The Libration Points

The dynamical boundary

Where Cislunar Space Begins: The Three-Body Transition

Below GEO, the Moon is a rounding error. Near the gateways, it rewrites the problem.

There is no painted line in space marking the beginning of the cislunar regime. The useful boundary is dynamical. Below geostationary orbit (GEO, 35,786 km), the Moon’s gravitational pull on a spacecraft amounts to roughly three parts in 100,000 of Earth’s central acceleration — a perturbation small enough to correct away. Near the Earth-Moon libration points, that ratio climbs to about 1.5%, and the familiar two-body Keplerian picture gives way to the circular restricted three-body problem.

The geometry is visible in one number. The Moon’s Hill radius — the region in which lunar gravity dominates local orbital behavior — is about 61,500 km. Earth-Moon L1 (EML1) sits roughly 58,000 km from the Moon, just inside that boundary. Earth-Moon L2 (EML2) sits roughly 64,500 km away, just outside it. That near-symmetry is not a coincidence. EML1 and EML2 occupy the necks of the Earth-Moon gravitational landscape, where families of unstable trajectories thread through and create natural transit corridors between Earth-centered, Moon-centered, and escaping orbits.

That is why EML1 and EML2 are gateways rather than merely points on a diagram.

Industrial topology

Warehouses and Junctions: Stability as an Industrial Property

Stability and departure cost are the same property viewed from opposite sides.

Every periodic orbit in the three-body problem has a measure of stability — how quickly a small error grows. Engineers express it through the eigenvalues of the orbit’s monodromy matrix. The industrial interpretation is simpler:

Unstable orbits cost propellant to hold, but the same instability creates natural pathways away — invariant-manifold tubes that let celestial mechanics carry much of the transfer. Depots, transshipment nodes, customs houses, and high-turnover logistics belong at unstable junctions.

Stable orbits keep inventory with little or no station-keeping, but there is no manifold waiting to carry it away. Departure requires a deliberate burn. Strategic reserves, spares, quarantined hardware, and anything meant to wait belongs at stable warehouses.

That single trade divides cislunar infrastructure into two kinds of nodes. The same logic returns around Jupiter and Saturn. The orbit families change. The industrial topology does not.

The freight rule: Route cargo through junctions. Park inventory in warehouses. A vehicle with no schedule rides a manifold for days or weeks and saves propellant. Inventory that needs to wait belongs in a stable orbit instead.

Earth-Moon families

Earth-Moon Orbit Families: From Lyapunov to Distant Retrograde

The names sound esoteric. Each one describes a different piece of infrastructure.

EML1 Halo Orbits — The Transshipment Node

EML1 halo orbits maintain Earth visibility while providing access to the lunar near side. The orbit is unstable enough to offer useful departure manifolds but manageable enough for station-keeping measured in tens of meters per second per year. That makes EML1 a natural transshipment node, customs house, and propellant depot — only about 2.5 km/s from the lunar surface.

EML2 Halo Orbits — The Far-Side Gateway

EML2 halo orbits can see Earth while maintaining access to the lunar far side. China’s Queqiao-1 relay satellite operated in an EML2 halo orbit from 2018, proving the concept. EML2 is also the departure port of the Earth-Moon system: roughly 0.14 km/s from Earth escape. For any architecture that intends to leave Earth-Moon space — bound for Mars, the asteroid belt, or beyond — EML2 is the natural last waypoint.

Near-Rectilinear Halo Orbit (NRHO) — Crew Staging

NRHO is the high-amplitude end of the halo family. NASA selected a southern L2 NRHO for its Gateway station: a roughly 6.5-day orbit with perilune around 3,000–3,500 km and apolune near 70,000 km. It combines long dwell over one lunar pole, near-continuous Earth communications, and near-marginal stability — unusually good for crew staging and polar surface operations. Gateway was paused in March 2026 under NASA administrator Isaacman in favor of a surface-first lunar base strategy, but the NRHO itself remains the most operationally characterized cislunar orbit family. A 9:2 synodic resonance (period ≈ 6.56 days) keeps the orbit eclipse-free — the cheapest battery can be an orbit chosen so the worst contiguous solar outage never occurs.

Distant Retrograde Orbit (DRO) — The Warehouse

In the rotating Earth-Moon frame, DRO loops retrograde around the Moon and can remain dynamically stable over an enormous range of sizes. Station-keeping can approach zero. The price is exactly what the warehouse rule predicts: no useful departure manifold, and no inherent polar coverage. DRO is where one parks propellant reserves, spares, quarantined hardware, or strategic inventory intended to remain untouched.

Butterfly, Vertical, and Lyapunov Families

Planar Lyapunov orbits around EML1 and EML2 are the fundamental families from which halo and other orbits bifurcate. Operationally they define the walls of the transit corridors rather than places to occupy. Butterfly orbits can provide high-latitude and far-side dwell that comparable halo orbits cannot. Vertical families can see both lunar poles from one orbit. Neither has flight heritage yet, but both become plausible second-generation relay architectures.

EML4 and EML5 — The Distant Warehouses

EML4 and EML5 are stable for the Earth-Moon mass ratio — the classic O’Neill-colony locations. But they sit roughly 384,400 km from both Earth and Moon. Their stability makes them excellent warehouses and poor ports; their distance makes them awkward for surface logistics.

Low Lunar Orbit and Frozen Orbits

Mascons — the Moon’s uneven concentrations of mass — destabilize many otherwise convenient circular orbits. Frozen orbit families near inclinations of roughly 27°, 50°, 76°, and 86° are exceptions. This is why lunar communications architectures such as NASA’s LunaNet, ESA’s Moonlight, and China’s Queqiao series favor eccentric and geometrically peculiar orbits rather than a miniature version of GPS.

Orbit family reference

Earth-Moon Cislunar Orbit Families

Infrastructure separates into unstable junctions and stable warehouses.

Sun-Earth points

Sun-Earth Libration Points: The Other Five

Separated from the Earth-Moon L-points by 1.5 million kilometers and an entirely different job description.

Sun-Earth L1 (SEL1) — The Solar Watchpost

SEL1 sits roughly 1.5 million km sunward of Earth. It provides continuous solar viewing and space-weather monitoring. The current fleet includes SOHO (ESA/NASA, operating since 1995 — over 30 years), NASA’s Advanced Composition Explorer (ACE, since 1997), and the newest arrivals: NASA’s IMAP (Interstellar Mapping and Acceleration Probe, which reached L1 on 10 January 2026 and began its primary science mission on 1 February 2026) and NOAA’s SOLAR-1 (née SWFO-L1, which reached its final L1 position on 23 January 2026 — NOAA’s first satellite designed specifically for continuous operational space-weather observations). DSCOVR, which has monitored the solar wind and taken full-Earth images from L1 since 2015, is approaching end of service.

SEL1 is not merely a science address. For crews operating outside Earth’s magnetosphere — at the Moon, in transit to Mars, or at cislunar junctions — the spacecraft at L1 are the upstream warning network. They detect coronal mass ejections and solar energetic particle events before they arrive. That makes L1 infrastructure in the same sense that a weather radar is infrastructure.

Sun-Earth L2 (SEL2) — The Cold Observatory

SEL2 sits roughly 1.5 million km anti-sunward. There the Sun, Earth, and Moon all crowd into one direction, so a single sunshield blocks their combined heat and light. That geometry provides the cold, stable, wide-sky environment that infrared and survey astronomy requires.

The flagship resident is NASA’s James Webb Space Telescope, which arrived at SEL2 in January 2022 and — now four years into science operations — continues to rewrite the textbook on the early universe, exoplanet atmospheres, and star formation. ESA’s Euclid dark-universe surveyor, launched July 2023, shares the neighborhood on a six-year nominal mission. Earlier occupants include the Planck and WMAP cosmic-microwave-background mappers, the Herschel infrared telescope, and ESA’s Gaia star-mapper — which charted nearly two billion stars before being decommissioned in March 2025 and sent into a retirement heliocentric orbit.

Starship-class fairings could eventually relax the folding constraints that shaped JWST — allowing larger monolithic or less-folded optical systems — but the cold observing environment itself still belongs at SEL2. The address is permanent; the telescopes will get bigger.

Sun-Earth L4 and L5 — The Side-View Points

SEL4 (60° ahead of Earth in its orbit) and SEL5 (60° behind) can view the Sun-Earth line from the side. SEL5 is unusually valuable for space weather. An appropriately instrumented spacecraft there could observe solar eruptions from a side angle and potentially turn warning times measured in tens of minutes near Earth into warnings measured in days. For crews outside the magnetosphere, that is not a science mission — it is infrastructure. ESA’s Vigil mission (formerly Lagrange) is planned for the late 2020s to demonstrate this capability.

Sun-Earth L3 — Behind the Sun

SEL3 sits on the far side of the Sun from Earth. No current occupants and no near-term industrial role, but it appears repeatedly in planetary-defense studies as a location to detect asteroids approaching from the Sun’s direction — the blind spot of every Earth-based survey.

Spacecraft census

Current Sun-Earth Libration-Point Spacecraft

The infrastructure already in place at the Sun-Earth L-points, as of mid-2026.

SEL3 and SEL4 have no current occupants. Carruthers Geocorona Observatory also reached L1 in late 2025 as a rideshare with IMAP but is studying the Earth’s exosphere rather than performing solar monitoring.

Lunar export terminals

Where Lunar Mass Arrives: The Capture-Cost Inversion

The mass driver cares about capture Δv, not total Δv. That changes which destination is cheapest by a factor of 26.

A chemical vehicle pays propellant for the entire climb out of a gravity well. A lunar mass driver supplies the launch impulse electrically while the payload is still on the track. Once the package leaves the Moon, the economically relevant cost is capture at the destination — not total Δv from the surface.

A package arriving from lunar distance at a 400 km LEO perigee reaches about 10.75 km/s while local circular velocity is about 7.67 km/s — it must shed roughly 3.08 km/s. At GEO the capture burn falls to about 1.05 km/s. At EML1 or EML2, manifold insertion is on the order of 0.12 km/s.

Total surface-to-destination Δv makes the destinations look different by a factor of roughly three. Capture-only Δv makes them different by roughly 26.

The mass driver analyzed in the lunar architecture leaves the track at about 2.52 km/s. Reaching EML1 or EML2 from the lunar surface requires about 2.5 km/s. The machine as designed already reaches the cislunar gateways. Everything added to push cargo below GEO — a solar-electric capture stage, propellant, months of spiral time, and a huge in-transit population crossing operational conjunction-management shells — exists only because the cargo is being pushed past GEO.

EML1 and EML2 are the natural first export terminals of a lunar industrial economy.

Delta-v reference

Capture Δv from Lunar Mass Driver to Cislunar Destinations

The electrically supplied launch impulse is already paid on the surface. The relevant cost is capture at the receiving node.

The tug problem: A reusable electric tug cycling EML1 ↔ low orbit pays ~3 km/s each way. Hall-effect and gridded-ion thrusters can do this efficiently, but their propellants — xenon, krypton, argon — have no demonstrated off-Earth source. The noble gases are a gating terrestrial stock for electric propulsion until an off-Earth supply chain is demonstrated.

The rescue line

The Far Side of the Rescue Line: Deep-Space Risk at Conversational Distance

The three-second conversation makes cislunar space feel near. The return trajectory makes it deep space.

An EML1 crew can speak with Earth across a round-trip light time of only a few seconds. They can see the planet. Mission control can watch nearly everything they do. It feels attached.

It is not.

No standing vehicle can bring a crew home from beyond LEO on emergency timescales. Orion cannot simply descend into and climb back out of low lunar orbit — that propulsion constraint was one reason NASA’s Gateway architecture existed in the first place, and pausing Gateway in 2026 does not change the underlying physics. A crewed facility at EML1, EML2, or NRHO inherits the full no-rescue architecture: consumables sized against an abort clock, medical capability that cannot assume rapid evacuation, and redundancy designed around unreachability.

That gap between perceived and actual proximity is one of the region’s defining hazards.

The unwatched volume

Stability as a Surveillance Property: The Unwatched Volume

A spacecraft in a distant retrograde orbit can remain dynamically stable with almost no station-keeping, tens of thousands of kilometers from the Moon.

Cislunar space has one final industrial property that does not appear in a Δv table: it is barely monitored.

A spacecraft in an unstable halo orbit has to maneuver periodically — those station-keeping events announce activity. A spacecraft in a DRO can remain dynamically stable with almost no maneuvers, tens of thousands of kilometers from the Moon and hundreds of thousands from Earth. China has already flown test articles in cislunar space, and every major space power now has reason to care about what can be parked there, moved through there, or hidden in plain orbital geometry.

Stability is therefore not only a propellant property. It is a surveillance property.

The strategic output of a lunar industrial base may eventually be more than oxygen, water, shielding, radiator panels, or structural mass. A lunar outpost also sits high in Earth’s gravity well with access to the junctions of the three-body network. It can sell mobility. The warehouses hold mass. The junctions sell Δv. And unlike low Earth orbit, much of the volume between them remains a place nobody can continuously watch.