As the number of Earth-orbiting spacecraft continues to grow, the efficient utilisation of near-Earth space may be facilitated by the adoption of certain traffic rules. Dalbello, for example, has proposed six "rules of the road" that might contribute to the peaceful use of outer space: "keep-out zones", limits on minimum separation distance between satellites, rights of inspection, restrictions on low-orbit overflight, advance notice of launch activities, and a "hot line" for space activities (1).

Another possibility is for several spacecraft to use the same orbit, as for instance do Navstar satellites. This would involve allocating 'slots' in certain low Earth orbits to specified users, as currently occurs in the use of the geostationary Earth orbit ( GEO), and is discussed in principle in (2).

The allocation of orbital positions in GEO between users is a relatively straightforward process due to the orbit's geophysical uniqueness, and to the possibility of simply defining a satellite's position in it with respect to the Earth. Thus the procedure involves primarily the allocation of a fixed number of orbital 'slots' to potential users, according to what is agreed in international negotiations to be a fair reflection of competing claims.

It is notable, however, that even in the context of the easily-defined geostationary Earth orbit, a number of technical matters have to be considered in detail. These include deciding on the optimal size of the slots in degrees of longitude; the accuracy with which satellite positions within these slots must be maintained; the amount of electromagnetic radiation that satellites may be permitted to emit, in terms of power, frequency and geographical spread; the amount of pollution in terms of thruster exhaust that satellites may produce; rights of passage of satellites arriving at GEO; and procedures to be followed at the end of GEO satellites' operational lifetimes.

Achieving internationally acceptable solutions to these questions requires consideration not only of legal principles but also of the technology of telecommunications satellite systems, since the decisions made place demands on manufacturers and users. For example, if increased crowding of certain regions of GEO leads to the decision to reduce the longitudinal width of slots, satellite manufacturers and users are thereby obliged to incur expenses to comply with these higher standards, both of physical accuracy of station-keeping and of narrower beamwidths of uplink and downl ink.

Reaching similar international agreement on the utilisation of low Earth orbits ( LEO) would be more difficult than the same process relating to GEO for a number of reasons: First, the definition of other Earth orbits than GEO in physical terms is more complex: spacecraft orbits in near-Earth space evolve through time due to perturbations caused by the Earth's oblateness, by the influence of other celestial bodies, by solar radiation and by the Earth's atmosphere. Thus two dissimilar spacecraft moving passively in a given low Earth orbit would not, except in special cases, remain in the same orbit over an extended period of time.

Because of this, if a particular orbit was to be used by many different spacecraft, users would be required to take actions to maintain their positions in the orbit that were substantially more complex than those required to maintain a constant position in GEO. In addition, the cost of maintaining a defined position in a specified low Earth orbit will differ between spacecraft in that orbit due to their different physical geometry leading to different reactions to atmospheric drag, solar pressure, and so on. The selection of an optimal orbit would thus involve economic trade-offs between the requirements of different potential users of the orbital zone under consideration.

A second reason why reaching agreement on the regulation of low Earth orbits would be more difficult than agreeing on GEO utilisation is that there are innumerable mutually exclusive ways of dividing near-Earth space into particular orbits or orbital zones. Consequently, in addition to balancing different users' claims within any given orbit, as in the allocation of slots in GEO, it would also be necessary to make judgements as between different means of partitioning near-Earth space, since the allocation of a particular orbital zone to certain users would preclude the use of a range of other orbits that intersected this zone, at least for certain purposes.

In the following sections I examine a range of considerations that would need to be taken into account in any attempt to define one or more low Earth orbital traffic zones. These fall into two categories:

• considerations relating to the precise legal status that such orbits and their users should be accorded; and

• factors that would be relevant in selecting appropriate values for each of the parameters required to define such a zone.

To date, much of the discussion of the possibility of introducing traffic regulations in near Earth space has been in connection with military systems. For example Schwetje (3) and Bittlinger (4) have discussed the concept of "keep-out zones" and their possible use for protecting military systems.

The present discussion is concerned specifically with civilian systems, and in particular with a 'nominal' case of a specified orbit to be used by some tens of commercially-owned, civilian facilities, both national and international, each providing accommodation for hundreds of people, and receiving up to several visit per day from launch vehicles. Such facilities would be used in either of the two non-military scenarios which it has been proposed could require the early establishment of LEO traffic systems, namely the construction of satellite solar power stations and the operation of orbital tourist facilities (2).

There are a large number of consideration that would enter into any decision on the legal status that might be accorded for the purposes of traffic regulation to selected low Earth orbits and their users.

A) Legal Rights of Registered Users of LEO Slots

Perhaps the most important consideration is the question of the legal rights that should be given to the registered user of a specified LEO 'slot', the most important such right being the form of legal protection that they would have against interference by other space vehicles. The possible levels of protection that might be given to a registered space vehicle in a particular LEO slot can usefully be divided into three:

1. A minimal agreement might state that registered users of internationally agreed LEO 'slots' would be protected to the extent that any spacecraft interfering or colliding with them would be absolutely liable for any damage caused.

2. A more extensive agreement could establish an exclusion zone around the registered vehicle which other vehicles were not permitted to enter without prior permission.

3. The most extreme form of protection would be to establish an exclusion zone around registered facilities within which registered users would be permitted to use force to prevent the approach of other vehicles, even including their destruction.

1. The minimal case would be the simplest to agree, being effectively a straightforward extension of the existing GEO regime to specified regions of LEO space. Even in this case, though, there would he scope for difficulties in defining 'interference'. Some spacecraft equipment, such as astronomical telescopes, is extremely sensitive, and could suffer interference from a satellite many hundreds of kilometres away. Such extreme cases would need to be accommodated in some manner in any allocation of LEO slots.

2. The principle of recognising an exclusion zone around operating satellites has been advocated for some years by Soviet authors (5, 6). Defined as the region within which other space vehicles would interfere with a registered vehicle, it is in effect the operational interpretation of the legal definition of the spacecraft itself - damage caused by electromagnetic interference is as real as that caused by mechanical impact, which is illegal under the 1972 Convention on Liability (7). Dudakov has proposed that

   "Without the consent of the State holding jurisdiction over the space object in orbit, not a single State is permitted either to investigate using its own spacecraft the satellites of other States, or to get too close to them violating a certain distance limit, which may lead to interfering with scientific devices carried aboard those satellites measuring procedures and normal functioning of telemetric and communications means.

   "... Even short-duration stationing (without any activities on the part of the pirate) in the vicinity of the satellite, which, as a rule, is equipped with rather sensitive scientific devices, communications and telemetric facilities, may result in interference and substantially affect the satellite performance" (5).

   However, as mentioned above, the zone within which interference may be caused to a satellite may be impractically large, and thus the practical realisation of the principle would require this difficulty to be resolved.

3. The concept of exclusion zones becomes more controversial when the possible right to self-defence is included. By the nature of orbits, collisions between orbital bodies generally occur at extremely high velocities (typically several kilometres per second). As a consequence, collisions other than those between vehicles in stable (and so predictable) orbits, would tend to occur with very little warning; they would be very difficult to prevent; and they would cause extreme damage.

   If exclusion zones were intended to ensure the safety of registered vehicles against collision, it would therefore be necessary to include the right to self-defence. Given the nature of the danger, however, defensive systems capable of preventing collisions would themselves have to be capable of extreme damage. It therefore seems unlikely that such an extreme form of legal protection would be readily agreed in international fora.

Bittlinger concluded (4) that unilateral declaration of "keep-out zones" around military facilities, which would include the implicit threat of self-defence, would contravene Article 2 of the 1967 Outer Space Treaty (8). It is not clear, however, that Soviet space lawyers would oppose such a development. Dudakov, for instance, discusses the need for self-defence against space piracy:

"We consider an attempt to get aboard other State's satellites to inspect or seize them an act of piracy granting the owner-States rights to resort to any means to protect its space objects" (5). (Dudakov also points that under existing space law such actions would constitute 'State piracy'.)

In practice, provided that internationally agreed regulations were generally obeyed, there should not be a need for such a development - as indeed it has not been found necessary for vehicles to carry defensive weapons in the cases of road, sea or air traffic. It is to be hoped, therefore, that the growth of space traffic will follow these precedents in being generally peaceful and law-abiding, and will thereby avoid the need for self-defence capabilities on civilian facilities, and the additional costs that would be incurred.

Whether or not exclusion zones of some kind were formally established by international agreement, any discussion of legal recognition of LEO slots leads naturally to the question of how long such arrangements should last.

B) LEO Slots and Appropriation of Space

The meaning of "appropriation" of space in the context of the 1967 Treaty is the subject of continuing discussion. In the present context it is clear that the longer the period of time for which a user had the right of exclusive occupancy of a particular LEO 'slot', the stronger the grounds for considering that the activity constituted "appropriation".

In general, in normal civilian commercial projects, investors require security of tenure of productive facilities for the lifetime of their investment - typically at least twenty years, and preferably for longer. Such a length of time is longer than the operational lifetime of any orbiting satellite to date, and could well be argued to constitute "appropriation".

However, Schwetje concluded his discussion of "Keep-Out Zones" as follows:

"For a variety of reasons, space mines and keep-out zones may be bad ideas whose time will never come. The categorical rejection of space zones for various purposes of safety, security, and traffic management, however, would be a grave mistake. If we truly believe that space can be colonized by earthkind, these concepts should be considered to promote safety, security, and stability" (3).

Thus, in contrast to the case of appropriation of particular regions of space by unilateral declaration of "keep-out zones" for military purposes, it would seem likely to be mutually beneficial for space-faring nations to agree even permanent exceptions to the principle of non-appropriation of space in the case of commercial facilities, provided that users fulfilled certain internationally agreed legal duties.

C) Legal Duties of Registered Users of LEO Slots

Obtaining internationally-guaranteed and exclusive use of defined positions in LEO space would be commercially valuable, and as a quid pro quo applicants for such privilege would be expected to abide by certain rules that would make the utilisation of LEO more efficient for all users.

Rules to be imposed on registered users would be likely to include at least the following: strict limits in three physical axes on the position to be occupied by every facility; prohibition on the creation of all types of space refuse; controls on emissions of electromagnetic radiation (both electronic and optical); controls on releases of gases; contribution to the costs of maintenance of the traffic system; and acceptance of penalties for infringement of the above, the ultimate sanction for non-compliance being loss of the right of occupancy of their orbital slot, and of the accompanying legal protection.

D) Traffic to and from LEO Slots

An issue related to that of the regulations governing orbital facilities occupying registered LEO slots is the question of regulations governing space vehicles both approaching and departing from these facilities, and passing through a defined orbital zone. On the minimal interpretation of orbital slots (discussed in A) i) above), any space vehicle would be permitted to pass through a regulated orbital zone, but would be liable for any resulting collision with a registered user's facility. In view of the very high cost of damage caused by orbital collisions due to the high kinetic energies usually involved, this might not offer sufficient assurance of safety to registered users. Stricter control of space vehicles moving within or through the zones might therefore be required. For instance, vehicles might be permitted within the zone only with specific

permission obtained after prior inspection. Such a procedure would evolve naturally into a system of registration of orbital vehicles that included tests of "spaceworthiness", similar to the regulation of land, sea and air vehicles.

In addition, it would clearly be essential that space vehicles creating debris that itself passed through a regulated LEO zone should also be liable for any damage that it caused. In practice, in order to determine responsibility for creation of debris, continuous radar monitoring of orbital zones might become necessary. This would not be difficult to perform from spacecraft within the orbit, but would represent an additional communal expense for registered users. It might, however, be subsumed within the cost of a wider system of control of space refuse.

The problem of ensuring the safety of orbiting facilities against collisions includes the important danger posed by collisions with orbital debris, or space refuse, which comprises most orbital objects. Baker has discussed the liability for damage caused in space by space refuse in detail (9), and he argues convincingly that there is already an increasingly urgent need to make producers of space refuse liable for any damage that it causes. This subject is beyond the scope of the present discussion, but it is clear that major growth in human occupancy of near-Earth space would make such a development even more necessary. It should therefore be seen a necessary adjunct to the creation and allocation of LEO slots.

A desirable side-effect of introducing liability for damage caused by space refuse, as Baker proposes, is that the high cost of any resulting reparations would act as an incentive for the establishment of orbital salvage and repair services, which would enhance the ability to prevent collisions in cases of breakdown of spacecraft. In a discussion of salvage in the context of Outer Space Law, DeSaussure links it to the general subject of near-Earth traffic regulation:

"It has often been mentioned that a space navigation code will someday be needed to assure flight safety, and it would be such a code that could provide for the sweeping of hazardous space debris and clearing the most frequently used orbital paths" (10).

DeSaussure distinguishes sharply between the need for international rules for the safety of spaceflight agreed by public negotiation, and salvage where flight safety is not a factor, arguing that in the latter case the rules governing astrosalvage should be left to the field of private law and to be derived from the practice of space venturers and, preferably, through the conclusion of a private international law convention which draws on principles of the mercantile law of the admiralty" (10).

Thus, despite the likely difficulty of persuading producers of space refuse to accept liability for damage caused by it in space, such a development would indirectly have the substantial advantage of harnessing market forces to the improvement of orbital safety.

Agreement would also need to be reached concerning the application of any regulation of LEO traffic zones to spacecraft using tethers. The reason for treating the use of tethers as a separate case is because in the foreseeable future tethers tens and even hundreds of kilometres in length may come to be used in LEO due to their commercially and scientifically valuable capabilities (11).

A relatively simple approach to accommodating such systems would be to enlarge the physical dimensions of defined orbital zones. Such an approach would be appropriate for more or less permanent tether systems. However, it is clearly limited by the economic need to minimise the size of any particular zone, other things being equal.

In cases where a registered user extended a tether beyond the limits of a defined zone, it would seem natural that their privileges as registered user of an orbital slot should not apply, and they should be liable for any damage caused to other vehicles outside the zone by the tether. This would clearly be appropriate for temporary tethers, such as those used for de-orbiting launch vehicles, or for launching satellites into higher orbits. However, the case of very long, permanent tether systems overlaps both these approaches, and would require further consideration.

As discussed above, low Earth orbital traffic zones will not in general comprise passive orbits. That is, users will be obliged to take active guidance measures in order to stay in their allocated positions within a particular orbital zone, though an important objective of the process of selecting orbits will be to minimise the extent, and so cost, of such actions.

The process of maintaining the position of a registered facility in its correct slot

in a defined low Earth orbit will be more complex than simply maintaining the facility at a fixed position relative to the surface of the Earth. It will involve keeping its altitude (apogee and perigee), orbital inclination, rate of nodal regression and rate of apsidal rotation within an agreed range of specified nominal values, while its position relative to the Earth changes constantly. The only bodies in relation to which its position would remain more or less constant would be other facilities occupying slots in the same orbit.

The need to make physical adjustments of a facility's position and velocity would not be fundamentally different from station-keeping in GEO; what would be significantly different is that the procedure for determining whether the facility's current position was correct or not would involve comparing it with a calculated nominal position which was continually changing relative to the Earth, rather than with a fixed position. With the use of computers this is not of course difficult, although the optimisation of station-keeping manoeuvres would be more complex due to their greater extent.

The reaching of international agreement to regulate the use of a particular LEO zone as proposed in the previous section would require decisions to be made on each of the parameters necessary to define the zone. In determining the values for these parameters there are many factors, both physical and economic, that would need to be considered. The most appropriate values would depend inter alia on the use to which the facilities occupying the orbit were to be put: the more similar these uses, the more similar would be the operators' requirements, and the fewer the compromises that would have to be made between them.

The proposed orbital zone's altitude, eccentricity, inclination, rate of nodal regression, and rate of apsidal rotation would all be chosen in order to minimise the costs to users of maintaining their orbital positions, subject to certain compromises that might be necessary to accommodate users of nearby orbital zones. The main considerations that would influence the choice of these parameters include the following:

The lower the altitude of an orbital zone, the lower the cost of launching payloads, whether passengers or material, to facilities using the orbit. In contrast, the lower the altitude, the faster the rate of orbital decay due to atmospheric drag acting on space vehicles, and the greater and more frequent the need for boosting of facilities.

The optimum altitude would also be influenced by the use to be made of facilities. If they were to receive daily visits from launch vehicles, there would be considerable potential for regularly boosting them in their orbits through momentum transfer to the orbital facilities, by using tethers for deorbiting returning vehicles. Thus, for example, it would be possible to site such facilities in a lower orbit in which, if they were not frequently reboosted, they would have a relatively short orbital lifetime (of the order of a year). The rate of orbital decay of a space vehicle in a given orbit depends on its density and geometry, and would therefore differ between facilities. Reaching agreement on such a low orbit would therefore require further compromise between different users' or potential users' preferences.

The greater the eccentricity of the orbit, the greater the cost of launch for a given perigee, but the wider the range of views of the Earth that are possible. That is, a facility in an eccentric orbit would offer a wider field of view of Earth at its apogee than at its perigee. At the same time, however, the total volume of space swept by craft in the orbit would be greater due to rotation of the apsides, making its establishment more expensive in terms of the amount of the limited resource of LEO space dedicated to such a zone - at least in the case of granting exclusion zones to users.

The lower the inclination of the orbit, the lower the cost of launch, but the less the proportion of the Earth that is visible from orbit. In the extreme case of equatorial orbits, the cost of launch is lowest, but only the equatorial belt of the Earth is visible at all.

The rate of nodal regression depends on the orbital inclination and period (and hence on the altitude and eccentricity), and together these factors determine how frequently an orbital vehicle passes the same point on the ground. As a result, certain orbits are very much more convenient than others for launching from given launch sites. This is a particularly important consideration in determining combinations of altitude, eccentricity and inclination that would be preferred by prospective users of a regulated orbital zone.

In non-circular orbits the apsides also rotate around the orbit at a rate dependent on the other orbital parameters In order to prevent the possibility of collision between users of the orbit it would therefore be necessary for them to agree to maintain a common rate of apsidal rotation. As mentioned above, the apsidal rotation enlarges the total volume of near-Earth space swept by the facilities in an eccentric orbit, in effect creating a zone within the orbital plane bounded by the circular orbits at the altitudes of the perigee and apogee respectively.

The accuracy of station-keeping required of users of a specified orbital zone would determine the range of movement allowed around the nominal parameter values. By analogy with the utilisation of GEO it seems likely that at earlier stages the permitted range would be relatively wide, and that later, as competition for the low Earth orbit resource increased, it would become narrower.

Concerning facilities' altitude, from the point of view of users of the orbit, the optimum variation permitted would be determined by the economically optimal pattern of orbital decay and reboost. This would in turn depend both on the economics of reboosting facilities and on geophysical factors, and would also vary according to solar and atmospheric conditions. It would also be influenced by the need to minimise inconvenience caused to users of registered facilities' by boost operations. The calculations of facility managers would be made even more complex by the possibility discussed above of reboosting facilities by using tethers to deorbit departing launch vehicles.

In order to minimise the amount of low Earth orbital space occupied by a given zone, the permitted altitude range should of course be as narrow as possible. The optimal value would therefore be a compromise between these conflicting preferences.

As discussed above, the optimum size and definition of possible exclusion zones around registered facilities would depend on the purpose which they were to serve. Economy in the use of low Earth orbital space would lead to smaller exclusion zones, however, as mentioned above, for some purposes larger exclusion zones would be required. Clearly exclusion zones several hundred kilometres across are not practical in low orbits.

Dalbello offered a sample definition of a "keep-out zone":

• Keep 100 kilometres and three degress out-of-plane from foreign satellites below 5000 kilometres;

• Keep 500 kilometres from foreign satellites above 5000 kilometres except those within 500 kilometres of geosynchronous altitude;

• One pre-announced close approach at a time is allowed;

In the event of a violation of the rules above, the nation of registry of the satellite which most recently initiated a manoeuvre "burn" is at fault and guilty of trespass (1).

This contains a number of useful ideas. However the final proposal to accord liability for any interference to the satellite which last made a manoeuvre "burn" is too simple to be more than a rough guide, as discussed in (2).

The use of tethers, whether permanently to extend facilities, or intermittently for docking and separation of launch and/or re-entry vehicles, introduces further complexities into decisions on the definition of orbital zones. Where these were taken into account, the degree of compromise required between different users would increase with greater variety in their use of tethers. For example, a facility comprising several segments connected by vertical tethers one hundred kilometres or more in length would have very different orbital preferences from a small facility a few tens of metres in dimensions. These problems overlap those relating to large space structures more generally.

It is clear from the above discussion that there are a very large number of factors that would have to be taken into account in any negotiations on the establishment of LEO traffic zones.

In practice, movement in the direction of agreeing on traffic zones in low Earth orbit is more likely initially to involve incremental changes to present arrangements than the introduction of a comprehensive system. Thus, for example, when a number of users of similar orbits perceive it in their interest to coordinate their activities, they may initially agree informally to harmonise their orbital paths, and subsequently agree rights and obligations through formal international negotiations, rather than attempt to establish a global system in which every such zone is defined ab initio. However, the ultimate evolution of such incremental steps would be the establishment of a system of global space traffic control, that would integrate with existing air traffic control.

A matter of great significance for the feasibility of large scale civilian use of orbital facilities is the future density of orbital debris, or space refuse. This is currently the subject of much serious study, and a range of actions including improved monitoring and research, measures to minimise debris production, and the possibility of reducing existing levels, are under discussion.

However, the prospect of facilities being developed with increasingly large cross-sections entails that not only is the need for such steps becoming more urgent but, as Baker has argued (9), there is also an increasing need for changes in Space Law. These should specifically extend liability for damage caused by space refuse to its creator. Reaching a binding international treaty to this effect will be difficult due both to the complexity of the issues, and to the potential cost to States and other Parties of accepting the liabilities involved. Nevertheless, it is clearly essential that progress be made in this direction as soon as possible.

Although the need for regulation of traffic in near-Earth space still lies some years in the future, it may not be more than a decade or two away (12, 13, 14, 15). Negotiation of satisfactory agreements will necessarily be a long-term process, and it is desirable to avoid the possibility that incremental developments in the utilisation of space may preclude the subsequent adoption of more efficient arrangements. Thus it is useful to begin to consider these longer-term possibilities already.

The achievement of international agreement along the proposed lines would clearly be a formidable task. However the mutual interest of the space-faring nations, the Soviet Union, the USA, Europe, Japan, India and others, in the efficient utilisation of space for peaceful purposes seems likely to provide a common motivation for success in this area.

1. R Dalbello, 1985, " 'Rules of the Road': Legal Measures to Strengthen the Peaceful Uses of Outer Space", IISL-85-03, pp 8-12
2. P Q Collins and T W Williams, 1986, "Towards Traffic Systems for Near-Earth Space", IISL-86-31, pp 161-170
3. F K Schwetje, 1987, " Protecting Space Assets: A Legal Analysis of 'Keep-Out Zones'", Journal of Space Law, Vol 15, No 2, pp 131-146
4. H Bittlinger, 1938, " "Keep-out Zones" and the Non-Appropriation Principle of International Space Law", IISL-88-002, pp 6-12
5. B G Dudakov, 1981, " On International Legal Status of Artificial Earth Satellites and the Zone Adjacent to Them", IISL-81-12, pp 97-101
6. V D Bordunov, 1981, " Rights of States as Regards Outer Space Objects", IISL81-10, pp 89-92
7. Convention on International Liability for Damage Caused by Space Objects, March 29 1972, 24 U.S.T. 2389, T.I.A.S. 7782, U.N. GAOR, 26th Session, Supp. 29 (Dcc. A/8429) (effective 9 October 1973)
8. Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, Including the Moon and Other Celestial Bodies, 27 January 1967, 18 U.S.T. 2410, T.I.A.S. 6347, 610 U.N. TS 205 (effective 10 October 1987)
9. H A Baker, 1988, " Liability for Damage Caused in Outer Space by Space Refuse", Annals of Air and Space Law, Vol 8, pp 183-227
10. H deSaussure, 1985, " The Application of Maritime Salvage to the Law of Outer Space", IISL-85-24
11. J A Carroll, 1985, " Guidebook for Analysis of Tether Applications", Final Report RH4-394049, Energy Sciences Laboratories, San Diego
12. T Yamanaka and M Nagatomo, 1986, " Spaceports and New Industrialized Areas in the Pacific Basin", Space Policy, Vol 2, No 4, pp 342-354
13. T F Rogers, 1989, "Space Settlements: Sooner Than We Think?" Ad Astra, Vol 1, No 1, pp3O-34
14. P Q Collins, 1989, " Stages in the Development of Low Earth Orbit Tourism", Space Technology, Vol 9, No 3, pp 315-323
15. J Yagi, 1989, Press Release, Shimizu Construction, Tokyo