADC-00-01, Presented to the 37th Session of theScientific and Technical Subcommittee, Committee on the PeacefulUses of Outer Space (COPUOS), United Nations, Vienna, Austria,2000.

Inter-Agency Space Debris Coordination Committee (IADC). IADCSpace Debris Mitigation Guidelines, IADC-02-01, October 15,2002.

Jehn, R., Klinkrad, H., Trends in Reorbiting of Geostationary Satellites,25th International Symposium on Space Technology and Science,Paper 2006-r-2-13, CD-ROM, The Organizing Committee of the 25thISTS and the Japan Society for Aeronautical and Space Sciences,Kanazawa, Japan, June 2006.

Kwok, J.H. The Long-Term Orbit Predictor (LOP), Version 2.0, NPO-17052, Jet Propulsion Laboratory, Pasadena, California, USA, June1986.

Lewis, H.G., Swinerd, G.G., Martin, C.E., Campbell, W.S. The stabilityof disposal orbits at super-synchronous altitudes. Acta Astronautica55, 299–310, 2004.

Lewis, H.G., Swinerd, G.G., Darby, R., Martin, C.E., Saunders, C. GEOdisposal study, in: Proceedings of the 24th IADC Plenary Meeting,JAXA, Tsukuba Space Center, Tsukuba, Japan, CD-ROM, April 10–14, 2006.

Liou, J.-C., Weaver, J.K. Orbital evolution of GEO debris with very higharea-to-mass ratios. The Orbital Debris Quarterly News 8 (3), 6–7,2004.

Liou, J.-C., Weaver, J.K. Orbital dynamics of high area-to-mass ratiodebris and their distribution in the geosynchronous region, in: Danesy,D. (Ed.), Proceedings of the Fourth European Conference on SpaceDebris, ESA SP-587. ESA Publications Division, Noordwijk, TheNetherlands, pp. 285–290, 2005.

Martin, C.E., Stokes, P.H., Walker, R., Klinkrad, H. The long-termevolution of the debris environment in high earth orbit including theeffectiveness of mitigation measures, in: Bendisch, J. (Ed.), SpaceDebris 2001, Science and Technology Series, A supplement toAdvances in the Astronautical Sciences, vol. 105. Univelt Inc., SanDiego, California, USA, pp. 141–154, 2002.

L. Anselmo, C. Pardini / Advances in Space Research 41 (2008) 1091–1099 1099

Pardini, C., Anselmo, L. On the effectiveness of end-of-life re-orbiting fordebris mitigation in geostationary orbit. Space Debris 1, 173–193,2001.

Pardini, C., Anselmo, L. Long-Term evolution of geosynchronous orbitaldebris with high area-to-mass ratios, in: 25th International Symposiumon Space Technology and Science, Paper 2006-r-2-10, CD-ROM, TheOrganizing Committee of the 25th ISTS and the Japan Society forAeronautical and Space Sciences, Kanazawa, Japan, June 2006.

Schildknecht, T., Musci, R., Flury, W., Kuusela, J., de Leon, J., de FatimaDominguez Palmero, L. Optical observations of space debris in high-altitude orbits, in: Danesy, D. (Ed.), Proceedings of the FourthEuropean Conference on Space Debris, ESA SP-587. ESA Publica-tions Division, Noordwijk, The Netherlands, pp. 113–118, 2005.

Yaoxiang, J. Effect of initial eccentricity vector on GEO disposal orbits,in: Proceedings of the 24th IADC Plenary Meeting, JAXA, TsukubaSpace Center, Tsukuba, Japan, CD-ROM, April 10–14, 2006.