The Space Reviewin association with SpaceNews
 

NSRC 2020

 
asteroid impact illustration
The threat of asteroid impacts is often misrepresented in the general media because of the difficulties understanding impact probabilities. (credit: NASA)

<< page 1: asteroid turns hot potato

2004 VD17, impact assessment scales, and a natural impact warning clearinghousee

So Rick Binzel (CCNet, Mar. 7) is happy enough to have 2004 VD17 continue to languish at Torino Scale 2 (or just as inconsequentially for the public as if it were at any other value from zero to seven), while Tim Roberts (CCNet, Mar. 9) is rather understandably concerned about the general problem of communicating and translating relevant information about possible NEO impacts from astronomers to policy makers who need “to develop a rational response, especially if that response is ‘act casual, do nothing’”.

And languish 2004 VD17 does, despite the publication of potential precovery positions from 2002. Sure, these 2002 images are so very weak that doubt can be expressed about their reality, but for the JPL and NEODyS webpages to ignore them, when they are being mentioned in the Asteroid/Comet Connection and the Minor Planet Mailing List, seems somehow to invite the idea of a “conspiracy” to hide something the public might not wish to hear.

There is absolutely no need for the present generations of humanity to worry about 2004 VD17. There are plenty of opportunities for astronomers to continue to make observations between now and 2102, and it is very likely indeed that, sooner or later, the potential danger will disappear.

The point is that our knowledge of the orbit computed from observations of 2004 VD17 since its discovery in November 2004 is so very good that we can predict almost precisely where it was on those fortuitous 2002 NEAT frames, and this provides an incentive to imagine that one can actually detect faint traces of the object. Sure enough, a couple of people have “independently” measured the possible detections. What—if anything—does this do to the assessment of the 2102 impact possibility? One set of measures is so very close to the nominal prediction that one could reasonably conclude that the potential impact event is still “on”; furthermore, since the alleged 2002 measures would increase the length of time spanned by the observations by a factor of three, one could anticipate a rather larger impact probability, with consequent increases also in the Palermo Scale and perhaps also the Torino Scale values. On the other hand, another set of measures suggests a 0.5-arcsec shift, say, in the object’s 2002 position. This will of course lead to a further change in the computation of the random number (currently advertised as 1-in-1300) that is the 2102 impact probability—perhaps to the extent that the whole impact possibility will go away.

Either way, there is absolutely no need for the present generations of humanity to worry about 2004 VD17. There are plenty of opportunities for astronomers to continue to make observations between now and 2102, and it is very likely indeed that, sooner or later, the potential danger will disappear.

I have suggested on several occasions during the past two and a half years that a better—and more understandable—assessment of potential NEO hazards is provided by the ratio of the time interval spanned by the observations to the time remaining until the first possible impact date. For 2004 VD17 and the 2102 date, this “Purgatorio Ratio” (PR) is 0.014, increasing to 0.042 if one accepts the 2002 precovery observations and the 2102 threat continues to be valid. Even the 0.042 value is smaller than the PRs of 0.062 for (99942) Apophis in 2036 and 0.063 for (29075) 1950 DA in 2880. The next smaller values of PR are 0.022 for 2000 SG344 in 2068 and 0.011 for 2003 DW10 in 2046. Whatever one may want to think of these numbers, it is pointless even to consider PR values that are less than 0.01, which is the case for all the 95 other objects that currently register on the JPL risk page.

Okay, so “currently” is an important word, if only because the PR values increase as we approach one of those possible impact dates. Consider, for example, the object 2004 FU162. As I write this—on the Ides of March—its PR value is 0.002. Yet in less than two weeks, on March 28, it will rise to 0.01, apparently up and out of what I considered in the previous paragraph to be the “pointless” range. On April Fool’s Eve, I think for the first time, we actually reach an unrescinded date for a potential impact! The impact probability is two in a million, and there’s another chance for a whack at us the next day, on “FUel’s day” itself. Sure, 2004 FU162 is almost certainly too small to allow it to reach the Earth’s surface, but what if it were larger? Should we be worried? Indeed, what will happen to the impact risk pages, other than the appearance of dramatically increasing Palermo Scale values (perhaps), and… then?

It’s actually very likely that 2004 FU162 will be 100 million kilometers or more away at the appointed time. We really have no idea where it is, because we had it under observation—on April Fool’s Eve two years ago—for only some 45 minutes! On the other hand, we do know for sure that 2004 VD17 will be within a million or two kilometers of the Earth on 2102 May 4, and that makes for a different situation. The JPL risk page now partially recognizes this sort of thing with its separation of objects into those that have been recently observed (within the past 60 days) and those (like 2004 FU162) that have not. The real point is that there is a need to separate those objects that can be observed (or could be, if we put our minds to it), because we know rather accurately where they are, from those that are quite lost, and will remain so unless there is an entirely accidental rediscovery.

So while most astronomers understand that, when it comes to assessing impact predictions, it makes a difference whether the object is or is not lost, it is my impression that the public has little appreciation of this point. But after all, if the object is not being observed, any meaningful computation of the probability of its impact on some specific date is not going to change until the date arrives and we can ascertain that it does not seem to have hit.

After all, if the object is not being observed, any meaningful computation of the probability of its impact on some specific date is not going to change until the date arrives and we can ascertain that it does not seem to have hit.

A suggested quantitative way of expressing this observability is to use what I shall now call the “Paradiso Parameter” (PP). This is the ratio of the time interval spanned by the observations to the time that has elapsed since the first observation was made. For an object still under observation, like 2004 VD17, the PP is therefore essentially unity. 1950 DA and Apophis currently have PP = 0.98 and 0.94, respectively. Of course, these two objects are going to be deliberately reobserved, and then we’ll have PP = 1.0 again. But a PP of unity is not paradise, given the purgatory of a possible impact during the next century. Paradise requires a small PP, like, say, the 0.00004 value for 2004 FU162. With PP = 0.20 and 0.14, 2000 SG344 and 2003 DW10 are in an intermediate range, the PP being small enough while the PR is itself quite small. With nothing happening but the passage of time, PR increases (generally rather slowly) and PP decreases (typically more rapidly). By 2028, for example, we can expect the PR for 2000 SG344 to have risen only to 0.034, whereas the PP will likely be down to 0.05. But in that year we can expect more observations, so the PP will presumably jump to 1.0. Of course, the PR could also then jump to a rather worrisome 0.7, although it is again probable, not only that the 2068 impact possibility will disappear, but that all others during the foreseeable future will too. So the PR also disappears.

At what point does the combination of PR and PP cause us distress? That is the question, for one would want such a combination also to yield an impact probability well on the way to unity. But the need to consider mitigating action depends also on other factors, such as further opportunities to make clarifying observations, the possibility of using an earlier close approach to the Earth to accomplish the desired deflection using a probe that can expend less energy, etc. Returning to 2004 VD17, I note that there are particularly good opportunities for observations in 2032 and 2041. If the 2102 impact possibility still exists at the end of 2041, the combination of PP = 1.0 and PR = 0.6 might provide a sufficient “Inferno Index” (so to speak) for thinking then about evasive action. After all, we should still have 60 years to accomplish the task. On the other hand, because of the infrequent opportunities for observation and the desirability of utilizing the 2029 close approach, if action seems necessary in the case of Apophis, this might start with an II of PP = 1.0 and PR = 0.4 following the observations in 2013. Combined also with a frighteningly high impact probability (of several tenths, perhaps?) and an estimate of mitigation energy (rather than the energy of the Earth impact we are going to avoid and that goes into the calculation of the PS and TS), an appropriately high Dantean combination would be a simple but useful tool for the policy makers Tim Roberts wants to advise.


Home