Research

Geographos vs Sisyphus: what your 2050s–2070s “unlock windows” really look like, and what it would take to build a permanent NEA base

Executive summary

You can absolutely justify a “rabid vacuum rush” in the early–mid 2050s for Geographos, but not for 1866 Sisyphus. The hard limiter is geometry: Sisyphus’ Earth MOID is ~0.103–0.104 AU (≈15.5 million km), which means it simply does not come anywhere near the “few million km” regime in the time period you’re aiming at. ESA’s NEO portal lists Sisyphus’ next Earth close approach as 2058‑11‑18 at 0.15329 AU (22,932,282 km). That matches Space Reference’s close-approach table (2058, 2071, 2157, 2170).

By contrast, 1620 Geographos has a much tighter orbit-to-orbit proximity: its Earth MOID is ~0.030 AU (Wikipedia quotes 0.0301 AU). Space Reference’s close-approach list shows a 2051‑08‑23 encounter at 7,163,068 km and a 2076‑08‑20 encounter at 11,724,111 km (your “eleven million km” decade), with relative encounter speeds around 11.0 km/s and 10.2 km/s respectively.

The important “feel” point: miss distance is not travel time. An 11–12 million km miss distance is not “a one-month commute” unless you accept high-energy departures and very high arrival relative speeds—which then makes rendezvous/landing the real challenge, not the cruise.

Close-approach reality check for your dates

Geographos has a real “2050ish hunger window”; Sisyphus does not

Geographos (1620) (Apollo, PHA) has “interesting” Earth passes in the timeframe you’re describing:

So yes: 2076 is closer than 2065, but note the irony: 2051 is much closer than 2076 in absolute distance. If your story wants the first “I’m going, fuck you all” moment in the 2050s, Geographos already provides it in 2051 at ~7.16 million km.

Now compare Sisyphus (1866) (Apollo, large, not PHA):

Space Reference’s predicted close approaches are:

And ESA’s NEO portal summary agrees on the 2058‑11‑18 next approach at 0.15329 AU / 22,932,282 km, and gives Earth MOID 0.10367 AU.

What a “2052 Sisyphus” would imply

To have a meaningful 2052 “Sisyphus unlock,” you’d need an object whose orbit comes within a few million km: that implies a MOID more like 0.02 AU or smaller for the “5 Mkm” regime (5 million km ≈ 0.033 AU, but practical “close” windows tend to be tighter still). Sisyphus simply isn’t that object: its orbit stays roughly 0.10 AU away at best.

If you want a “big rock + genuinely close flyby” vibe, Toutatis is the poster child (but later)

You called Toutatis a resonance bridge; that’s not just vibes—JPL’s radar page explicitly notes its 3:1 mean-motion resonance with Jupiter as a key dynamical pathway from the main belt into Earth-crossing space. Space Reference gives 4179 Toutatis a standout close approach in 2069‑11‑05 at 2,969,363 km with relative speed 8.406 km/s—that’s in the “few million km” class your brain is reaching for.

“Eleven million km in 2076” is not “one month away” unless you accept brutal energy and rendezvous pain.. The key distinction: flyby/intercept vs rendezvous/landing

At Geographos’ 2076 close approach, the Earth–asteroid relative speed at closest approach is ~10.225 km/s. That number tells you two things immediately:

1. A fast intercept can be short, because you can ride a high-energy trajectory and “arrive” when the asteroid is near Earth (in distance), but…
2. A base mission cannot be a fast intercept, because you can’t “land” or “dock” at ₁₀ km/s relative speed. You must bleed off most of that velocity (directly or indirectly) to match the asteroid.

So the narrative tension is real and useful: distance is the marketing slide, relative velocity is the engineering cliff.

Back-of-envelope travel-time scale (helps you write believable dialog)

If you (fictionally) time departure so you intercept near the close-approach epoch, then the distance scale is on the order of 1e7 km. Time-of-flight depends on your average closing speed relative to Earth (or relative to the asteroid, if you’re doing a proper chase).

For 11.7 million km:

• At 1 km/s average closing speed: ~135 days
• At 5 km/s average closing speed: ~27 days
• At 10 km/s average closing speed: ~13.5 days

This is why you can write characters saying “it’s only twelve million klicks” and somebody else replying “yeah, and we hit it at ten klicks per second.” The latter is the real problem, and Geographos’ catalogued flyby speed gives you that line for free.

Why NHATS implicitly argues against your “< 1 month, crewed rendezvous” plan

NASA’s NHATS framework (Near-Earth Object Human Space Flight Accessible Targets Study) is explicitly about round-trip human rendezvous targets and screens using mission constraints. The API documentation shows the common “standard” constraint example includes dv, dur, stay (and defaults include dv up to 12 km/s and dur up to 450 days). NHATS also defines what “total mission delta‑V” means: it includes the burn out of a notional Earth parking orbit plus the delta‑V to match the NEA at arrival, depart the NEA, and (if needed) control Earth-entry speed. Space Reference flags both Geographos and Sisyphus as “not considered a viable target” under NHATS—consistent with the idea that, under standard human round-trip assumptions, they’re energetically awkward. In plain story terms: your <1-month crewed rendezvous is outside the design space NASA’s “accessible target” tooling is built around—not because nobody wants it, but because the energy and operational risk explode.

How many people to build an “Antarctica-style permanent base” on an asteroid

Antarctic analogs give you two useful anchors: build crew and winter-over crew

For the Amundsen–Scott South Pole Station:

• NSF notes the original “Old Pole” station was built by an 18‑man U.S. Navy crew, who were also the first to winter-over.
• Modern operations are much larger; a common pattern is summer ~150 and winter-over “a few dozen” (e.g., 39 in 2021).
• NSF describes today’s station as a complex system with ~47 buildings and explicitly includes a food growth chamber among the facilities.

That’s a direct analog to your “bunker, in-situ, survive, grow food” requirement: you’re describing a base that is logistics + maintenance + life support, not just “a lab.”

NASA’s Mars reference architecture shows what “minimum skill mix” looks like for a first foothold

NASA’s Mars Design Reference Architecture 5.0 overview text (via NTRS download) states:

• The DRA 5.0 concept envisions sending six crewmembers to Mars, explicitly framing this as a compromise between skill mix / effort and the systems needed to support them.

That six-person number is very useful as a minimum viable expedition crew benchmark (not a colony), especially when paired with heavy predeployment.

A realistic “founding base” headcount depends on how much you predeploy robotically

If you truly can “send endless unmanned before and after,” your crew can be sized for:

• commissioning and integration of delivered systems
• routine operations + maintenance
• growth of local capability (excavation, ISRU, agriculture)
• resilience against single-person loss (illness, EVA accident, burnout)

A practical way to think about it:

You need redundancy by role, not by headcount. NASA’s human-systems standard (NASA‑STD‑3001 Vol 2) requires training a minimum of two crewmembers per vehicle/platform as Crew Medical Officers, specifically so there is backup medical capability. That’s one clear example of a broader rule: for any “must-not-fail” function, you need at least 2 qualified people.

For an asteroid base, the must-not-fail functions usually include:

• life support / atmosphere / water loop
• power generation and distribution
• thermal control
• comms and navigation
• EVA operations / suit maintenance
• robotics / teleops (especially if construction and mining are largely robotic)
• medical (including dentistry, surgery stabilization, psych)
• food production (horticulture + nutrient chemistry + pest/disease management)

Minimum psychologically and operationally credible permanent base crew (with heavy automation and robust predeployment): ~12–18 people

This gives you:

• at least two trained medical officers (per NASA logic)
• redundancy across power/ECLSS/robotics
• enough people to staff 24/7 monitoring without everyone being permanently on call

More “Antarctica-like” self-sufficiency with expansion capability: ₂₄–40 people

This starts to look like the modern winter-over scale at South Pole (dozens) rather than a Mars expedition crew of six. If you want the story beat of “they’re going for life,” then the social/psych factor pushes you toward the higher end, because miniature societies get fragile fast when the headcount is single digits.

“One-way for life” doesn’t remove the need for propulsion; it shifts it

You’re right that a return vehicle is a huge mass investment, and “one-way” can be an economic choice.

But one-way does not eliminate the propulsion problem. The crew still needs:

• a way to match velocity with the target (rendezvous)
• a way to control descent/contact in microgravity
• a way to move the base around the asteroid (hoppers, tethers, winches, reaction control, anchor systems)

Geographos’ close-approach relative speeds (≈10–13 km/s) mean you’re not “landing” without major energy management. Sisyphus is even worse at close approach: 20+ km/s class.

So your trade is really: “Return stage mass” vs “arrival/braking + local mobility mass”. If you underfund the second, your colonists arrive as a tragedy, not a base. Cislunar L4/L5 staging and what it adds to your 2050s “rush” narrative ... You mentioned “game out what else constellations appear, in particular L4/L5.” In practice, writers use Lagrange sites as:

• political choke points (ports, inspection, customs)
• industrial aggregation nodes (fuel depots, construction yards)
• rescue/abort staging

Two grounding facts you can use:

• NASA’s basics-of-space-flight primer explains that spacecraft can operate around Lagrange points in “halo” style orbits with stationkeeping; it gives real examples like SOHO at Sun–Earth L1 (₁.53 million km) and WMAP at L2.
• The Earth–Moon L4/L5 points are theoretically stable in the restricted three-body approximation, but solar perturbations matter. The Royal Astronomical Society notes L4/L5 in the Earth–Moon system are not completely stable due to the Sun’s gravitational disturbance.

The real-world cherry on top: the Kordylewski dust clouds (very faint dust concentrations) are associated with Earth–Moon L4/L5, with repeated modern polarimetric evidence in the literature.

Story implication: L4/L5 are plausible as civilization nodes, but they are not “natural asteroid parking lots.” Large structures would be there by stationkeeping and economics, not because rocks naturally gather there in stable swarms like Jupiter Trojans.

If you need “ALL other Apollo asteroids,” don’t list them—query them

There are tens of thousands of NEAs in modern catalogs (ESA’s portal page cites ~39,868 known NEAs at last update in late 2025). A massive fraction of those are in the Apollo/Aten/Amor families; “Apollo asteroids” alone are too many to enumerate manually in a sane workflow.

What you actually want is a filterable list by:

• orbit class = Apollo (APO)
• Earth MOID threshold (e.g., MOID < 0.02 AU)
• size proxy (H magnitude)
• and (for your narrative) close-approach date range around 2050–2080

NASA/JPL provides exactly the plumbing you need:

• Orbit-class code APO is documented in the SBDB filter list.
• The SBDB Query API lets you request sets of objects by constraints rather than one-by-one lookups.
• For “who passes Earth when,” the SBDB Close-Approach Data API (CAD) documents date filters and returns date, nominal distance, and relative velocity fields.

Here are example query patterns (URLs in code so you can copy/paste):

textCopy# JPL CAD API documentation
https://ssd-api.jpl.nasa.gov/doc/cad.html

# JPL SBDB Query API documentation
https://ssd-api.jpl.nasa.gov/doc/sbdb_query.html

# JPL NHATS API documentation (human-accessible target screening logic)
https://ssd-api.jpl.nasa.gov/doc/nhats.html

In your fiction workflow, that lets you “spawn” the 2050s rush by selecting:

Apollos with MOID < 0.02 AU and close approaches in 2050–2060 with dist < 0.05 AU …and then sort by relative velocity to decide which ones are “gold rush accessible” versus “death wish fast movers.”

Bottom line for your specific prompt

• No, there is no “1866 Sisyphus 2052 unlock.” The first major listed approach after mid-century is 2058‑11‑18 at ~22.93 million km, and its MOID (₀.1037 AU) means it won’t ever be a “few million km” rock without long-term orbital evolution.
• Yes, Geographos has a strong 2051 and 2076 narrative. 2051 is ~7.16 million km; 2076 is ~11.72 million km with ~10.2 km/s encounter speed.
• Travel time doesn’t come from miss distance; it comes from energy and rendezvous requirements. For a base, the hard part is not reaching “11 million km away,” it’s arriving slow enough to stay.
• A believable permanent base founding crew is ~12–18 with heavy automation, ~24–40 for Antarctica-like resilience, with “minimum of two medical officers” style redundancy as a good design heuristic.