An independent researcher has published a peer-reviewed engineering blueprint for keeping Earth habitable for up to 9.1 billion years — not by escaping the planet, but by re-engineering the solar system around it. The centrepiece: firing a continuous particle beam made of oxygen mined from Jupiter to slowly nudge Earth into a wider, safer orbit before the Sun swells into a red giant and swallows it whole.
The paper, by researcher Gabriel Harry, has been accepted for publication in the Journal of the British Interplanetary Society — a legitimate, peer-reviewed astronautics journal that has been publishing serious speculative engineering since 1934 — with a preprint already available on arXiv. It's worth being clear about that credibility upfront, because the plan itself sounds, at first read, like something out of a science fiction script.
The Problem Harry Is Actually Solving
Astronomers broadly agree the Sun will make Earth uninhabitable well before it destroys the planet outright. Rising solar luminosity is expected to boil Earth's oceans and strip its atmosphere within roughly a billion years, and in about five billion years, the Sun will exhaust its hydrogen fuel and expand into a red giant large enough to engulf Mercury and Venus — with Earth sitting dangerously close to the boundary of being swallowed too.
Harry's paper doesn't dispute any of that. Instead, it asks a genuinely interesting engineering question: if a sufficiently advanced civilisation had billions of years to prepare, what could it actually build to survive it?
The Three-Stage Plan
The proposal unfolds in stages, each solving a problem created by the one before it.
Stage one: block the aging Sun's excess heat, buying time as solar output climbs. But blocking the Sun also removes the light and warmth life depends on, which creates problem two.
Stage two: turn Jupiter into an artificial sun. Harry proposes placing fusion reactors roughly 6,500 kilometres into Jupiter's atmosphere, where they'd consume the planet's abundant hydrogen and helium-4 to generate power. That energy would be beamed to Jupiter's L1 Lagrange point and relayed to Earth by laser, functioning as a substitute light and heat source. It's a smaller-scale cousin of the "Dyson sphere" concept theoretical physicists have floated for decades — using a star system's own resources to engineer starlight rather than simply receiving it.
Stage three, the most ambitious: move Earth itself, from its current orbit at 1 astronomical unit out to a safer 1.2 AU, putting enough distance between Earth and the expanding Sun to avoid engulfment during the red giant phase. This is where the oxygen beam comes in — extracted from Jupiter and fired continuously past Earth, its momentum gradually widening the planet's orbit over the course of centuries. Harry calculates the beam would need a mass flow roughly equal to half the Amazon River's total outflow, sustained essentially without interruption, and requiring precise Sun-Earth-Jupiter alignment throughout.
Why This Isn't Actually About Next Week
It's worth being direct about what this paper is and isn't. It is not an engineering proposal anyone intends to start building. It's a theoretical exercise in what physicists call "megascale engineering" — testing the outer limits of what physics permits, using known energy and mass constraints, without worrying about present-day feasibility, cost, or political will. The scientific value lies in the constraint-checking: does the physics actually work, even if the technology and civilisation-scale coordination required don't exist yet?
Unsurprisingly, the proposal has drawn a mix of genuine scientific interest and open mockery online, with reactions ranging from serious discussion of the orbital mechanics involved to jokes about needing a stiff drink to process the scale of what's being suggested. Harry's broader body of work reportedly extends the same logic to other slow-motion threats — the eventual cessation of Earth's plate tectonics, long-term water loss, and shifts in the planet's axial tilt — treating each as an engineering problem with a theoretical solution, however many billions of years away.
Whether any of it is ever built is, by the paper's own logic, a question for a civilisation many thousands of generations removed from this one. What it offers today is something more modest but still valuable: a rigorous look at just how far the laws of physics actually stretch when the deadline is measured in billions of years rather than election cycles.