This proposal has been prepared in consultation with the former members
of the Anglo-Australian Near Earth Asteroid Survey team (Ken Russell, Gordon
Garradd, Rob McNaught and Duncan Steel) and scientists involved in NEO
and related work throughout the world. The views presented in this document
are those of the author and do not necessarily represent the views of any
other person or organisation.
This proposal has been prepared on a voluntary basis - the author is not associated with any organisation which might benefit financially from a NEO search program.
A major NEO impact is an infrequent but highly destructive event. An impact by a 1 kilometre object could kill 1,500 million people. The chances of such an impact in the next 50 years are about 1 in 2000 (much "better" than the odds of four of a kind in poker). For a low-lying area on the shoreline of an ocean the risk of death from an impact is probably greater than that of an inland location due to the additional hazard of tsunami.
A highly successful Australian search for NEOs was conducted between 1990 and 1996. The program ceased in 1996 when Australian Government funds were cut. We teach our children to look both ways before crossing the road - the decision to cut NEO funding was equivalent to saying "Don't bother looking for cars before you cross the road".
This document proposes the re-establishment of a major Australian NEO search program, as part of the international Spaceguard Survey. The proposed full Australian Spaceguard program would cost between AU$1.3 and AU$3.9 million to implement and about AU$600,000 per annum to run. On the basis of very a conservative estimate, the annual cost works out at about $750 per human life saved (or, more selfishly, $240,000 per Australian life saved).
Based on recent natural disasters in Australia, there is the potential for a $800 saving in direct economic losses in Australia (community, property and business losses) for every dollar spent on the Spaceguard program.
Within a few years the annual operating costs of the Spaceguard program could be funded, at least partially, through private and corporate sponsorship/donations. Government funds (national and international) will be needed to get the program underway.
We have the technology to both detect most of these threatening objects and to avoid or mitigate an impact, provided that sufficient warning is available. If, in a few decades, an imminent impact is detected when it is too late to do anything about it then the finger will be pointed at our generation - we had the ability to provide this gift of survival knowledge to the future but did nothing about it.
Read on...Chapman 1998a, Verschuur 1998).
During the 1980's several programs were initiated around the world to search for NEOs. These gave a better understanding of the size of the NEO population but were generally not sensitive enough to detect a major percentage of these objects. In 1992 the Spaceguard Survey was proposed, in response to a US Congress request to NASA to accelerate the discovery rate (NASA 1992). In May 1998 a US House of Representatives Committee conducted a hearing on the impact hazard. The evidence presented to the committee made a strong case for implementing a global Spaceguard program (Chapman 1998b).
The proposal for reviving the Australian
component of this important program has been under development since
1996, when the Australian Federal Government withdrew funding for the initial
project (which started as the Anglo-Australian Near Earth Asteroid Survey
- AANEAS). The demise of the Australian program
caused great concern among the NEO scientific community as it was the only
major search in the Southern Hemisphere. NEOs discovered in the Northern
Hemisphere might be "lost" if they move into southern skies:
Appendix A contains further information about NEOs.
Early detection of an impending impact by a large NEO would initiate an international effort to deflect the object (the earlier this is done the better the chance of success) and to prepare contingency plans in the event of an impact. In either case civilisation could be saved.
There are many more smaller NEOs which also present a regional hazard on Earth. An object about 100m in diameter can devastate the landscape for a 50km radius around the point of impact (as occurred in Siberia earlier this century). The chances of such an object hitting the Earth in the next 50 years are about 1 in 10 (perhaps much higher - one purpose of Spaceguard is to refine these estimates). The proposed worldwide Spaceguard program is expected to greatly increase the detection rate of these smaller NEOs. Sufficient prior warning of a small NEO impact would enable evacuation of the affected areas and contingency preparations for possible effects on crop production.
The chances of a US citizen dying from an asteroid or comet impact is about 1 in 20,000. This is compared with other causes of death in the US in the following table:
Table 1 - Summary of Accidental Causes of Death for US Citizens
It can be seen that asteroid/comet impact is relatively high in the list compared with other "civil disasters". Although major impacts are very rare the consequences of such impacts, in terms of loss of life, are extremely high. The significant difference is that, given an adequate detection program, a NEO impact can be predicted many years in advance and therefore the consequences can be minimised. In other words, the above table is based on current (minimal) NEO detection programs - the chances of death through NEO impact can be substantially reduced through improved NEO search programs.
On a worldwide basis, it is conservatively estimated that, averaged over many years, the global death rate from NEO impacts would be equivalent to 3,000 deaths per year or about one death per two million of population (see Appendix B). This rate will be used in the cost-effectiveness estimates set out below. Although the value seems relatively low, compared with other causes of death around the globe, the actual deaths from any one NEO impact event could range from zero to most of the population of the Earth - it has been pointed out that the impact of a 1km asteroid could cause the death of 25% of the human population (Steel 1995). Furthermore, the fragile global economy is likely to be affected by a NEO impact in a populated, agricultural or resource rich area.
Another factor not fully covered in this estimate are the consequences of an ocean impact. These consequences have only recently been determined through computer modelling. Tsunami and hurricanes resulting from the impact of a NEO can cause fatalities and damage equivalent to a large NEO striking land. Taking these and other factors into account Steel (1995) derives an annual expectancy of deaths from impacts in the range 3,000 to 15,000. As indicated above, the lower value will be used for cost effectiveness estimates. For comparison, the global long-term average death toll from earthquakes is about 10,000 per year.
The NEO detection program has many parallels with public health disease
prevention programs - early detection can prevent or mitigate the effects
of the threat.
The major NEO detection effort is currently within the US where there are several projects (Spacewatch, NEAT, LONEOS, LINEAR). In July 1998 NASA established a Near-Earth Object Program Office within JPL "to co-ordinate NASA-sponsored efforts to detect, track and characterise potentially hazardous asteroids and comets that could approach Earth". David Morrison from NASA recently reported on the progress of these US programs - about one tenth of the required global discovery rate.
Japan has just allocated a budget of US$15million for NEO-related research including an optical telescope and a radar system.
In Europe the OCA-DLR Asteroid Survey (O.D.A.S.) is located north of Nice, France. It began observing in October 1996.
Asteroid/comet detection is also undertaken by dozens of amateur astronomers throughout the world. Although generally these people do not have access to equipment which would detect the faint objects of most concern they form a very important part of the follow-up of anticipated detections by the Spaceguard program. In addition, there will be a need, from time to time, to use other professional observatories, particularly if recovery/follow-up of an object requires larger telescopes than that used by Spaceguard.
Co-ordination of all asteroid/comet detections (including NEOs) is carried
out by Minor Planet Centre (MPC) based in the USA. It is understood that
the US government intends to increase resourcing for the MPC in order to
cope with an anticipated vast increase in detection rates over the next
With this level of NEO search there would also be a greatly increased chance of detecting longer-period comets, which spend much of their time in the dimly lit outer reaches of the solar system. However, an effective (say 80%) detection program for these objects is much more challenging, as is an effective detection program for the region-threatening small NEOs (under 200m diameter). In either case the sampling by the Spaceguard program would provide important information that would enable these more sophisticated programs to be assessed and ultimately developed.
A key issue in setting goals is the prompt implementation of an effective program. Due to the nature of the orbits of NEOs they are generally only in favourable positions in the night sky for several days every few years. For an object that is visible today it could be several years before an opportunity arises to detect that object again. Furthermore, in order to precisely calculate the object's orbit and determine the long-term hazard to the Earth, it is necessary to have observations a year or more apart. That is why observations made today are so important - they can be used in several years time to redetect the object and make precise orbit calculations. Today's observations could provide a gift of survival information to the future.
The Spaceguard goals are realistic and the technology exists to implement
an effective program immediately.
The limiting magnitude (minimum brightness of a detected object) achievable by a given telescope depends primarily on the aperture (the diameter of the main mirror), the exposure time and the method of detection (photographic plate or CCD - see below). The larger the aperture the shorter the exposure time to detect on object of a given magnitude. Excessive exposure times would mean that the telescope must be pointing at a set spot in the sky for too long and there would be insufficient telescope time to survey the whole of the sky allocated to that telescope. For longer-term follow-ups (such as detecting an object discovered a year or more previously) the exposure times can be extended in order to detect the object of interest, which might be in a less favourable position for observation the second time.
A related issue is the observational strategy. An asteroid or comet generally moves very slowly against the background stars. If two exposures of the same portion of sky are taken 1 hour apart then the motion of the comet or asteroid should become evident. Experience has shown, however, that it is best to make a third exposure an hour later to help eliminate spurious signals (mainly from cosmic rays). This technique will show that an asteroid or comet is present in that portion of the sky. If it is not a previously catalogued object then follow-up observations will be needed over several days in order to provisionally determine the object's orbit (and the possibility of collision with Earth within a few decades). If the follow up observation is missed then there is a strong chance the object will be lost - a missed opportunity.
The necessary exposure times and the observational strategy place limits on the minimum aperture of a Spaceguard telescope. A telescope with a 1m aperture and an advanced CCD detector will be able to detect magnitude 20 objects after exposures of 20 seconds. However, the area of sky covered during a single exposure, coupled with time needed to make the second and third exposures, mean that the telescope would have insufficient free time for follow-up observations and a second telescope would probably need to be employed for this purpose. Given the importance of the follow-up observations, a second 1m telescope would be needed on a near-fulltime basis.
An alternative strategy, currently being implemented at the University of Arizona, is to use a 1.8m telescope, or larger. In this case magnitude 20 objects can be detected after exposures of about 10 seconds. The telescope would therefore have sufficient time to complete its own follow-up surveys of the sky.
Most of the NEO detection efforts in the early 1990s involved the time-consuming and tedious analysis of photographic plates. There has been a recent major breakthrough in technology with the application of Charge Coupled Devices (CCD) to astronomy. This is demonstrated by the recent success of the LINEAR project in the USA. CCDs detect faint light signals on an electronic matrix (the same type of device as that used in video cameras). The resulting electronic image can be directly processed by computer. For example, two images taken an hour apart can be automatically analysed to identify objects, such as NEOs, which have moved over that time (a fast-moving NEO might need to be slightly brighter than the magnitude 20 "target" value in order to be detected by the system).
Once a potential NEO is detected it will need to be checked against
data for previously discovered objects to determine if it is a new discovery.
This will mean that the observer will need computing facilities that can
perform the necessary orbital calculations and that have access to a database
of known objects. In either case the resulting astrometric data will be
forwarded to a central repository (currently the Minor Planet Centre) for
distribution to other NEO observers. The determination of the impact threat
to the Earth could be undertaken by any organisation with access to this
data, although guidelines have recently been developed to cover the public
announcement of the discovery of a potentially threatening object (to avoid
false alarms). Computer programs are now available to analyse CCD data
and calculate NEO orbits.
The cost of constructing a new 2m telescope with CCDs and other equipment
would be around $4 million. An alternative approach is to obtain most of
the equipment from redundant programs - preferably through donation. Possible
sources of equipment are the Automated Patrol Telescope at Siding Spring
and the US Air Force or NASA for the CCD system and computer software.
There would still be substantial costs involved in providing the facilities
(proposed at Siding Spring) and adapting the equipment for the Spaceguard
project. Estimates of these costs are set out below.
10% Contingency - allow $50,000 per year
It would be possible to stage the implementation to optimise the resources. For example the first phase might involve upgrading the Automated Patrol Telescope at Siding Spring for NEO search work. This telescope could then be effectively used as the main telescope in the initial years for an estimated operating cost of about $300,000 per year. The second phase would involve the commissioning of a larger telescope for dedicated NEO search work (this would be needed for detecting fainter objects as the Spaceguard Survey progresses). The estimated costs of $600,000 per year cover the full-time operation of this telescope and the part time operation of a follow-up telescope such as the APT.
The above calculation is based on saving lives around the globe. If the calculation is confined to Australian lives saved then it is estimated that, averaged over many years, 2.5 Australian lives would be saved each year (this is a somewhat selfish approach and ignores the grave economic, social and political consequences of a major impact). The Spaceguard program in Australia would therefore equate to $240,000 per life saved.
This level of cost per life saved is well within the "worth doing" category for road safety initiatives, airline safety programs and medical funding. For example, averaged over the past two decades, the worldwide number of fatalities in non-sabotage commercial airliner crashes is around 700 per year (Reynolds 1992). This is one quarter of the estimated death rate from NEOs under current search efforts. The relatively low rate for commercial airline crashes is primarily due to very effective preventative measures that apply in the commercial airline industry - measures which cost hundreds of millions of dollars per year and equate to millions of dollars per life saved. (Note that the chances of a USA citizen dying in a commercial airline crash are much higher than citizens from most other countries due to higher rates of airline travel in the USA - the value for "Passenger airline crash" in Table 1 only applies to residents of the USA). The annual cost of the Australian component of Spaceguard is probably well below the annual cost of weapon detection systems (including staffing) at just Sydney domestic airport.
These NEO cost effectiveness calculations are very conservative since they are based solely on preventing deaths from major impacts (NEOs 1km and larger). They do not include the benefits of preventing the global economic chaos which would result from a major NEO impact, the prevention or mitigation of the regional effects of smaller NEO impacts (including the potential for many additional deaths resulting from an impact above the ocean) or the protection of our satellite and communication systems through a better understanding of NEOs. Furthermore, the benefits would extend well beyond the ten year period proposed for achieving the Spaceguard goals. On the other hand, mitigation measures such as deflecting asteroids or creating global food and energy stores have not been costed but these could be expected to be much less than the potential economic losses.
Based on recent natural disasters in Australia, it is estimated that
the Australian component of an effective Spaceguard program would be associated
with direct economic savings of about $500 million per year - more than
eight hundred times the annual cost of the program (see Appendix
B). This estimate is likely to be very conservative since the physical
and economic effects of cyclones, floods and earthquakes tend to be confined
to regions and, to date, they have not seriously disrupted the Australian
Start-up costs could be shared between the Australian and US Governments (including the value of US agency-donated assets).
In the first year the operational costs could be met by the following
We now have the technology to detect a significant proportion potentially threatening objects. If an object is found to be on a collision course with the Earth in the foreseeable future then we can take action to avoid the collision or to mitigate the effects of an impact. If we do not look for these objects then an impact is likely to occur without warning and the consequences will be much graver.
There is an urgent need for a Southern Hemisphere component of the international Spaceguard program. Australia is ideally placed for this task, with a suitable location and experienced astronomers. Is is estimated that it would cost $1.3 million to establish an effective NEO search program in Australia (assuming the donation of some redundant assets) and the operating costs would be $600,000 per year, over the proposed 10 year Spaceguard program. Within a few years the annual operating costs of the Spaceguard program could be funded, at least partially, through private and corporate sponsorship/donations. Government funds (national and international) will be needed to get the program underway.
The NEO search is clearly a public health issue. Averaged over many years, the global death rate from NEO impacts is conservatively estimated to be 3,000 per year under current (minimal) NEO search programs. This could be cut by 2,400 per year through the proposed worldwide Spaceguard program (assuming that 20% of large NEOs remain undetected). In effect, the Australian component would contribute to about one third of the lives saved or about 800 lives per year. With an annual cost of $600,000 this is equivalent to $750 per life saved - a remarkably cost-effective program. Even if the costing is confined to Australian lives then the cost of the Australian program is equivalent to $240,000 per life saved - this compares very favourably with road safety and medical prevention programs. Based on recent natural disasters in Australia, there is the potential for a $800 saving in direct economic losses in Australia (community, property and business losses) for every dollar spent on the Spaceguard program.
Let us not wait until a catastrophic impact shocks to world into conducting a serious search for NEOs. In any case, the environmental and economic consequences of even a moderate impact could set civilisation back for decades or centuries, paradoxically making a NEO search a lower priority than the rebuilding of civilisation.
By conducting a serious NEO search now we can provide future generations
with otherwise unobtainable information which may prove crucial to their
survival - a rare gift to the future from the people of the 20th Century.
Chapman (1998a) The Asteroid/Comet Impact Hazard, Originally presented at the Workshop on Prediction in the Earth Sciences, Boulder CO, 10 July 1997. Updated 22 April 1998. http://www.boulder.swri.edu/clark/index.html>
Chapman C (1998b) The Threat of Impact by Near-Earth Asteroids, Statement before the Subcommittee on Space and Aeronautics, Committee on Science, US House of Representatives, May 21 1998. http://www.boulder.swri.edu/clark/hr.html. Also Action Plan Statement, 9 June 1998: http://www.boulder.swri.edu/clark/actnea.html
Hodges A (1997) 'Disasters and disaster issues - the Australian Experience', Australian Insurance Law Association National Conference, 14 August 1997. http://www.ema.gov.au/insclawc.html
Martel L (1997) Damage by Impact, Hawaii Institute of Geophysics and Planetology, http://www.soest.hawaii.edu/PSRdiscoveries/Dec97/impactBlast.html
Morrison D (1997) 'Is the sky falling?', Skeptical Inquirer, May/June 1997. http://www.csicop.org/si/9705/asteroid.html
NASA (1992) The Spaceguard Survey, http://ccf.arc.nasa.gov/sst/spaceguard/index.html
NASA (1996) Responding to the potential threat of a NEO impact, AIAA Position paper. http://impact.arc.nasa.gov/reports/aiaa/index.html
NASA (1995) The NEO Survey Workgroup Report, http://impact.arc.nasa.gov/reports/neoreport/executive.html
NRC (1998) Impact Cratering on Earth, Natural Resources Canada, (includes maps and database) http://gdcinfo.agg.emr.ca/paper/cratering_e.html
Steel D, McNaught R, Garradd G, Asher D and Russell K (1998) 'AANEAS: A Valedictory Report', Australian Journal of Astronomy, June 1998. http://www1.tpgi.com.au/users/tps-seti/spacegd4.html
Verschuur G (1998) 'Impact Hazards: Truth and Consequences', Sky and Telescope, June 1998. http://impact.skypub.com
Young R, Bryant E, Price D and Spassov E (1995)
'The imprint of tsunami in quarternary coastal sediments of Southeastern
Australia', Bulgarian Geophysical Journal, Vol.XXi, No.4.
Based on the relatively small sample of the NEO population by astronomers, the expected impact rate for these large NEOs is one every 100,000 years (thus the 1 in 2000 chance of an impact in the next 50 years). On the 15% of the Earth's surface that would have retained evidence of such an impact the average interval between events would therefore be about 700,000 years which is in reasonable agreement with the independent (and incomplete) observation of cratering on Earth (3 impacts causing craters over 10km diameter in the last 4 million years - NRC 1998). However, it is wrong to think that the "next major impact" will not occur for tens of thousands of years and is therefore not of current concern. It could occur at any time - there is a 1 in 100,000 chance that it will occur next year and the same odds in any subsequent year. See Poisson Distributions for more information about interpreting impact rates.
Craters from impacts with regional significance have been found in Algeria (100,000 years ago), Arizona (50,000 years ago), India (50,000 years ago), South Africa (200,000 years ago), Argentina (100,000 years ago) and Western Australia (300,000 years ago). Again this cratering rate is in reasonable agreement with estimates based on the known NEO population. Furthermore, as in the case of Tunguska, not all impacts with regional significance leave an impact crater.
The usual causes of tsunami are earthquakes and underwater landslides but there is now evidence that giant tsunami can be caused by NEO impacts. Tsunami and hurricanes resulting from the impact of a NEO can cause fatalities and damage equivalent to a large NEO striking land.
Research by the University of Wollongong suggests that the New South Wales South Coast has been struck by at least six large tsunami within the last 6,000 years - a typical interval of 1,000 years - perhaps much less (Archer 1998, Young et al 1995). One possible cause is giant underwater "landslides" on the edge of the continental shelf but NEO ocean impacts may have caused some (or all) of these tsunami.
Although we probably cannot detect most of the smaller NEOs with current technology the threat of tsunami from the smaller impacts could be mitigated by the installation of tsunami warning systems, as used in Japan and Hawaii.
The explosion of a small asteroid several kilometres above the Earth's surface is likely to throw tens of thousands of tonnes of debris into near space where it would pose a threat to spacecraft.Chapman 1998a) the number of US deaths from NEO impacts is estimated to be 115. This is equivalent to 0.5 NEO deaths per million of population. Applying this to the world population of about 6 billion gives an estimated death rate from NEO impacts of 3,000 per year. For comparison, the global long-term average death toll from earthquakes is about 10,000 per year.
An indication of the scale of direct economic losses of an impact can be gained from Australia's recent natural disasters: "Australian citizens rightly expect prevention of, or at least protection from, disasters which affect life and property. But though our disasters have caused relatively few deaths, very substantial economic losses are often sustained. For instance [in 1996] there were 23 events nationally with individual losses exceeding $5 million causing total losses of $1.3 billion." (Hodges 1997). The 1989 Newcastle Earthquake caused 15 deaths, 150 injuries and seriously damaged 33,000 buildings. The total estimated cost of the disaster was $4 billion including insurance losses of almost $1 billion. Using these events as a very rough rule of thumb, the estimated 2.5 Australian lives saved per year through an effective Spaceguard program would be associated with direct economic savings of about $500 million per year - more than eight hundred times the annual cost of the program.
The above analysis does not take into account the extra hazard to Australia's coastal areas from impact-generated tsunami. On the basis of the limited research into NEOs and tsunami to date, the death toll from a 10 metre tsunami striking the east coast of Australia is estimated to be at least 35,000.
An alternative approach could be considered for raising funds from the private sector: people and corporations could pay to have an Spaceguard-discovered asteroid named after them. This would involve a radical change to the way that asteroids are named* and would require international agreement. However, under Spaceguard there will be a huge increase in the discovery rate for "main-belt" asteroids, in addition to NEOs and comets and therefore a rethink about naming procedures will probably be necessary. The intention is that sponsored names would only apply to objects discovered through the Spaceguard program - in fact this proposal could be used to raise funds for the international Spaceguard effort.
*Note: The current practice is to name comets after their discoverer. Discoverers of asteroids can submit a name (not their own) for consideration.
There could be grades of sponsorship which depend on the "importance" of the object: the heirarchy might be: large NEOs, large main-belt asteroids, small NEOs, small main-belt asteroids. A possible sponsorship scenario for the Australian component of the Spaceguard program might be:
Type of object
Est.discovery Naming fee
Income per yr
rate per yr
Large NEO 100 $500 $50,000
Large MB Aster. 300 $200 $60,000
Small NEO 500 $100 $50,000
Small MB Aster. 2,000 $20 $40,000
Both the discovery rate and the possible naming fees are approximate
at this stage but they indicate that there is potential to cover at least
some of the Spaceguard annual budget through this mechanism. After all,
many people pay several hundred dollars for a headstone when they die -
a named asteroid would last much longer than a headstone. Care would, of
course, need to be taken to ensure that any sponsored object was not going
to impact the Earth, at least within the next fifty years.
The Australian Government should establish an Australian Spaceguard Working Party to address the following issues:
He has an amateur interest in planetary science. He is a member of The
Planetary Society (TPS) and is NSW Co-ordinator of the TPS
Australian Volunteer Coordinators.
He first became aware of the demise of the Australian NEO search program when TPS Executive Director Louis Friedman wrote to Australian TPS member in 1996. Michael then wrote to his local Member of Parliament (Bronwyn Bishop - the Minister for Defence Science) and, on receiving a poorly researched reply, subsequently wrote to several other Australian Ministers - the saga has continued from there!
This document has been in preparation for several months and Michael is grateful for the advice and comments provided by NEO scientists from around the globe. Several recent events have made it important to progress this issue: