Planetary Defense

Planetary defence is the only member of the catastrophic-risk family where the hazard can be enumerated object by object, the countermeasure has been demonstrated in flight, and the remaining gap has a name. That combination makes it unusually tractable and unusually easy to overstate. The framing under test on this page is that planetary defence is a solved capability.

Established It is the most nearly-true framing in this category and it is still wrong, in a way worth spending a whole brief on. One deflection technique has been demonstrated once, against one object, of one size class, in a binary system that made the result measurable, with years of warning — and the efficiency of that demonstration depends on a bulk density nobody has measured. Everything else in the chain is unsolved in a different way for each link.

Established The chain has four links and they should be kept apart throughout: find, characterise, decide, act. Find is a budget line and a denominator problem. Characterise is what Hera is flying to do. Act is demonstrated for one case. Decide has no owner at all — no named body anywhere can commit a deflection mission, and that absence is as well-attested as the period change.

Established The single most useful correction this page can make is to the headline number. The measured quantity from DART is a velocity change: Dimorphos's along-track orbital velocity fell by 2.70 millimetres per second, with a stated uncertainty of 0.10. The momentum enhancement factor β is that measurement divided by an assumed mass. The mass is a guess with a factor-of-two spread, and β inherits the spread. Frontier The honest statement of DART's efficiency is “somewhere between two and five times the incident momentum, depending on how dense the rubble pile turns out to be.”

Established And the number that reorganises the survey half of the subject. The Vera C. Rubin Observatory began full survey operations on 30 June 2026. It is the most capable ground survey ever built, and it is estimated to reach 62% completeness for the 140-metre-and-larger population — against a congressional mandate of 90% set in 2005 and missed in 2020. The gap between 62% and 90% is what NEO Surveyor is for, and it is why the survey constraint on this page is a funding token rather than a telescope-design token.

Established The demonstration, with its error bars. DART launched on 23–24 November 2021 from Vandenberg on a Falcon 9 and struck Dimorphos on 26 September 2022. Dimorphos is about 160 metres across; its primary, Didymos, about 780 metres. Impact speed was roughly 6.6 km/s, spacecraft mass at impact about 500 kg of a 610 kg launch mass. Established The measured result: the orbital period of Dimorphos about Didymos decreased by 33.0 minutes, with a 3-sigma uncertainty of 1.0 minute, from an initial period of about 11 hours 52 minutes. Two independent methods — ground-based lightcurve photometry and radar ranging — returned the same value. The pre-impact prediction, had momentum transferred with no enhancement, was about 7 minutes.

Established The paper to name when saying deflection is demonstrated is Daly, Ernst, Barnouin, Chabot, Rivkin and 139 others, in Nature. A news story about DART is not that paper. The period-change determination is Thomas and colleagues; the momentum accounting is Cheng, Agrusa, Barbee and colleagues. Established Ejecta dominated the transfer — significantly more momentum reached Dimorphos from escaping impact debris than arrived with the spacecraft — and the lowest-speed ejecta formed a sustained tail morphologically consistent with previously observed active asteroids.

Established Now the conditional, which is the sentence the previous version of this page did not carry. Cheng and colleagues measured an instantaneous along-track velocity reduction of 2.70 millimetres per second (uncertainty 0.10) and derived a momentum enhancement factor of β = 3.61, with an upper bound of +0.19 and a lower bound of −0.25 — conditional on assuming equal bulk densities of 2,400 kg/m³ for both bodies. Established Across the plausible density range of 1,500 to 3,300 kg/m³, the same paper gives β spanning 2.2 to 4.9. Frontier A brief that quotes 3.61 without that conditional has reported a derived number as a measurement. The velocity change is the observation; the mass is an assumption; β is the quotient.

Established Which is precisely what Hera is for, and it is not a victory lap. ESA's Hera launched 7 October 2024 at 14:52 UTC on a Falcon 9 from SLC-40, performed a Mars gravity assist on 12 March 2025 within 5,000 km of the surface, and arrives at Didymos in November 2026 — about a month earlier than originally planned, roughly four years after impact. Construction contract: €129.4 million, signed September 2020. It carries twelve payloads including two Asteroid Framing Cameras, JAXA's TIRI thermal infrared imager, the PALT micro-lidar altimeter and the HyperScout-H hyperspectral imager, plus two CubeSats: Juventas, carrying a low-frequency radar and a gravimeter for internal structure, and Milani, carrying a hyperspectral imager and a dust detector. Established The quantity Hera exists to supply is the mass. Determining the masses of both bodies to high accuracy collapses the density range, and collapsing the density range collapses β from an interval to a number.

Frontier So the correct framing of the four-year gap is this: between September 2022 and the Hera results, planetary defence possesses a demonstrated deflection and an unmeasured efficiency. You can say “we moved it.” You cannot yet say “and here is how hard you must hit the next one.” Frontier Interest to mark: ESA's own statement is that Hera “turns NASA's DART impact into a well-understood and repeatable planetary defence technique.” That is the agency flying the mission describing its own importance. The independent support for the claim is the density spread in the Cheng paper, which is a third party's arithmetic.

Established The survey picture, and the honest version is stronger than the usual one. The 2005 George E. Brown Jr. Near-Earth Object Survey Act directed NASA to detect, track and characterise 90% of near-Earth objects of 140 metres and larger by 2020. The deadline passed unmet; no dedicated funding accompanied the mandate; the Planetary Defense Coordination Office continues toward the goal with no statutory deadline. Established As of 30 December 2024, about 867 near-Earth asteroids larger than 1 km are catalogued against population estimates of 900 to 981 — roughly 90% complete at the kilometre scale. At 140 metres, about 11,167 are known.

Frontier And the denominator is the problem. Published estimates of the 140-metre-and-larger population are 13,200 (2011), 27,100 (2017), under 20,000 (2021) and about 35,000 (2023) — a spread of roughly a factor of 2.7. Dividing the count by each in turn gives implied completeness of about 85%, 41%, over 56% and about 32%. Frontier Those four percentages are arithmetic on published estimates, not findings, and they are presented here as arithmetic. The honest sentence is that we do not know how complete the 140-metre catalogue is, because we do not know the denominator to better than a factor of about 2.7 — which is a much more interesting statement than “fewer than half.”

Established Rubin is now operating and its contribution is bounded. System first-light images were released on 23 June 2025; full survey operations began 30 June 2026; construction cost was about $680 million. Its expected haul is around 100,000 new near-Earth objects among more than five million asteroids. Frontier Rubin alone is estimated to reach 62% completeness for the 140-metre population. The most capable ground survey ever built, in full operation, does not on its own close a twenty-year-old mandate.

Established NEO Surveyor is the intended remainder, and its history is a funding history. NASA's mission page gives launch no earlier than September 2027, with the spacecraft under construction and critical design review passed on 11 February 2025. Design: a 50 cm aperture infrared telescope in two bands, 4–5.2 and 6–10 micrometres, at Sun–Earth L1, about 1,300 kg launch mass, a 12-year planned survey, expected to add 200,000 to 300,000 objects down to about 10 metres. Infrared is the point: dark objects reflect little sunlight but glow when warmed, and an L1 platform works closer to the Sun's direction than any ground survey can. Established The instrument concept competed unsuccessfully for NASA Discovery Program selection in 2006, 2010, 2015, 2016 and 2017 — five losses in eleven years — before the Planetary Defense Coordination Office approved full funding on 23 September 2019, outside NASA's science budget, on national-security grounds. Reported cost figures range from a $500–600 million estimate to a $1.2 billion development baseline; the authoritative figure is the agency baseline commitment, which this rewrite could not open.

Frontier Read the five Discovery losses correctly. They are not evidence that the science was weak. They are evidence that a survey instrument competing against science missions inside a science programme loses, because its output is a catalogue rather than a discovery. The mission advanced only once it was reclassified out of that competition. That is what “survey capacity is a funding line” means in practice.

Established Chelyabinsk calibrates the part of the hazard the mandate does not cover. On 15 February 2013 an object of roughly 17 to 20 metres and about 10,000 tonnes entered over Chelyabinsk Oblast and airburst at high altitude — peak brightness placed at about 23.3 km in the first analyses and about 29.7 km in later work. Energy: a peer-reviewed best estimate of 460 to 470 kilotons TNT equivalent, against preliminary NASA figures of 400 to 500. 1,491 people sought medical treatment, essentially all from indirect effects — overwhelmingly glass driven by the blast wave. Established It was not detected in advance. The object approached on a radiant close to the Sun; no survey saw it; the first notice was the fireball. The primary literature is Popova and colleagues in Science and Brown and colleagues in Nature, both 2013, neither of which this rewrite was able to open directly.

Frontier The structural point deserves saying flatly. A 140-metre survey mandate says nothing about a 17-metre object, and a 17-metre object sent fifteen hundred people to hospital. The sunward blind spot is geometric: ground-based optical surveys cannot look where Chelyabinsk came from. An infrared telescope at L1 improves this and does not eliminate it. The class of object most likely to hurt someone in any given decade is the class the mandate does not cover.

Established 2024 YR4 is the best end-to-end test the system has had, and it deserves the dates. Discovered by ATLAS at Río Hurtado, Chile, on 27 December 2024. Rated Torino 3 on 27 January 2025 — the second-highest rating ever assigned. IAWN notification issued 29 January 2025, 33 days after discovery. Peak Earth impact probability 3.1% on 18 February 2025 for 22 December 2032, Palermo Scale −0.18; ESA's independent assessment the same day was 2.8%. Torino downgraded to 0 on 23 February. NEODyS ruled out Earth impact on 1 March, ESA on 8 March, NASA on 2 April on a 91-day observation arc. JWST observed on 8 and 26 March and 11 May 2025; the 26 March measurement gives a diameter of 60 metres, uncertainty 7 metres. Established Then the risk moved to the Moon: lunar impact probability peaked at 4.3% on 11 May 2025, was refined by further JWST observations on 17 and 26 February 2026, and reached zero on 5 March 2026, with a predicted miss distance of 21,200 km from the lunar surface, uncertainty 2,300 km.

Established Two details from that episode are worth more than the headline. First, NASA and ESA computed 3.1% and 2.8% independently on the same day — two agencies converging within the uncertainty is a genuine institutional strength and should be said as such. Second, the system's own output was non-monotonic: the risk fell for Earth, rose for the Moon, and fell again ten months later. Frontier A size discrepancy worth carrying rather than resolving: NASA's page describes the object as approximately 50 metres across from February 2025 Webb data, while the March 2025 Webb measurement is quoted at 60 metres with a 7-metre uncertainty. Both are Webb-derived. Give the range and attribute each figure.

Frontier The verdict on 2024 YR4 is the spine of this brief. It establishes that detection, dual-agency computation, notification and public communication work: 33 days from discovery to formal IAWN notification, 96 days from discovery to Earth-risk elimination. It establishes nothing whatever about decision, because the probability went to zero before anyone had to decide anything. It was a full-dress rehearsal of every act except the one nobody has scripted. Speculative The counterfactual, flagged as one: had the probability risen past 10% rather than falling, the next required act would have been for some body to commit a deflection mission, and no named body could have done that.

Established The open questions in this subject sort cleanly into the four links, and the first one has a delivery date. Hera arrives at Didymos in November 2026. Its measurement of Dimorphos's mass is the single question whose answer is scheduled.

Frontier Question one: does β near 3.6 generalise? Cheng and colleagues published the value and the planetary defence community works with it. The paper itself gives the 2.2–4.9 range across plausible densities, so the value is conditional by its authors' own statement. Whether a number derived from one rubble pile transfers to the next rubble pile is a separate question again, and nothing in the record answers it.

Frontier Question two: how complete is the 140-metre catalogue? Four published population estimates span a factor of 2.7 and the field has not converged. Published completeness figures range from roughly 25% to 50% depending on year and method. Frontier Whether Rubin plus NEO Surveyor together close the 90% mandate is an open question and must not be asserted: Rubin alone is estimated at 62%, and no combined figure was obtainable for this rewrite. Anyone quoting one should be asked for its source.

Established Question three, and it is not a research question at all: who decides. Two UN-endorsed bodies exist, both established in October 2013 under COPUOS following the Action Team-14 recommendations. IAWN, the International Asteroid Warning Network, is a clearinghouse for shared information, with a public notification threshold of impact probability above 1% within 20 years for objects larger than 10 metres. SMPAG, the Space Missions Planning Advisory Group, coordinates joint studies of deflection technologies and provides oversight of missions. Established IAWN's verb is inform. SMPAG's verbs are study and advise. Neither can commit a mission.

Frontier Question four: which technique for which object? Kinetic impact is demonstrated and works best against small, solid bodies — less well against loosely aggregated rubble piles, which is what Dimorphos appears to be. Gravity tractors work irrespective of composition, spin state or structure, which are exactly the properties we are worst at measuring in advance, and they have the lowest technical readiness and the highest assessed cost of the mainstream options. Ion beam shepherds are at concept stage. Nuclear standoff options are assessed at ten to a hundred times the effectiveness of non-nuclear alternatives.

Frontier Question five: is disruption ever the right answer? Work from 2019 indicates asteroids may be substantially harder to destroy than earlier thought, because a disrupted body can gravitationally reassemble. Earlier supercomputer simulations from 2011–2012 found that with sufficiently high and well-matched energy delivery, fragments do exceed escape velocity rather than recoalescing. Frontier These results are not in contradiction; they bracket a threshold, and this page presents them as a threshold problem rather than as a controversy.

Frontier Question six: is the sub-100-metre population the one that actually matters? The Chelyabinsk facts are settled — a 17-to-20-metre object, undetected, 1,491 people treated. The generalisation that sub-100-metre objects dominate the near-term casualty risk is plausible and was not sourced for this rewrite, so it is carried as a hypothesis and not as a finding.

Frontier Question seven, which is political rather than technical: can deflection be delegated at all? Rusty Schweickart's objection is the cleanest statement of why not. During a slow deflection the most likely impact point drags across different countries' territory, so somebody has to decide whose sky carries the residual risk while the operation runs. Speculative Whether IAWN and SMPAG would, in a real event, converge on a recommendation that states then acted on has never been tested, and the fetched record is explicit that as of 2025 there are no firm commitments from spacefaring nations regarding response protocols.

Handwave And the framing itself, stated as the claim it is. “Planetary defence is a solved capability” is a common reading of the DART result. One technique, one target, one size class, years of warning, an unmeasured efficiency and no decision authority. The claim does the work by assertion at the word solved.

Established The first bottleneck is warning time, and the arithmetic is what makes the institutional gap bite. Most deflection approaches require from a year to decades of notice. NASA testimony in 2013 gave at least five years of preparation before a mission could launch. The 2021 Planetary Defense Conference tabletop exercise concluded that five to ten years of preparation may be needed to avoid a virtual impactor. Frontier Set that against the record. 2024 YR4 was discovered in December 2024 for a 2032 date — about eight years, inside the five-to-ten band with almost no margin, for an object found unusually early. Chelyabinsk gave zero.

Established The second bottleneck is the denominator, not the telescope. Detection hardware is not the limiting factor at 140 metres; Rubin exists and works. What limits the field is that the population estimate against which completeness is measured is uncertain by a factor of 2.7, so the completeness number itself has an error bar wider than the policy debate about it. Improving that requires more sky, more time and more infrared — all of which are appropriations.

Established The third bottleneck is characterisation, and it is the one Hera addresses. Deflection design needs mass, porosity, internal structure and spin state. None of these is available from a discovery observation; all of them require either a dedicated rendezvous or a radar pass. We currently possess a demonstrated technique whose scaling parameter is bracketed between 2.2 and 4.9 because we do not know one asteroid's density.

Established The fourth bottleneck is the one with no engineering content: decision authority. Every deflection mission that has flown or been funded — DART and Hera — was authorised through an ordinary national agency budget process, on a decade-scale timeline, against a target chosen for convenience rather than threat. Established NASA's Planetary Defense Coordination Office, established January 2016 within the Planetary Science Division, catalogues, tracks, funds surveys and coordinates response planning. There is no described mechanism by which it launches anything on its own authority.

Frontier The fifth bottleneck is rapid-response launch and deep-space tracking, and it belongs to a neighbouring brief. A deflection mission is a spacecraft; spacecraft consume antenna hours and launch slots that Deep Space Infrastructure shows are already contended, with a documented case of one uncrewed test flight degrading a flagship observatory's operations. This page needs only the warning-time consequence and hands the capacity argument across.

Speculative And a sixth that is genuinely unsolved rather than merely unfunded: the short-warning case. Nuclear standoff options are the only family credible on short notice, assessed as effective against 100-to-500-metre objects with about two years of warning and larger objects with at least five. Anything shorter than that has no demonstrated answer, and the concept-study claims at the aggressive end of the literature are exactly that.

Established This brief records no dependency on another brief, and that is a finding rather than an omission. Nothing in planetary defence waits on a physics result, a new material or a demonstration that some other page produces. The technique is flown. What the field waits on is a survey budget and a body that can decide, which is why the adjudication here is constraints-only.

Established The first standing requirement is survey capacity, and it is measured in appropriations. The mandate is twenty years old and was missed. Rubin reaches an estimated 62%. NEO Surveyor lost five Discovery competitions before being funded outside the science budget on national-security grounds, and its launch is no earlier than September 2027. Every number in that sequence is a budget decision, not an engineering one.

Established The second is a deflection decision authority that can act inside the warning time. IAWN informs; SMPAG studies and advises; the Planetary Defense Coordination Office catalogues and coordinates; national agencies authorise spacecraft on decade timescales. No body can commit a deflection mission, and the preparation times in the literature are comparable to or longer than the warning that a realistic case would give.

Frontier Two dependencies are deliberately handed away. Existential Risk Governance owns the general question of how a society decides under low-probability, high-consequence uncertainty — the decision theory, the discounting, the institutional design. This page owns the asteroid case, and owns it separately because the asteroid case is the one member of that family where the hazard is enumerable, the countermeasure is demonstrated, and the missing piece is a named institutional gap rather than a conceptual one. Frontier Space Law and Governance owns the general observation that no forum has compulsory jurisdiction over space activities; this page supplies the sharpest instance of it and the warning-time arithmetic that makes the absence load-bearing.

Frontier And one shared instrument heritage rather than a dependency. Hera's radar, gravimetry and hyperspectral imaging are the same measurements a prospector wants. Asteroid Mining and Space Resource Economies measure an asteroid to value it; this page measures one to hit it correctly. Same instruments, different purpose, and no edge is claimed on that basis.

Established The decisive experiment of the last decade has already run and its result is a number with an uncertainty. DART struck Dimorphos on 26 September 2022; the period fell by 33.0 minutes with a 3-sigma uncertainty of 1.0 minute, against a no-enhancement prediction of about 7 minutes; two independent measurement methods agreed. That is the experiment this whole subject rests on, and it worked.

Established The next decisive experiment has a date. Hera arrives at Didymos in November 2026 and will determine the masses of both bodies, measure the DART crater, and characterise Dimorphos's internal structure and porosity. Frontier The specific result to watch for is a bulk density. A density near the assumed 2,400 kg/m³ leaves β near 3.6; a density at either end of the 1,500–3,300 range moves it toward 4.9 or 2.2. Until that number exists, every deflection design carries a factor-of-two uncertainty in how hard to hit.

Established An unplanned experiment ran in parallel and is worth as much as a designed one. 2024 YR4 exercised discovery, independent dual-agency orbit determination, IAWN notification within 33 days, escalation to Torino 3, public communication at a 3.1% peak, JWST characterisation, and de-escalation to zero — first for Earth, then, ten months later, for the Moon. Frontier Its result is a pass on every link that was tested and no data at all on the one that was not.

Frontier The experiments that have not been done are enumerable. No deflection has been attempted against a solitary body on a heliocentric orbit — DART's result was measurable because Dimorphos orbits Didymos, which is a far easier measurement problem. No gravity tractor has flown. No ion beam shepherd has flown. No nuclear deflection option has been tested in any form, and none can be without addressing a separate treaty question.

Frontier The institutional experiment is the tabletop exercise, and its outputs are honest about their own limits. The 2021 Planetary Defense Conference exercise concluded that five to ten years of preparation may be needed to avoid a virtual impactor. Frontier Planning for scenario classes — whole-planet impacts, single-nation impacts, impacts on economically disadvantaged regions — was described as still underway as of 2025. An exercise that ends with “planning continues” is a real result about the institution.

Speculative And the experiment nobody will run deliberately. The only complete test of the chain including decision would be a real object with a rising probability and a warning window inside the preparation time. The field's own literature treats that as the case it is not ready for, and no exercise substitutes for it.

Established Kinetic impact is the mature option and its requirements are now specified by a flight. A 610 kg spacecraft, about 500 kg at impact, delivered at 6.6 km/s with autonomous terminal navigation onto a 160-metre target, produced a 2.70 millimetre-per-second velocity change on a body whose mass is estimated rather than known. Established The engineering that made it work was the guidance, not the impactor — hitting a small, previously unresolved body at that speed is a closed-loop imaging and navigation problem.

Established Characterisation hardware is the second requirement and Hera is the specification. Two framing cameras, a thermal infrared imager supplied by JAXA, a micro-lidar altimeter, a hyperspectral imager, and two CubeSats carrying a low-frequency radar, a gravimeter and a dust detector. Internal structure and porosity are radar and gravimetry problems, and they cannot be solved from Earth.

Established Survey hardware divides on wavelength and vantage. Rubin is a wide-field ground optical survey costing about $680 million. NEO Surveyor is a 50 cm infrared telescope in two bands, 4–5.2 and 6–10 micrometres, at Sun–Earth L1, about 1,300 kg, with a 12-year planned survey. Established The vantage is as important as the aperture: an L1 infrared platform can look closer to the Sun's direction than any ground survey, which is the direction Chelyabinsk came from.

Frontier The gravity tractor is the option whose engineering is easiest to state and hardest to justify. Proposed by Edward T. Lu and Stanley G. Love, it uses the mutual gravitational attraction of a hovering spacecraft to tow an asteroid, which requires years to decades of station-keeping at close range. NASA assessed it in 2007 as the most expensive of the mainstream options and it has the lowest technical readiness. Its one real advantage is that it works irrespective of composition, spin state or structure — the properties we are worst at measuring in advance.

Frontier The ion beam shepherd is the intermediate concept. A low-divergence ion thruster fired at the asteroid from a hovering spacecraft applies a slow continuous force with a lighter spacecraft than a gravity tractor requires. It remains at concept stage with no flight.

Frontier Nuclear options are the only family credible on short notice, and the engineering choice inside the family matters. They are assessed at ten to a hundred times the effectiveness of non-nuclear alternatives. Standoff detonation is preferred over contact or subsurface precisely to avoid fragmenting a rubble pile — the aim is to ablate a surface layer and produce recoil, not to break the object. Speculative The HAIV concept pairs a kinetic leader with a subsurface nuclear follower; the claimed capability of destroying a 300-metre asteroid on 30 days' warning is a concept-study figure and is flagged as one here.

Frontier And the requirement that spans all of them: the response has to be buildable inside the window. A deflection mission is a spacecraft with a launch vehicle, a trajectory, a tracking allocation and an integration schedule. NASA's own 2013 testimony put at least five years between decision and launch. No amount of technique maturity compresses that, which is why the engineering answer and the institutional answer are the same answer.

Established The boundary that most needs drawing is with general catastrophic-risk governance. Existential Risk Governance owns how a civilisation ought to decide about low-probability, high-consequence hazards: the decision theory, the discounting, the design of the institutions. This page owns the asteroid case, and it is worth owning separately because the asteroid case is the only member of that family where the hazard is enumerable object by object, the countermeasure is demonstrated in flight, and the missing piece is a named institutional gap rather than a conceptual one. That page asks how you decide under those conditions; this one reports that for asteroids nobody has the authority to decide at all.

Frontier The second boundary is with the infrastructure that a response would consume. Deep Space Infrastructure owns rapid-response launch, deep-space tracking capacity and the contention record. This page takes only the consequence: the antenna hours and launch slots a deflection mission would need are already contended, and adding a crash programme to a rationed network is a scheduling problem with a documented history.

Frontier The third is instrument heritage shared with the resource briefs. Hera's radar, gravimeter and hyperspectral imager measure exactly what Asteroid Mining and Space Resource Economies want measured. The seam is purpose, not hardware: those pages measure an asteroid to value it, this one measures it to hit it correctly, and the same rendezvous produces both datasets.

Established The fourth is legal. Space Law and Governance owns the general finding that no forum has compulsory jurisdiction over space activities, and owns the treaty question that any nuclear option would raise. This page supplies that brief's sharpest worked example of an unassigned decision.

Frontier And one adjacency of technique rather than subject. The photometry, radar ranging and astrometric orbit determination that produced the 33.0-minute result are the same tools that support every small-body mission, and the survey pipeline that finds impactors is the same pipeline that finds targets for Interstellar Probes flyby science and for the mining prospectors. Planetary defence is unusual in this category for being a consumer of ordinary astronomy rather than a producer of novel physics.

Established The institutional finding here is the whole page, and it is well-attested rather than pessimistic. Two bodies were created in October 2013 under the aegis of COPUOS following the Action Team-14 recommendations, and they were created deliberately without decision power. IAWN informs. SMPAG studies and advises. Neither can commit a mission, and neither was ever given that job.

Established IAWN's mandate is concrete and its threshold is public: it is a clearinghouse for shared information on hazardous objects, and it notifies on an impact probability above 1% within 20 years for objects larger than 10 metres. It publishes potential impactors. SMPAG's mandate is to coordinate joint studies of deflection technologies and to provide oversight of missions. Frontier Describing either as having failed at a job it was never given would be unfair and inaccurate — and describing them as decision-making bodies would be simply wrong.

Established NASA's Planetary Defense Coordination Office does not close the gap either. Established January 2016 within the Planetary Science Division, its functions are cataloguing, tracking, funding surveys and coordinating response planning. DART was a joint NASA and Johns Hopkins Applied Physics Laboratory project authorised through NASA's ordinary programme machinery. Nothing in the record describes a mechanism by which the office launches a deflection mission on its own authority.

Frontier The UN endorsement of IAWN and SMPAG is carried here with an explicit caveat. The retrieval path for this rewrite attributes the endorsement to a General Assembly resolution numbered 68/75 but dates it to 2023, which is almost certainly a transcription error — A/RES/68/75 is a 2013 resolution. The number and the year should both be verified against the UN Office for Outer Space Affairs before this claim is relied on, and that office was not reachable for this rewrite. The substance — that both bodies were endorsed through the COPUOS process in 2013 — is not in doubt.

Established The funding institution behaves exactly as the record predicts. A survey instrument lost five Discovery Program competitions in eleven years, then was funded within a year of being moved out of that competition and reclassified on national-security grounds. Frontier That is a fact about how science-mission selection treats infrastructure: a catalogue is not a discovery, and a programme built to rank discoveries will rank a catalogue last, however useful it is.

Established And the cluster-level observation, which this page states in one line and Space Law and Governance states in full. Every brief in this operations cluster carries constraints-only adjudications and not one brief-to-brief edge. Planetary defence is not waiting on a result. It is waiting on an appropriation and a decision rule, and neither is produced by an experiment.

Established The sharpest ethical problem in this subject is not whether to deflect but where the residual risk goes while you are deflecting. Rusty Schweickart's objection states it exactly: during a slow deflection the most likely impact point drags across different countries' territory. Deflection does not remove risk; it moves it, and for a period it moves it deliberately across populated places that had no say. That is why the decision cannot be delegated to a technical body, and it is the strongest argument that the missing institution is missing for a reason.

Frontier The distributive version of the same problem is on the record as unfinished business. Planning for scenario classes — whole-planet impacts, single-nation impacts, impacts on economically disadvantaged regions — was described as underway as of 2025. A regime that has not decided how to treat an impact on a country with no space programme has not decided the hardest case.

Established Risk communication is the part that has been tested in public and largely passed. 2024 YR4 was rated Torino 3, notified through IAWN 33 days after discovery, assessed at 3.1% by NASA and 2.8% by ESA on the same day, and communicated with those numbers attached. Frontier The honest reading is that the risk was real, was resolved by observation, and resolved downward — which is the system working, not the risk having been illusory. Describing it afterwards as “never a threat” teaches the public the wrong lesson about the next one.

Frontier The nuclear question is ethical before it is technical. Standoff detonation is assessed as the only credible short-warning option and as ten to a hundred times more effective than the alternatives. Testing it, developing it, or pre-positioning it engages arms-control commitments that no planetary-defence body has authority over. The capability most useful in the worst case is the one hardest to prepare in advance, and this page does not resolve that; Space Law and Governance owns the treaty question.

Established And an ethical point about the mandate itself. The 2005 Act set a threshold at 140 metres, which is a reasonable line for regional devastation. Chelyabinsk was 17 to 20 metres, undetected, and sent 1,491 people for treatment. Frontier A survey mandate is a statement about which harms are worth spending public money to prevent, and the current one is calibrated to the tail rather than to the mode.

Established The civilizational case for this subject is genuinely strong and does not need overstating. The impact hazard is the one catastrophic risk that is fully enumerable in principle: a finite population of objects, each with a computable orbit, against which a demonstrated countermeasure exists. Nothing else in the existential-risk family has that shape.

Established And the achievement is real. In 2022 a species deliberately changed the orbit of another body in the solar system and measured the change by two independent methods. That is worth stating plainly before any qualification, because it is the qualification that usually gets dropped rather than the achievement.

Frontier The deflating half is that the capability is demonstrated and the institution is unassigned, and the second half moves more slowly than the first. A technique was proven in one flight. A decision authority has not appeared in the thirteen years since IAWN and SMPAG were created, and the fetched record is explicit that as of 2025 no firm national commitments on response protocols exist. The gap between those two clocks is the subject.

Frontier The cost comparison that would settle whether this is proportionate cannot be made here, and saying so is more useful than guessing. The programme costs are known: NEO Surveyor's development baseline is reported at about $1.2 billion, Rubin's construction at about $680 million, Hera's construction contract at €129.4 million. Handwave Any statement of the expected annualised loss from impacts, against which those figures would be judged, requires an impact-frequency model this rewrite did not obtain, and none is asserted. The comparison is the natural one and the denominator is missing.

Established The sentence to leave a reader with. The physics is finished, the technique is flown, the surveys are running, and the answer is still not available — because what remains is a budget line and a decision rule. Planetary defence is the clearest case in this category of a frontier where the science stopped being the hard part.

These horizons track the four links — find, characterise, decide, act — because they move at completely different rates and only one of them has a schedule:

• 10 yr: Established Hera arrives at Didymos in November 2026 and measures the masses; expect the β interval of 2.2 to 4.9 to collapse toward a number within a year of arrival. Established Rubin, in full survey operations since 30 June 2026, accumulates toward its estimated 62% completeness for 140-metre objects. Frontier NEO Surveyor launches no earlier than September 2027 on a 12-year survey; the combined completeness figure does not exist yet and should be watched for rather than assumed. Frontier The decision-authority gap is the item most likely to be unchanged at the end of this decade.
• 25 yr: Frontier This is the window in which the 140-metre mandate either closes or is formally revised, and in which the population denominator should converge as two surveys sample the same sky. Frontier It is also the window in which a second deflection demonstration — against a solitary body, or with a different technique — either happens or does not; nothing currently funds one. Speculative A decision protocol with named signatories is possible in this window and has no current sponsor.
• 50 yr: Frontier On the survey side this is long enough for the catalogue to be substantially complete at 140 metres and materially better below it, if funding persists at anything like present levels. Speculative On the response side, a stockpiled or rapidly-assemblable deflection capability — the thing that would actually convert a short-warning case into a survivable one — remains a concept in every published form. Speculative Whether anything with authority to commit a mission exists by then is a political forecast, not a technical one.
• 100 / 250+ yr: Handwave Beyond useful forecasting for a subject whose institutions have not changed shape in the thirteen years since they were created. Speculative The structural observation that survives at this range is that the impact hazard is the one civilizational risk that a sufficiently patient survey retires by enumeration — every object found and tracked is a risk permanently converted from unknown to computed, and that is a monotonic process as long as somebody pays for it.

• Depends on This brief records no dependency on another brief, and the absence is deliberate. Nothing in planetary defence waits on a physics result, a new material or a demonstration that some other page on this map produces. The kinetic impactor has flown; the surveys are operating; the characterisation mission is in cruise. Where the subject is uncertain, the reason is one of three things and none of them is a missing discovery: a parameter measured once and conditional on an assumption nobody has checked (β, conditional on Dimorphos's density), a rate that is a budget line (survey completeness), or an institution that does not exist (a body that can commit a deflection).
• Requires (not on this map) Both constraints are institutional, and both are as well-attested as the measurements on this page. The first is a body that can commit a deflection mission inside the window that matters. IAWN informs, SMPAG studies and advises, and NASA's Planetary Defense Coordination Office catalogues and coordinates; every mission that has flown was authorised through an ordinary agency budget process on a decade timescale. Against that, NASA's own 2013 testimony gives at least five years from decision to launch and the 2021 tabletop exercise gives five to ten years of preparation — while 2024 YR4, discovered unusually early, offered about eight years and Chelyabinsk offered none. The second is a survey budget. The 90%-at-140-metres mandate dates from 2005, was missed in 2020, and carried no dedicated funding; Rubin reaches an estimated 62% alone; NEO Surveyor lost five Discovery competitions in eleven years and was funded only once it was moved out of the science programme and reclassified on national-security grounds. Neither constraint is produced by an experiment.
• Enables No typed enabling edge is claimed. The nearest thing to one is negative: a deflection mission would consume launch and tracking capacity that Deep Space Infrastructure documents as already contended, so this brief is better read as a future consumer of that page's subject than as an enabler of anything. The characterisation instruments Hera carries are the same ones Asteroid Mining wants, which is shared heritage rather than an edge.
• Adjacent Existential Risk Governance owns the general theory of deciding under low-probability, high-consequence uncertainty; this page owns the asteroid case and hands the general question across. Space Law and Governance owns the observation that no forum has compulsory jurisdiction, of which the deflection gap is the sharpest instance, and owns the treaty question any nuclear option raises. Asteroid Mining and Space Resource Economies share the characterisation instruments and differ in purpose. Outside this map: ground-based survey astronomy, radar ranging, small-body geophysics, and the launch industry that would have to fly a response.

Handwave “Planetary defence is a solved capability.” One technique, demonstrated once, against one object of one size class, in a binary system that made the measurement possible, with years of warning — and with an efficiency still bracketed between 2.2 and 4.9 pending Hera. Established The most capable survey ever built reaches an estimated 62% of a mandate set at 90% in 2005 and missed in 2020, and no body can commit a deflection mission. Calling that solved describes the physics and skips the institution, the budget line and the error bar.

Established “DART measured a momentum enhancement factor of 3.61.” It did not. It measured an along-track velocity change of 2.70 millimetres per second with an uncertainty of 0.10. β = 3.61 is that measurement divided by a mass derived from an assumed bulk density of 2,400 kg/m³, and the same paper states that across the plausible range of 1,500 to 3,300 kg/m³ the value spans 2.2 to 4.9. Established The headline figure is a quotient with an assumption in the denominator, and Hera exists to remove the assumption.

Established “DART proved we can deflect asteroids.” With qualification, yes; without it, no. It proved that a kinetic impactor can measurably change the orbit of a 160-metre moonlet in a binary system, where the change was observable because it was a binary. Deflecting a solitary body on a heliocentric orbit is a harder measurement problem and has not been demonstrated.

Frontier “We know roughly how complete the hazardous-asteroid catalogue is.” At the kilometre scale, yes — about 867 known against 900 to 981 estimated, roughly 90%. At 140 metres, no: about 11,167 are known against published population estimates of 13,200, 27,100, under 20,000 and about 35,000, a spread of a factor of 2.7 in the denominator. Frontier Implied completeness therefore ranges from about a third to about six-sevenths depending on which estimate you divide by, and that arithmetic is presented here as arithmetic.

Frontier “NEO Surveyor plus Rubin will close the 90% mandate.” Rubin alone is estimated at 62%. No combined figure was obtainable for this rewrite and none is asserted. The claim is plausible, widely repeated, and unsourced here.

Established “IAWN and SMPAG are the decision-making bodies.” They are not, and they were never designed to be. IAWN informs; SMPAG studies and advises. Established The opposite error is just as wrong: describing them as having failed is unfair to bodies doing exactly the job they were given, and 2024 YR4 shows the coordination function performing well.

Speculative “A nuclear option could handle a short-warning object.” Nuclear standoff is assessed at ten to a hundred times the effectiveness of the alternatives and is the only family credible on short notice — against 100-to-500-metre objects with about two years of warning. The HAIV claim of destroying a 300-metre asteroid on 30 days' warning is a concept-study figure with no demonstration behind it, and it is flagged here as one.

Frontier “If deflection fails you can just blow it up.” Work from 2019 indicates asteroids are substantially harder to destroy than earlier thought, because a disrupted body can gravitationally reassemble; earlier simulations from 2011–2012 found fragments do exceed escape velocity above a sufficient and well-matched energy delivery. Both results are real and they bracket a threshold rather than contradicting each other. Disruption is not a fallback you select at the last minute; it is an energy-delivery problem with a hard edge.

Established “2024 YR4 was never a threat.” It was assessed at 3.1% by NASA and 2.8% by ESA on the same day and rated Torino 3, the second-highest rating ever assigned. Established The correct statement is that the risk was real, was resolved by observation, and resolved downward, which is the system working rather than the risk having been imaginary.

Frontier “The surveys cover the objects that actually hurt people.” The mandate is 140 metres. Chelyabinsk was 17 to 20 metres, arrived from a radiant near the Sun, was not detected in advance, and sent 1,491 people for medical treatment. Frontier An infrared telescope at L1 improves the sunward blind spot; it does not remove it, and the smallest class remains the one most likely to produce casualties in any given decade.

Established And the framing verdict. The capability is demonstrated and the decision is unassigned. 2024 YR4 exercised every link in the chain except decision, and the only reason no link failed is that the probability reached zero before anyone had to decide whether to launch anything. That is the most accurate one-sentence description of the state of planetary defence available.