Building Large Radio Observatories: Space vs. the Far Side of the Moon

Key Takeaways

• Space-based radio telescopes extend interferometric baselines to tens of thousands of kilometers
• The lunar far side is the only location offering natural radio silence from Earth’s ionospheric noise
• Both approaches target low-frequency astronomy below 30 MHz, a band effectively invisible from Earth’s surface

A Universe in Frequencies

The universe is broadcasting on frequencies that Earth cannot hear. Below roughly 30 megahertz, the planet’s ionosphere acts as a reflective ceiling, bouncing most incoming radio waves back into space before they can reach any ground-based receiver. This isn’t a minor inconvenience for astronomers. The low-frequency radio band carries information about some of the oldest structures in the cosmos, about the period before the first stars formed, about the magnetic fields of planets orbiting distant suns. Listening at those frequencies means getting away from Earth entirely, and that constraint has driven two separate engineering strategies that are now both approaching the point where hardware gets built.

The first strategy is to place receivers in space. The second is to build on the far side of the Moon. Both approaches have serious engineering support behind them, both have attracted significant funding from multiple national space agencies, and both are moving from paper studies toward deployment planning. Comparing them isn’t simply a matter of cost or complexity. It touches on fundamental questions about what kind of science is actually possible, what location enables it, and which engineering trade-offs can realistically be managed.

What Makes Low-Frequency Radio Astronomy Different

Ground-based radio astronomy has a long and productive history. The Karl G. Jansky Very Large Array in New Mexico, the Green Bank Telescope in West Virginia, and the Parkes Observatory in Australia have contributed to discoveries ranging from pulsars to measurements of the cosmic microwave background. These instruments all operate in frequency bands that reach the ground without significant ionospheric distortion, generally above 100 MHz and extending into the gigahertz range.

Below about 30 MHz, the situation changes entirely. The ionosphere becomes opaque, and the same layer that allows shortwave radio broadcasts to bounce around the globe also prevents cosmic signals in that range from getting through. This means a large and scientifically rich swath of the electromagnetic spectrum has never been properly mapped. No large-scale survey of the sky at 5 MHz has ever been conducted from Earth’s surface, because it’s physically impossible. The ionosphere blocks it.

What’s down there scientifically? Quite a lot. The 21-centimeter hydrogen line is one of the most studied signals in radio astronomy, corresponding to 1420 MHz when emitted by nearby hydrogen. But when that same line is emitted by hydrogen gas in the very early universe, billions of years of cosmic expansion redshift the signal down into the low-frequency band. The hydrogen clouds that existed during the Epoch of Reionization and even earlier, during the Cosmic Dark Ages, emit signals that arrive today below 200 MHz, and the very earliest detectable hydrogen signatures fall below 30 MHz. Getting a clean view of the sky in that band is effectively a scientific prerequisite for understanding the universe’s first few hundred million years.

Planetary radio emissions represent another compelling target. Jupiter emits powerful bursts of radio energy below 40 MHz, generated by interactions between its magnetic field and its moon Io. Analogous signals from exoplanets, Jupiter-sized worlds orbiting other stars, would arrive in the same frequency range. Detecting those signals would allow astronomers to directly measure the magnetic field strengths of worlds that can’t otherwise be imaged or characterized.

The fundamental challenge isn’t building a sensitive enough detector. Simple dipole antennas can pick up low-frequency signals reasonably well. The challenge is finding a location where Earth’s own electromagnetic activity doesn’t drown out faint cosmic signals and where the ionosphere doesn’t scramble incoming data before it can be recorded.

Space-Based Radio Telescopes: The First Generation

The idea of extending radio telescope baselines into space predates the current discussion about lunar arrays by several decades. Very Long Baseline Interferometry (VLBI) works by combining signals from two or more widely separated antennas to create the resolution equivalent of a single dish as large as the distance between them. The longer the baseline, the finer the angular resolution. By the 1980s and 1990s, researchers were exploring seriously whether a radio antenna in orbit could extend a baseline far beyond Earth’s diameter, producing angular resolutions that no conceivable ground-based array could achieve.

HALCA, launched by JAXA in February 1997, was the first dedicated space VLBI satellite. The spacecraft carried an 8-meter parabolic dish antenna and operated as part of the VSOP (VLBI Space Observatory Programme). It flew in an elliptical orbit ranging from about 560 km to 21,400 km above Earth, giving it baselines with ground stations of up to roughly 30,000 km. HALCA successfully completed observations at 1.6, 5, and 22 GHz, producing some of the highest-resolution astronomical images ever made at those frequencies. The program demonstrated not only that space VLBI was technically feasible but that it could produce scientifically useful results in routine operation over a multi-year mission.

RadioAstron, launched by Russia in July 2011 under the designation Spektr-R, took the concept substantially further. The spacecraft carried a 10-meter deployable antenna and operated in a highly elliptical orbit with an apogee initially around 350,000 km, roughly comparable to the Earth-Moon distance. At its most distant, it provided VLBI baselines of nearly 400,000 km. RadioAstron operated at frequencies from 0.3 to 22 GHz and produced some remarkable results, including the highest-resolution radio images ever achieved of active galactic nucleus jets, quasars, and pulsars. The mission ran until 2019, when controllers lost contact after problems with the spacecraft’s attitude control system.

These two missions, HALCA and RadioAstron, proved several things at once. Space VLBI produces unprecedented angular resolution. The engineering challenges are manageable, at least for a single spacecraft in Earth orbit or nearby space. The scientific returns justify the investment. Neither mission, though, operated below 300 MHz. Both worked well above the ionospheric barrier. For low-frequency astronomy below 30 MHz, a different architecture is required.

The Ionospheric Barrier and Why Orbit Alone Doesn’t Solve It

A satellite in low Earth orbit, or even in a highly elliptical orbit like RadioAstron’s, can observe frequencies below 30 MHz without the direct ionospheric blocking that affects ground telescopes. In that narrow sense, any space-based antenna can access the low-frequency band. The problem is that Earth itself becomes the interference source.

The planet’s ionosphere doesn’t just block signals from space. It also reflects terrestrial radio transmissions outward, creating a shell of reflected noise that surrounds Earth in all directions. AM radio stations, shortwave broadcasts, lightning strikes generating atmospheric sferics, and auroral kilometric radiation all produce strong signals in the low-frequency band that propagate into space and arrive at a satellite antenna from every direction. A receiver in Earth orbit trying to observe a faint cosmic signal at 5 MHz would be competing with enormous interference from terrestrial and Earth-connected sources, even with no direct line of sight to the ground.

The Moon’s far side solves this problem geometrically. From any point on the hemisphere permanently facing away from Earth, the planet is always below the horizon. Earth’s ionospheric reflections, the lightning atmospherics, the shortwave broadcasts, none of it can arrive directly. The Moon’s own bulk provides roughly 60 dB of shielding from Earth. That’s a factor of one million in power. No electromagnetic shielding available to an orbiting spacecraft can match that figure.

This is the core argument for the lunar far side, and on the merits it’s compelling. The question is whether the engineering challenges of operating on the Moon are worth the scientific advantage of that radio silence, and the practical assessment of that question has been shifting in recent years as lunar exploration infrastructure has accelerated.

The Lunar Far Side: Geography and Radio Environment

The far side of the Moon never faces Earth due to tidal locking. The Moon completes one rotation on its axis in exactly the same time it takes to orbit Earth, so the same hemisphere always points toward the planet. For most of the lunar far side, Earth never rises above the horizon. An antenna placed near the center of the far side sees Earth at 0 degrees above the horizon at best, and usually below it.

The radio environment on the lunar far side has now been measured directly. Chang’e 4, which landed in the Von Kármán crater in January 2019 and remains the only spacecraft to have soft-landed on the lunar far side, carries the Netherlands-China Low-Frequency Explorer (NCLE). This instrument, developed at Radboud University in the Netherlands in partnership with Chinese institutions, has conducted observations in the frequency range from about 80 kHz to 80 MHz. Early published results confirmed the anticipated radio silence during lunar night, when the spacecraft is shielded from solar wind noise, and showed that Earth-related interference was suppressed to a degree consistent with theoretical models. The data from NCLE represents the first systematic in-situ characterization of the radio frequency environment on the lunar far side.

What NCLE confirmed, and what theoretical predictions had already suggested, is that the lunar far side during lunar night is one of the quietest radio environments accessible to any near-future mission. The solar wind generates a low background during the day, but during the approximately 14-day lunar night, the combination of the Moon’s shielding bulk and the absence of solar activity creates conditions where truly faint cosmic signals in the 1 to 30 MHz range could be detected with relatively modest antenna arrays. The challenge is building hardware that can survive that environment and operate reliably throughout it.

The lunar night lasts roughly 14 Earth days at any given location, with surface temperatures dropping to around -170 degrees Celsius. No sunlight means no solar power. The extreme thermal cycling between day and night, which swings from about +120 degrees Celsius to -170 degrees Celsius, creates mechanical stress on every material in any instrument. These are the environmental realities that have defined the engineering requirements for every far-side observatory concept proposed to date.

Proposed Lunar Far Side Observatories

Several serious proposals have been developed since roughly 2010, and a few have progressed from initial concept to funded study or early hardware development.

FARSIDE

The FARSIDE concept, which stands for Farside Array for Radio Science Investigations of the Dark ages and Exoplanets, was proposed as a NASA mission concept and received support for formal study. The design calls for 128 dual-polarization dipole antennas deployed across a roughly 10 km area on the lunar far side. Each antenna connects to a central base station via cables laid on the surface by a rover. The array would operate between 0.1 and 40 MHz, covering the full frequency range blocked by Earth’s ionosphere and extending significantly below it.

FARSIDE’s primary scientific targets are the Cosmic Dark Ages, the period from about 380,000 years after the Big Bang to roughly 150 million years later, when hydrogen gas filled the universe but no stars had yet formed. During this period, the hydrogen gas should produce a distinctive pattern of emission that, redshifted to the frequencies FARSIDE would observe, carries information about the density fluctuations that eventually grew into galaxies. No observatory has ever directly detected this signal, because it requires access to frequencies below about 100 MHz and exceptional sensitivity in a radio-quiet environment.

The secondary target, exoplanet radio emissions, is potentially even more immediately productive. FARSIDE’s design simulations suggest it could detect radio bursts from Jupiter-like planets at distances of up to about 20 parsecs, roughly 65 light-years from Earth. That may not sound like much, but it represents an entirely new detection channel for characterizing the magnetic environments of planets orbiting nearby stars.

The logistical requirements for FARSIDE are significant. Deploying 128 antennas and their connecting cables across a 10 km area requires either a highly capable rover or multiple rovers working sequentially. Power for the base station during lunar night would require a small nuclear power source, as solar panels produce nothing for the 14-day night period. Communications from the far side require a relay satellite in a halo orbit around the Earth-Moon L2 Lagrange point, approximately 61,500 km beyond the Moon, since there’s no direct line of sight to Earth.

Lunar Crater Radio Telescope

The Lunar Crater Radio Telescope (LCRT) concept, developed by Saptarshi Bandyopadhyay at NASA‘s Jet Propulsion Laboratory and funded through the NASA Innovative Advanced Concepts (NIAC) program, takes a different approach. Instead of a distributed phased array, LCRT proposes using an existing lunar crater as the natural structural bowl for a single large parabolic reflector. The dish would be constructed by draping a wire mesh across the crater’s inner walls, with DuAxel robotic systems doing the installation work.

The target crater would be about 3 to 5 km in diameter, giving a dish of comparable size. For reference, the Arecibo Observatory, which collapsed in December 2020, had a 305-meter dish. A 3 km lunar crater dish would be nearly ten times larger in diameter, and roughly 100 times larger in collecting area. The LCRT concept envisions operating at frequencies from about 1 to 100 MHz, covering the full low-frequency band in a single instrument.

There are obvious trade-offs compared to the distributed array approach. A single dish provides high sensitivity in one direction but can only observe the sky region above it, and a crater-based dish can’t be steered mechanically. The crater’s shape determines the pointing direction, which for a crater near the lunar equator would be roughly toward the zenith. An array like FARSIDE can, by processing signals from multiple antennas with carefully calculated time delays, synthesize a beam that can be steered electronically across a wide sky region, giving much greater flexibility for survey work.

The LCRT received NIAC Phase II funding, which indicated that NASA considered it sufficiently credible and novel to warrant detailed engineering study. Whether it progresses beyond study phase depends on cost estimates, the availability of mature robotic deployment systems, and competition from other lunar science priorities that are themselves accumulating.

LuSEE-Night

LuSEE-Night, the Lunar Surface Electromagnetics Experiment-Night, represents a more near-term step than either FARSIDE or LCRT. It’s a joint NASA and Department of Energy project designed to land a modest radio receiver package on the lunar far side and survive an entire lunar night, the first time any NASA-led instrument has attempted that specific thermal challenge. LuSEE-Night is not a full observatory. It’s a pathfinder, intended to demonstrate that electronics can operate through the extreme thermal environment of a lunar far-side night and to characterize the low-frequency radio environment in greater detail than NCLE has so far provided.

The science package includes four monopole antennas with approximately 2.3-meter deployable elements and can measure signals between about 0.1 and 50 MHz. If it operates as designed, it would provide the most detailed spectrum of the lunar far-side radio environment ever measured, complementing the NCLE data from Chang’e 4 with finer frequency resolution and a broader operating range. LuSEE-Night’s survival of a full lunar night would also validate the thermal engineering approach that any larger observatory would need to use, a demonstration with value well beyond the instrument’s own science return.

Space-Based Arrays: The Distributed Approach

While lunar surface concepts have attracted substantial recent attention, serious work has also gone into space-based arrays designed specifically for low-frequency astronomy. These differ fundamentally from single-dish missions like HALCA or RadioAstron and instead use swarms or constellations of simple antennas flying in coordinated arrangements.

The key engineering advantage of a space-based array is deployment flexibility. Antennas in orbit aren’t fixed to any surface, and their baselines can in principle be arranged in configurations optimized for specific science targets. A cluster of small satellites, each carrying a dipole antenna, could form different geometric arrangements over time, changing sensitivity patterns or expanding baselines as missions evolve. In theory, the aperture of such an array could grow beyond what any surface-based instrument could physically accommodate.

The Netherlands Institute for Radio Astronomy (ASTRON) and academic partners in Europe have studied several distributed space array concepts for low-frequency work. The general consensus from those studies is that space-based low-frequency arrays are technically possible but that the combination of simultaneous signal timing across multiple spacecraft, individual antenna calibration, and filtering of RFI from Earth’s electromagnetic environment makes them substantially more complex than surface-based alternatives for equivalent sensitivity.

China’s approach to the problem is worth noting separately. CNSA has discussed placing a cluster of small satellites in lunar orbit as part of future Chang’e program missions, building on the experience with NCLE on Chang’e 4. These satellites would form a space-based low-frequency array, with the Moon providing partial shielding when the constellation is positioned on the anti-Earth side of its orbit. This architecture is interesting because it’s neither fully a space-based array nor a lunar surface array. The satellites are in space and offer deployment flexibility, but their orbital geometry uses the Moon for partial shielding. It’s a reasonable compromise given the engineering challenges of full far-side surface deployment, and it uses infrastructure China is already developing.

A Direct Comparison

Comparing space-based and lunar far-side approaches requires separating the question of scientific capability from the question of engineering feasibility, and being clear about which science goals each architecture actually serves.

On scientific capability for low-frequency observations, the far side of the Moon wins, and it isn’t close. The radio shielding provided by the Moon’s bulk is unique and cannot be replicated by any space-based architecture without also blocking the cosmic signals you’re trying to detect. If the goal is measuring the redshifted 21-cm signal from the Cosmic Dark Ages, surveying the sky below 10 MHz, or detecting exoplanet radio bursts, a surface array on the far side is the right instrument. An orbiting array in any Earth-accessible orbit will always be competing with interference from Earth’s leaky ionosphere and terrestrial transmitters, regardless of how well it’s engineered.

For high-angular-resolution imaging at higher frequencies, the space-based approach as demonstrated by RadioAstron and HALCA still has no competitor. Surface arrays, whether on Earth or on the Moon, are limited in baseline length by physical geography. A space-based antenna in a very elongated orbit achieves baselines of hundreds of thousands of kilometers that no surface arrangement can replicate.

The following table summarizes key characteristics of the representative missions and concepts discussed throughout this article.

The Role of Interferometry

One aspect of this comparison that doesn’t get adequate attention is interferometry, specifically how differently it works in the two settings and what that means for what each type of observatory actually measures.

Ground-based VLBI with a space extension, as demonstrated by HALCA and RadioAstron, works because all the antennas share a common timing and data recording infrastructure. Each station records incoming signals alongside extremely precise timing data from hydrogen masers synchronized to global standards. The recordings are then correlated in software to extract the interference pattern that gives the array its effective resolution. The Earth-based stations can be large dishes of 25 to 100 meters or more. The space element is one antenna, providing one very long baseline. The result is extraordinary angular resolution but limited sensitivity, because the space antenna is physically small and collects little total signal power.

A lunar far-side array like FARSIDE would work differently, closer in principle to the Square Kilometre Array under construction in South Africa and Australia than to a VLBI network. With 128 antennas distributed over 10 km, it functions as a dense aperture synthesis array. The short baselines between nearby antennas provide sensitivity to extended, diffuse emission. The longer baselines between antennas at opposite ends of the array provide angular resolution for compact structures. The maximum baseline is 10 km, far shorter than what RadioAstron achieved, but FARSIDE isn’t competing on angular resolution. It’s competing on sensitivity to faint, extended emission from the very early universe, emission that couldn’t be detected any other way because no other location provides the necessary combination of frequency access and radio quiet.

An orbiting low-frequency array could achieve much longer baselines, potentially extending to the Earth-Moon distance or beyond. The limitation is collecting area. Sensitivity, the ability to detect faint signals, depends on how much electromagnetic radiation the array actually captures, which means antenna size multiplied by antenna count. A swarm of CubeSats, each carrying a small dipole antenna, adds up to very little collecting area regardless of baseline length. Combining high collecting area with very long baselines requires many large antennas flying in precise formation, a capability that doesn’t exist at meaningful scale for radio astronomy.

This creates a real asymmetry. Space-based arrays have a theoretical path to angular resolution that lunar surface arrays can’t match for the same baseline length. Lunar surface arrays have a demonstrated path to sensitivity in a radio-quiet environment that space arrays can’t match at any accessible frequency below 30 MHz. The two approaches may not be competing for the same scientific territory at all.

Getting the Data Back

Any observatory on the lunar far side faces a communications challenge that space-based systems don’t encounter in the same way. From the far side, there’s no line of sight to Earth. Getting data back requires a relay satellite in a halo orbit around the Earth-Moon L2 Lagrange point or some comparable relay architecture.

China’s Queqiao relay satellite, launched in May 2018 specifically to support Chang’e 4, proved the concept works at operational scale. A second relay, Queqiao-2, launched in March 2024, provides improved coverage and higher data rates for subsequent Chinese missions to the far side and polar regions. Both satellites demonstrate that the relay infrastructure problem is solvable with existing technology. ESA‘s Lunar Pathfinder mission was being developed to provide independent communication and navigation services to lunar surface missions, including those operating on the far side, as part of a broader commercial lunar infrastructure strategy.

The bandwidth constraint is the harder problem for a large observatory. A full-scale array generating continuous scientific data from 128 antennas can produce data volumes that strain relay satellite link capacity. FARSIDE’s design addresses this through on-board correlation and compression, where the raw antenna signals are partially processed on the lunar surface before being transmitted to Earth. This reduces the required data rate to something a relay satellite link can handle, but it adds processing electronics that must survive the thermal environment of the lunar far side. Each added electronic subsystem that must survive a lunar night is a non-trivial engineering problem.

A space-based array faces its own version of the data return challenge. Multiple spacecraft each recording low-frequency signals and combining them interferometrically either requires a high-bandwidth inter-satellite link, a central processing node that the spacecraft fly past, or a return to Earth-based downlinks, which reintroduces proximity to the interference environment. There’s no fully clean solution in either architecture.

Power and Thermal Survival

The Moon’s surface environment is demanding in specific ways that differ from what Earth-orbiting spacecraft face. There’s no atmospheric drag, no orbital period thermal cycling, and the regolith provides thermal inertia. But the roughly 14-day lunar night imposes a power generation problem with no solar solution.

Any observatory that needs to operate continuously through lunar night, which low-frequency radio astronomy prefers because the night is when solar wind background noise is lowest, needs a non-solar power source. A radioisotope power system or a small fission reactor is the credible engineering option at the power levels an observatory would need.

NASA‘s Kilopower project demonstrated a small fission reactor system called KRUSTY in 2018, producing up to 10 kilowatts of electrical power from a system designed to operate for a decade without maintenance. The follow-on Fission Surface Power program targets the 10 to 40 kilowatt range for lunar and Mars surface applications, with flight system development ongoing as of 2025. The technology is real and approaching flight readiness, though no such system has yet operated on the lunar surface. For a FARSIDE-class observatory, this power level would be more than sufficient to run array electronics and the data relay system through continuous lunar night operations.

Surface-mounted nuclear power has some advantages over space-based nuclear systems that don’t get enough acknowledgment in the comparison. A surface unit can radiate waste heat into the regolith, has no orbital mechanics concerns associated with nuclear material, and can in principle be accessed for servicing by future robotic or crewed missions. A nuclear-powered spacecraft is fixed for its operational lifetime with no servicing option, carrying the same mass and cost overhead as any other spacecraft but with no maintenance path.

The Seismic and Electromagnetic Environment

Radio observatories are sensitive instruments, and their operation can be disrupted by physical and electromagnetic disturbances that might not be immediately obvious from the outside.

For a lunar far-side array, both types of disturbance are relevant. Moonquakes are real and documented. The Apollo seismometers that operated on the lunar surface between 1969 and 1977 recorded four categories of seismic activity: deep moonquakes at 700 to 1200 km depth, shallow moonquakes in the upper crust, meteorite impacts, and thermal quakes from the day-night temperature cycle. Shallow moonquakes were the most energetic, with some reaching magnitude 5.5 on a comparable scale to Earth earthquakes. A seismic event of that magnitude on the lunar far side would disturb surface-deployed antenna cables, potentially shifting antenna positions and corrupting the calibration that aperture synthesis imaging depends on.

The electromagnetic environment includes some local sources that must be understood for precision calibration. The interaction of the solar wind with the lunar surface generates faint radio emissions. Charged particle interactions with regolith create low-level electrostatic noise. These effects are small compared to the shielding benefit from Earth, but they need detailed characterization before a precision cosmological observatory can separate them from the signals it’s hunting.

A space-based array carries different disturbance sources. Spacecraft attitude control thrusters generate vibrations. The power systems of the spacecraft itself, particularly switching regulators, produce interference across a broad frequency range. CubeSat power electronics are notoriously noisy at radio frequencies. Any space-based low-frequency array requires careful electromagnetic isolation of its own subsystems, which adds mass, complexity, and design risk. A surface array can physically separate its electronics from the antennas with long cables, keeping the interference sources away from the sensitive receivers.

What’s Actually Being Built Right Now

The Square Kilometre Array Observatory (SKAO) is currently under construction in two locations: the SKA-Low array in the Murchison region of Western Australia, using hundreds of thousands of log-periodic dipole antennas operating from 50 to 350 MHz, and the SKA-Mid array in South Africa’s Karoo desert, using parabolic dishes from 350 MHz to 15.3 GHz. SKA-Low’s low frequency limit of 50 MHz approaches but doesn’t enter the sub-30 MHz band that requires space or lunar deployment. It will produce data and experience directly relevant to interpreting what a lunar array would see.

LOFAR, operated by ASTRON and headquartered in the Netherlands with stations across Europe, currently operates from 10 to 240 MHz and represents the closest ground-based approach to the true low-frequency regime. Its core is in Exloo in the northeastern Netherlands. LOFAR’s stated lower limit of 10 MHz actually crosses the 30 MHz threshold, though ionospheric effects degrade sensitivity substantially below about 20 MHz. LOFAR has produced important early-universe science in the 120 to 200 MHz range and its software correlation infrastructure, which processes several terabits of data per second from hundreds of distributed antenna stations, has directly informed the design of proposed lunar arrays. In 2021, researchers using LOFAR data published a candidate detection of radio emission from the stellar system Tau Boötis, suggesting that exoplanet-driven radio bursts at low frequencies might be at the edge of current ground-based detectability.

This existing body of LOFAR work sets an interesting baseline. If the best current ground-based instrument can find candidate exoplanet signals at the edge of its sensitivity, a purpose-built observatory in a radio-quiet environment with no ionospheric degradation would push the detection threshold substantially further. The scientific motivation for the next step isn’t speculative; it’s backed by data from operating systems.

The Geopolitical Dimension

The discussion about who builds the next generation of large radio observatories is inseparable from the broader geopolitics of the new lunar exploration period. China’s Chang’e program has already demonstrated far-side landing capability with Chang’e 4, and China has outlined plans for Chang’e 7 and Chang’e 8, with the latter designed to test construction and resource utilization techniques near the lunar south pole as part of the International Lunar Research Station program. Chinese scientific consortiums have expressed interest in deploying radio astronomy instruments as part of these missions, building on the NCLE experience.

NASA‘s Artemis program, while centered on crewed surface access at the lunar south pole, has also supported a range of scientific payloads through the Commercial Lunar Payload Services (CLPS) program. CLPS has now delivered hardware to the lunar surface through multiple contractors, and the program’s manifest includes far-side destinations. LuSEE-Night, as a CLPS-eligible payload, could reach the far side within a relatively near-term timeframe if funding and manifest positions align.

ESA has been a consistent partner in lunar communications infrastructure projects that would underpin any European-led or European-partnered scientific observatory. ESA’s participation in the Lunar Gateway space station program, its work on the Lunar Pathfinder communications relay mission, and its cooperation with NASA on Artemis all position European institutions to contribute instruments to a far-side radio observatory even if ESA doesn’t lead such a mission independently.

The result is a competitive and partially cooperative environment in which multiple national space programs are simultaneously developing pieces of the capability needed for a large lunar far-side observatory. Whether those pieces get assembled into a single coordinated international facility or whether multiple independent instruments land at separate sites is partly a scientific planning question and partly a geopolitical one that depends on funding cycles and program priorities not yet settled.

The Science in Detail

Specificity matters when describing what these observatories would actually measure. The language around cosmological radio astronomy can sound abstract, so it’s worth grounding it in concrete observational goals.

The primary target for most far-side proposals is the redshifted hydrogen signal from the Cosmic Dark Ages. Before the first stars formed, the universe was filled with neutral hydrogen gas at a temperature slightly above or below the cosmic microwave background. That small temperature contrast produced an absorption or emission signal at the 21-cm line. As the universe expanded over 13 billion years, the apparent frequency of that signal from the Dark Ages era has shifted to roughly 10 to 100 MHz. Detecting it would provide the first direct observational measurement of matter density and temperature in the universe before any structure had formed, a data point that current cosmological models predict but that nobody has ever observed. The detection is not easy even from the lunar far side. The expected signal is somewhere between 1 and 100 millikelvin in amplitude, sitting on top of a galactic synchrotron background that is orders of magnitude brighter. Extracting it requires precise calibration and excellent understanding of foreground emission, which is why pathfinders like LuSEE-Night are so valuable.

The exoplanet radio emission target is potentially more immediately productive. The interaction between a planetary magnetic field and stellar wind drives cyclotron radio emission at frequencies directly proportional to the magnetic field strength at the planet’s surface. For Earth, this emission falls around 30 kHz. For Jupiter, it falls below 40 MHz. An exoplanet with Jupiter’s mass orbiting a star 20 parsecs away would produce analogous emission that a sensitive far-side array could potentially detect directly. This would be the first direct magnetic characterization of an extrasolar planet, a qualitatively new category of exoplanet data.

There’s also a class of transient events, including solar radio bursts, interplanetary plasma structures, and potentially magnetospheric activity on other planets in the solar system, that a low-frequency array could monitor continuously in ways no Earth-based instrument can. The scientific return on a lunar far-side observatory is broad and wide-ranging in ways that extend well beyond the headlining cosmological goal.

The Question of Timing

There’s a reasonable engineering argument that building a large-scale low-frequency observatory on the lunar far side is premature. The relay infrastructure for far-side communications is nascent, though Queqiao-2 has proven the basic concept. Nuclear surface power systems haven’t flown to the Moon. Robotic deployment of a 10 km array across the lunar surface hasn’t been demonstrated at anything like the required complexity. The total cost of a FARSIDE-class facility, once transportation, deployment, relay satellites, nuclear power, and operations are included, would likely be in the range of several billion US dollars, comparable to a flagship NASA space telescope.

The scientific case has been building for decades and is now compelling enough that multiple national agencies are funding study work seriously. The engineering gap between current capability and what’s needed for a FARSIDE-class mission is real but not categorically unbridgeable. Every required technology has a development program running somewhere. The gap is closing.

There’s also an argument that deserves more attention in the planning community: the window for a scientifically pristine radio environment on the lunar far side may not remain open indefinitely. Increased human and robotic activity on the Moon will bring more radio frequency emitters with it. Lunar surface vehicles, communication systems, surface power infrastructure, and future crewed habitats all generate electromagnetic emissions across a broad frequency range. A sufficiently sensitive low-frequency observatory could find its operating environment degraded by nearby lunar activity, much the way ground-based radio telescopes have seen their environments degrade as terrestrial wireless infrastructure expanded.

Whether that prospect is a near-term concern or a distant worry depends entirely on how quickly lunar development accelerates, and that timeline is not yet settled. The pace of commercial lunar activity between 2020 and 2025 has exceeded most projections from a decade ago, and it continues to accelerate. The argument that a far-side radio observatory needs to establish itself before the lunar electromagnetic environment gets complicated is one that deserves to be taken seriously in programmatic planning, even if it’s not yet urgent.

Summary

Space-based radio astronomy, as demonstrated by HALCA in 1997 and RadioAstron from 2011 to 2019, established that moving telescope infrastructure off the ground produces results that can’t be obtained any other way. For high-frequency VLBI, orbit is the right environment, and the results from both missions proved it. For low-frequency astronomy below 30 MHz, the calculus is different. Orbit doesn’t provide the required radio silence. The Moon’s far side does, and no other accessible location does.

The proposals now in various stages of development, FARSIDE, LCRT, LuSEE-Night, and China’s lunar orbit constellation plans, represent a serious collective attempt to access that environment and the science it enables. Each makes different engineering trade-offs. FARSIDE offers survey capability and sky coverage through electronic beam steering. LCRT offers extraordinary collecting area in a fixed pointing direction. LuSEE-Night offers near-term pathfinder feasibility to reduce risk for everything that follows. China’s orbital constellation approach offers a middle path that uses existing mission infrastructure.

The full vision, an observatory capable of detecting Dark Ages hydrogen emission or confirming exoplanet radio bursts from multiple nearby stars, remains a project measured in decades and measured in billions of dollars. The scientific return is proportional. A direct observational window into the universe’s first hundred million years would be among the most significant advances in observational cosmology since the detection of the cosmic microwave background.

What remains uncertain is whether any of these proposals will attract the sustained institutional commitment needed to go from concept to operational instrument. The history of proposed space and lunar observatories includes many compelling science cases that never found their funding. What’s different now is that the Moon is returning to active exploration for the first time since Apollo, and the relay satellites, nuclear power systems, and robotic surface mobility that a far-side radio observatory needs are all being developed for other reasons. That convergence creates an opening. Whether the scientific community and the space agencies that fund it can act on that opening before the lunar electromagnetic environment becomes complicated is the real question of timing that the next decade will answer.

Appendix: Top 10 Questions Answered in This Article

Why can’t ground-based radio telescopes observe frequencies below 30 MHz?

Earth’s ionosphere reflects and absorbs radio waves below roughly 30 MHz, preventing cosmic signals in that frequency range from reaching the ground. This makes a substantial portion of the electromagnetic spectrum inaccessible from any surface-based observatory on Earth, regardless of its size or location.

What is the primary scientific advantage of placing a radio telescope on the Moon’s far side?

The Moon’s bulk provides approximately 60 dB of shielding from Earth’s ionospheric reflections and terrestrial radio transmissions, creating a radio-quiet environment unmatched anywhere else accessible to near-future missions. This shielding is essential for detecting faint cosmic signals in the low-frequency band between 0.1 and 30 MHz.

What did the RadioAstron mission demonstrate about space-based radio astronomy?

RadioAstron, launched in 2011 and operating until 2019, demonstrated that a radio antenna in a highly elliptical orbit extending to approximately 350,000 km could produce VLBI baselines of extraordinary length, yielding the highest-resolution radio images ever achieved of AGN jets, pulsars, and quasars. It confirmed that space-based interferometry is scientifically productive and operationally feasible at that scale.

What are the Cosmic Dark Ages and why are they a target for low-frequency radio observatories?

The Cosmic Dark Ages refers to the period from about 380,000 to roughly 150 million years after the Big Bang, when neutral hydrogen filled the universe before the first stars formed. The 21-cm hydrogen line emitted during this period is redshifted to frequencies below 30 MHz today, placing it exactly in the band accessible only from space or the lunar far side.

What is the FARSIDE mission concept?

FARSIDE is a NASA mission concept proposing 128 dual-polarization dipole antennas deployed over a roughly 10 km area on the lunar far side, operating from 0.1 to 40 MHz. Its primary targets are the Cosmic Dark Ages hydrogen signal and radio emissions from magnetized exoplanets, and it requires a relay satellite at the Earth-Moon L2 point, a nuclear power source, and rover-based deployment.

How does the Lunar Crater Radio Telescope differ from distributed array concepts like FARSIDE?

The Lunar Crater Radio Telescope proposes using a natural lunar crater as the structural bowl for a single large parabolic dish, with wire mesh draped across the crater walls by robotic systems, potentially creating a dish several kilometers in diameter. Unlike a phased array, it offers enormous collecting area in a fixed pointing direction rather than electronically steerable sky coverage.

What has China’s Chang’e 4 mission contributed to lunar far-side radio astronomy?

Chang’e 4 carries the Netherlands-China Low-Frequency Explorer, which provided the first direct in-situ measurements of the radio frequency environment on the lunar far side. The data confirmed anticipated radio silence during lunar night and established a baseline characterization of the electromagnetic environment for future observatory planning.

Why is power generation a particular challenge for lunar far-side observatories?

The lunar far side experiences roughly 14 days of continuous darkness during each lunar night, making solar power generation unavailable for extended periods. Low-frequency radio observations are most scientifically productive during lunar night when solar wind background noise is lowest, so any operational observatory needs a nuclear power source such as a radioisotope generator or small fission reactor.

What is LOFAR and how does it relate to proposed lunar far-side observatories?

LOFAR is a distributed radio telescope array operated by ASTRON in the Netherlands with stations across Europe, operating from 10 to 240 MHz. Its software correlation architecture and distributed antenna processing experience have directly informed the design of proposed lunar and space-based low-frequency arrays, and its candidate exoplanet radio detection in 2021 provided empirical motivation for a dedicated low-frequency facility.

What relay satellite infrastructure does a lunar far-side observatory require?

A relay satellite in a halo orbit around the Earth-Moon L2 Lagrange point, approximately 61,500 km beyond the Moon, is required to maintain communication between a far-side surface facility and Earth. China’s Queqiao and Queqiao-2 satellites have demonstrated this architecture operationally in support of Chang’e 4, and ESA’s Lunar Pathfinder mission was in development to provide similar independent services for future far-side missions.