Miniaturization and new technologies in ESA sounding missions AI-pedia

Atmospheric sounding is the measurement of physical properties—such as temperature, moisture, pressure, and chemical composition—at varying altitudes throughout an atmospheric column. Sounding data is a fundamental input for numerical weather prediction (NWP), climate modeling, and atmospheric chemistry research. For decades, satellite-based sounding required multi-tonne spacecraft operating in low Earth orbit or geostationary orbit. These spacecraft carried massive instruments demanding hundreds of watts of electrical power, heavy mechanical cryocoolers, and large structural benches engineered to prevent thermal expansion.

The European Space Agency (ESA), working alongside meteorological organizations such as EUMETSAT, has historically developed sounding capabilities through flagship platforms. Notable examples include Envisat, the MetOp series carrying the Infrared Atmospheric Sounding Interferometer (IASI), and the MetOp Second Generation (MetOp-SG) satellites. While these missions set international benchmarks for radiometric and spectral accuracy, their multi-billion-euro budgets and decades-long development cycles make rapid iteration difficult and limit revisit frequency over any single geographic coordinate.

In response to emerging meteorological requirements and rapid advances in semiconductor physics, ESA has shifted significant resources toward miniaturization and alternative instrument architectures. By replacing bulk optics with flat metasurfaces, integrating free-space interferometers onto photonic integrated circuits, adopting high operating temperature detectors, and reducing microwave receivers to single-chip packages, sounding payloads can now be deployed on small satellites and CubeSats. This paradigm shift enables distributed satellite constellations that provide high temporal resolution while lowering mission risk and development costs.

Historical Context and the Need for Miniaturization

Spaceborne atmospheric sounding began with simple radiometers that measured radiation across a few discrete spectral channels. As atmospheric models advanced, meteorologists demanded higher spectral resolution. High spectral resolution allows scientists to resolve individual absorption lines of atmospheric gases such as carbon dioxide, water vapor, and ozone, which in turn permits fine vertical profiling of the atmosphere.

Achieving this performance historically required Fourier Transform Spectrometers (FTS) and high-frequency microwave radiometers built with heavy, rigid hardware. The first-generation IASI instruments, developed by the French space agency (CNES) and flying on ESA/EUMETSAT MetOp satellites, each weigh over 200 kilograms and consume more than 200 watts of electrical power. The instrument relies on a Michelson interferometer with moving corner-cube retroreflectors, massive off-axis parabolic mirrors, and a dedicated active cryogenic cooling system to lower its mercury cadmium telluride (MCT) photodetectors to roughly 80 Kelvin.

Large spacecraft architectures create several operational limitations:

• Low temporal revisit rates: A single polar-orbiting flagship satellite generally samples a specific location on Earth twice per day. Severe, fast-evolving weather phenomena—such as convective storms and tropical cyclones—develop on hourly timescales and can escape timely observation.
• Prolonged development timelines: Flagship missions typically require 10 to 15 years from initial conception to launch. This long gestation period delays the space qualification and operational use of modern hardware innovations.
• High single-point mission risk: Concentrating numerous high-value instruments on a single bus creates a catastrophic vulnerability. A launcher failure or on-orbit bus anomaly can compromise an entire generation of national meteorological forecasting assets.

To overcome these limits, numerical weather prediction centers requested denser temporal sampling. Providing global revisit times of 30 to 60 minutes using multi-tonne satellites is financially impossible. Consequently, ESA established programs to investigate whether modern physics and micro-engineering could shrink sounding payloads down to small satellite scales—typically platforms weighing between 20 and 150 kilograms—without crippling their radiometric sensitivity.

Core Technological Innovations

The miniaturization of sounding instruments is not merely a matter of building smaller copies of legacy designs. Shrinking traditional optical and mechanical assemblies often causes severe physical bottlenecks, such as optical throughput reduction and diffraction limits. True miniaturization has instead required new physical principles and manufacturing techniques across photodetectors, optical elements, waveguiding, and high-frequency electronics.

High Operating Temperature Photodetectors and Type-II Superlattices

Infrared sounders measure atmospheric thermal emissions in the medium-wave infrared (MWIR, 3 to 8 micrometers) and long-wave infrared (LWIR, 8 to 15 micrometers) regions. At these wavelengths, the bandgap of the detector material must be narrow. Historically, thermal excitation at room temperature easily knocked electrons across this small bandgap, creating massive dark current (spurious electrical current generated by thermal energy in the absence of light) that completely overwhelmed the faint atmospheric signal.

To suppress dark current, traditional infrared detectors based on mercury cadmium telluride (HgCdTe or MCT) must be cooled to between 60 K and 80 K (-213°C to -193°C). In space, maintaining these temperatures requires either large passive radiator shields pointed permanently at cold space or active closed-cycle mechanical cryocoolers (such as Stirling or pulse-tube coolers). Mechanical cryocoolers are heavy, draw dozens to hundreds of watts of power, and introduce mechanical micro-vibrations that degrade optical pointing.

ESA has supported research into High Operating Temperature (HOT) photodetectors, specifically Type-II Superlattice (T2SL) devices. A T2SL is an artificially engineered semiconductor composed of alternating, extremely thin layers (often only a few nanometers thick) of materials such as indium arsenide (InAs), gallium antimonide (GaSb), or indium arsenide antimonide (InAsSb). In a Type-II band alignment, the conduction band minimum is located in one layer while the valence band maximum is in the adjacent layer.

This synthetic band structure allows physicists to design the effective bandgap precisely by adjusting layer thicknesses rather than altering bulk chemical alloy compositions. More importantly, T2SL architectures allow engineers to suppress Auger recombination, an intrinsic semiconductor process where an electron and hole recombine and transfer their energy to a third charge carrier instead of emitting or absorbing light. Suppressing Auger processes dramatically reduces thermal dark current, allowing T2SL detectors to achieve radiometric performance at 120 K to 160 K that matches what legacy MCT detectors achieved at 70 K to 80 K.

Raising the detector operating temperature to 130 K or higher yields immense engineering advantages. The required cooling power drops according to thermodynamic Carnot efficiency principles. At these higher temperatures, miniature, low-power single-stage thermoelectric coolers (solid-state Peltier devices) or very small, lightweight passive radiators become viable, eliminating the mass, volume, and vibration associated with mechanical cryocoolers.

Optical Metasurfaces and Flat Optics

Traditional optical systems in infrared and visible atmospheric sounders rely on sequences of precision-ground, bulk glass or crystal lenses and reflective mirrors. These optical elements ensure light is collected, focused, and corrected for chromatic and spherical aberrations. Such assemblies are inherently heavy, consume significant spatial volume, and require robust mechanical mounts to survive launch vibrations.

Metasurfaces are planar, artificial optical structures composed of sub-wavelength nanostructures arranged on a flat surface. These structures—often called "meta-atoms"—take the form of microscopic pillars, blocks, or ridges made from high-refractive-index dielectric materials such as silicon, titanium dioxide, or gallium nitride. Each meta-atom acts as a localized phase-shifter. By varying the diameter, geometry, and placement of these pillars across a substrate, engineers can manipulate the wavefront, polarization, phase, and amplitude of incoming light across a surface that is less than a micrometer thick.

For spaceborne sounding payloads, metasurfaces offer several distinct benefits:

• Extreme mass and volume reduction: A single flat metalens can replicate the focusing power of a multi-element refractive lens system, reducing the optical train's weight and length by up to 90 percent.
• Integrated dispersion and filtering: Metasurfaces can be patterned to simultaneously focus light and disperse it spectrally, functioning as both an imaging objective and a diffraction grating.
• Elimination of thermal optical defocusing: Bulky optical lenses expand and contract unevenly during spacecraft thermal cycles, shifting the focal point. Flat metasurfaces patterned on low-expansion substrates significantly reduce thermal defocusing across wide orbital temperature swings.
• Direct focal-plane integration: Metasurfaces can be lithographically fabricated directly onto the protective windows or surface arrays of photodetector chips, eliminating free-space optical gaps entirely.

Photonic Integrated Circuits

In standard Fourier Transform Spectrometers, an incoming beam of light is divided into two paths by a beam splitter, directed toward mirrors (one fixed, one moving), and recombined to generate an interference pattern (an interferogram). The spectrum of the atmosphere is recovered by taking the mathematical Fourier transform of this interferogram. The mechanical translation mechanisms, optical mirrors, and alignment systems required for this process are physically large and sensitive to mechanical shock.

Photonic Integrated Circuits (PICs) replace free-space optical paths with microscopic waveguides fabricated directly on semiconductor chips using silicon photonics, silicon nitride (SiN), or indium phosphide (InP) platforms. Light captured by an external collection aperture is coupled directly into the chip's waveguides. Inside the chip, directional couplers, optical delay lines, and phase modulators guide and manipulate the light waves on a micron scale.

Using PIC technology, an entire interferometer can be built on a chip measuring a few square centimeters. Static Fourier transform architectures, such as spatial heterodyne spectrometers (SHS) or arrayed waveguide gratings (AWGs), split the guided light into dozens or hundreds of discrete optical channels with progressively staggered optical path delays. These channels interfere with one another without requiring any moving mechanical parts. The resulting interference fringe pattern is recorded directly by a linear photodetector array bonded to the chip edge, creating a solid-state spectrometer with high vibration resistance.

Monolithic Microwave Integrated Circuits and Sub-Millimeter Radiometry

While infrared sounders deliver exceptional vertical resolution, infrared radiation cannot penetrate thick clouds. Microwave sounders are indispensable for global weather forecasting because microwave radiation penetrates deep into non-precipitating clouds, permitting continuous profiling of the troposphere regardless of cloud cover.

Legacy microwave sounders—such as the Advanced Microwave Sounding Unit (AMSU-A) and the Microwave Humidity Sounder (MHS)—relied on bulky, machined metal hollow waveguides, discrete horn antennas, and separate mixer modules. These metal parts contributed heavily to instrument weight and thermal inertia. Miniaturization in this domain has been enabled by developments in Monolithic Microwave Integrated Circuits (MMICs) based on gallium arsenide (GaAs) and indium phosphide (InP) semiconductor substrates, with high-electron-mobility transistor (HEMT) technologies.

Modern InP HEMT low-noise amplifiers (LNAs) operate at frequencies well above 100 GHz, extending beyond 300 GHz and into the sub-millimeter regime. These components amplify faint atmospheric signals while introducing very low intrinsic electronic noise. An entire microwave receiver frontend—including the radio frequency amplifier, local oscillator, mixer, and intermediate-frequency filter—can now be integrated onto a single semiconductor die smaller than a fingernail.

These advances allow modern radiometers to monitor critical atmospheric water vapor absorption lines (such as 183.31 GHz) and oxygen absorption complexes (such as 50 to 58 GHz and 118.75 GHz) using compact planar antennas and microstrip circuitry. This eliminates kilograms of machined aluminum waveguide hardware.

Technology Domain	Legacy Approach	Miniaturized Approach	Primary Impact on Mission Architecture
Infrared Detection	Mercury Cadmium Telluride (MCT) cooled to ~75 K via closed-cycle Stirling/pulse-tube cryocoolers.	Type-II Superlattice (T2SL) / High Operating Temperature (HOT) operating at 120 K to 160 K.	Elimination of heavy, power-hungry mechanical cryocoolers; enables passive or micro-thermoelectric cooling.
Optical Assemblies	Bulk reflective/refractive multi-element systems with precision-machined mechanical mountings.	Planar optical metasurfaces (sub-wavelength dielectric meta-atoms).	Over 80% reduction in optical volume and mass; simplifies optical bench structural rigidity.
Spectrometry Mechanism	Michelson interferometer with moving scanning mirrors and corner-cube retroreflectors.	Photonic Integrated Circuits (PICs) and static Spatial Heterodyne Spectrometers (SHS).	Removes all moving parts; improves resilience against launch loads and mechanical failure.
Microwave Frontends	Discrete hollow metallic waveguides, separate horn assemblies, and bulky downconverters.	InP and GaAs Monolithic Microwave Integrated Circuits (MMICs) and planar patch arrays.	Integrates receiver frontends onto millimeter-scale chips, permitting deployment on 6U–12U CubeSats.

Key ESA Programs and Demonstrator Missions

To transition these component-level technologies into space-qualified systems, ESA has introduced rapid prototyping initiatives, technology demonstration frameworks, and targeted small-satellite missions.

The Arctic Weather Satellite

The Arctic Weather Satellite (AWS) is ESA's flagship demonstrator for miniaturized microwave sounding. Developed under an ESA contract with an industrial consortium led by OHB Sweden, AWS was conceived to address a severe forecasting gap: the sparse temporal coverage of the Arctic region by traditional geostationary and low-Earth-orbit satellites.

The AWS spacecraft belongs to the microsatellite class, having a total launch mass of approximately 120 kilograms. It carries a highly integrated, cross-track scanning passive microwave radiometer that measures across 19 channels distributed across four main frequency bands:

• 50 to 58 GHz: For atmospheric temperature profiling from the lower troposphere into the stratosphere.
• 89 GHz: For surface characterization, snow cover assessment, and cloud liquid water estimation.
• 166 GHz: For measuring atmospheric moisture and identifying frozen precipitation particles.
• 176 to 183.31 GHz: Centered on the strong water vapor absorption line, providing fine vertical humidity profiles.

The AWS radiometer employs an ultra-compact rotating antenna system that cross-scans the Earth's atmosphere beneath the satellite. Despite its small mass and power budget—which are a fraction of the legacy Microwave Warning Instrument on MetOp—the AWS payload delivers radiometric resolution (Noise-Equivalent Delta Temperature, or NEΔT) comparable to full-scale operational sounders.

Following the successful deployment and evaluation of the AWS prototype (launched in August 2024), the mission serves as the technical baseline for EPS-Sterna. Proposed in collaboration with EUMETSAT, EPS-Sterna is designed as an operational constellation of small satellites deployed across multiple orbital planes. This constellation aims to deliver high-revisit-rate temperature and humidity soundings globally, providing numerical weather prediction models with frequent, hourly atmospheric state updates.

ESA Scout Missions

Within the Earthnet and FutureEO programs, ESA established the "Scout" mission framework. Scout missions explore whether small satellites (typically CubeSats or SmallSats under 100 kg) can gather scientifically useful Earth observation data rapidly and on strict budgets (typically capped at €35 million, including launch, with development cycles under three years).

Miniaturized atmospheric sounding concepts have featured prominently in Scout program development:

• CubeMAP: A proposed Scout mission designed to study the atmospheric interface between the upper troposphere and stratosphere. The concept centers on a constellation of 12U CubeSats carrying miniaturized infrared limb sounders and solar occultation spectrometers. These instruments aim to quantify greenhouse gases, ozone, and volcanic aerosol concentrations with high vertical resolution using High Operating Temperature infrared sensors.
• Hydroterra: A mission concept assessing geostationary radar and microwave sounding principles to understand rapid hydrological cycle transformations and severe convection processes over Europe and Africa.
• Miniaturized GNSS Radio Occultation: Scout and related technology projects have funded high-gain, compact patch-antenna arrays and low-power software-defined radio (SDR) receivers. These packages allow small satellites to calculate atmospheric temperature, pressure, and refractivity profiles by tracking the phase delay of GPS, Galileo, and BeiDou navigation signals as they slice through the Earth's limb.

Airborne Heritage and Instrument Transfers: From GLORIA to GLITE

Before launching novel sounding concepts into space, ESA and its national partner agencies test new instrument architectures on high-altitude airborne research platforms, such as the German DLR HALO (High Altitude and Long Range Research Aircraft) or the French Safire aircraft.

A key baseline for miniaturized limb sounding is the GLORIA (Gimballed Limb Observer for Radiance Imaging of the Atmosphere) instrument. Developed jointly by the Karlsruhe Institute of Technology (KIT) and Forschungszentrum Jülich, GLORIA is an imaging Fourier Transform Spectrometer that pairs a classical Michelson interferometer with a two-dimensional infrared focal plane array. By measuring atmospheric thermal radiation across thousands of spatial and spectral channels simultaneously, GLORIA demonstrated that atmospheric dynamics and trace-gas distributions could be resolved in three dimensions with fine spatial scales.

However, the original GLORIA instrument weighs several hundred kilograms and relies on an extensive gimbal system and cryogenic systems. ESA has supported research to scale this imaging FTS technology down into compact instruments, often referred to under airborne development pathways like GLITE (Gimballed Limb Observer for Radiance Imaging of the Atmosphere - Infrared Trace-gas Experiment concepts) and related compact FTS designs. These efforts evaluate static interferometers, photonic chips, and advanced thermal detectors to capture identical multi-channel limb radiance images within payloads small enough for microsatellites or atmospheric drones.

Limb Sounding Versus Nadir Sounding Miniaturization

Miniaturization strategies differ fundamentally based on whether the instrument observes in a nadir-viewing or limb-viewing geometry.

Nadir sounders look straight down toward the Earth's surface (or scan across the ground track). The measured signal originates throughout the entire atmospheric column, integrated vertically. Nadir sounding requires moderate optical aperture sizes to achieve adequate spatial footprints on the ground (typically 10 to 50 km). The primary miniaturization challenge for nadir sounders lies in maintaining high radiometric sensitivity: because the ground footprint moves quickly beneath the spacecraft, detector integration times are short, requiring low-noise detectors and wide spectral channels.

Limb sounders do not look downward; instead, they point toward the atmospheric horizon, scanning through thin horizontal slices of the atmosphere against the cold background of space. Because the path length of light through the atmosphere along the line of sight is long (hundreds of kilometers), limb sounding provides exceptional sensitivity to trace gases and pollutants present in tiny quantities (parts-per-billion or parts-per-trillion). It also achieves fine vertical resolution (1 to 2 kilometers) in the upper troposphere and stratosphere.

However, miniaturizing a limb sounder is exceptionally difficult due to three physical constraints:

1. Attitude Determination and Control (ADCS): Because the atmosphere thins exponentially with height, a pointing error of just 0.05 degrees translates to an altitude error of several kilometers at the limb horizon. Miniaturized CubeSats have low mass moments of inertia and often experience mechanical disturbances from reaction wheels. Deploying a limb sounder on a small spacecraft requires sub-arcminute star trackers, fine-steering piezo-electric mirrors, and active jitter compensation.
2. Stray Light Rejection and Optical Baffling: The solid surface of the Earth is intensely bright in the infrared compared to the tenuous upper atmosphere directly above it. If unrejected light from the Earth's disk leaks into the instrument aperture, it blinds the limb detector. Limb sounders require large, complex, knife-edged baffling tubes to suppress off-axis light. Shrinking the volume of these baffle assemblies without compromising stray-light rejection requires specialized, ultra-black light-absorbing surface coatings (such as carbon-nanotube arrays).
3. Aperture Limitations: The vertical resolution of a limb sounder is ultimately constrained by diffraction at the collection pupil ($D \propto \lambda / \theta$, where $D$ is aperture diameter, $\lambda$ is wavelength, and $\theta$ is angular resolution). Compressing an instrument into a CubeSat structure forces the use of small entrance apertures, which can cause diffraction blur to spill across vertical atmospheric layers.

Engineering Challenges and Physical Constraints

While laboratory demonstrations confirm the validity of T2SL detectors, metasurfaces, and photonic chips, transitioning these technologies into operational space environments reveals strict physical constraints that cannot be designed away by software or clever mechanics.

Diffraction Limits and Antenna Sizing

The Rayleigh criterion defines the minimum angular resolution $\theta$ achievable by any optical or electromagnetic aperture:

       λ
θ ≈ 1.22 —
       D

where $\lambda$ represents the observing wavelength and $D$ represents the physical diameter of the antenna or primary collection mirror. In the microwave domain, wavelengths are long. For example, a 54 GHz temperature-sounding channel has a wavelength of approximately 5.5 millimeters. To produce an acceptable 30-kilometer ground footprint from a standard 600-kilometer low Earth orbit, the collecting aperture must be at least 15 to 20 centimeters across.

If an engineer attempts to shrink a microwave sounder down to a 3U CubeSat (which measures 10 by 10 centimeters across its face), the maximum physical antenna diameter is constrained to under 10 centimeters. At 54 GHz, this small aperture broadens the angular footprint to over 70 kilometers on the ground. Such coarse spatial resolution limits the instrument's ability to locate localized weather fronts, deep convective cores, or microbursts.

To overcome this diffraction barrier, ESA-backed consortia are developing deployable mesh reflectors, deployable reflectarrays, and synthetic aperture radiometers. Synthetic aperture systems use cross-correlated arrays of tiny antennas spread along deployable booms, synthesizing the resolving power of a large single dish without its mass and launch volume.

Radiometric Noise and the Integration Time Trade-off

The quality of meteorological sounding data depends heavily on its Noise-Equivalent Delta Temperature ($NE\Delta T$), which characterizes the minimum temperature difference the instrument can distinguish from background thermal noise. Radiometer performance is governed by the radiometer equation:

        T_sys + T_ant
NEΔT ≈ ———————————————
          √(B · τ)

where $T_{sys}$ is system noise temperature, $T_{ant}$ is antenna temperature, $B$ is channel bandwidth, and $\tau$ is the integration (dwell) time spent collecting photons over a specific point.

When an instrument is shrunk onto a small satellite, two things happen that compromise this equation:

• Physical antenna aperture size decreases, reducing collected photon flux.
• If the satellite uses cross-track scanning to sweep across the Earth's surface, the integration time $\tau$ per ground footprint is limited to a few milliseconds.

Consequently, the instrument's noise level ($NE\Delta T$) rises. If the data becomes too noisy, data-assimilation systems in operational weather forecasting centers will systematically reject the soundings. Instrument teams must continuously balance how much they can reduce instrument mass and integration time before the data becomes too degraded for meteorological use.

Thermal and Structural Instabilities

Small satellites possess very low thermal mass. In low Earth orbit, a satellite passes from direct, unattenuated solar radiation into the cold shadow of Earth every 90 minutes. This creates violent internal temperature swings, with structural temperature changes sometimes exceeding 40°C over an orbit.

In a large, heavy flagship satellite, massive structural rings of titanium or invar (a specialized iron-nickel alloy with a near-zero coefficient of thermal expansion) hold optical components in rigid alignment. CubeSats and SmallSats cannot afford kilograms of invar. Without strict thermal control, thermal expansion and contraction distort the alignment between the optics, metasurfaces, and detectors. This causes defocusing, optical aberration, and severe radiometric calibration drift.

To solve this without adding heavy insulation, modern miniaturized instruments employ athermalized optomechanical designs. In an athermal setup, structural chassis elements and reflective mirrors are precision-machined from the exact same alloy—typically space-qualified aluminum alloys like Al 6061-T6, or matching silicon carbide (SiC) ceramics. When the instrument heats or cools, all components expand and contract at precisely identical rates, maintaining the focal point on the detector array across the entire orbital thermal cycle.

Comparison of Sounding Architectural Paradigms

The shift from traditional multi-instrument flagship spacecraft toward disaggregated small-satellite constellations represents a complete change in mission trade-offs, operational logistics, and payload engineering.

System Parameter	Legacy Flagship Model (e.g., MetOp-SG)	Miniaturized Constellation Model (e.g., EPS-Sterna / AWS)
Total Spacecraft Mass	3,000 kg to 4,500 kg	30 kg to 150 kg per satellite
Payload Mass Allocation	800 kg to 1,200 kg (multiple combined payloads)	10 kg to 30 kg (single dedicated sounding payload)
Average Electrical Power	2,000 W to 4,000 W	50 W to 150 W
Cooling Methodology	Active mechanical Stirling or pulse-tube cryocoolers (down to 65 K) + large space-facing radiators	Passive cold-space radiative panels or solid-state thermoelectric coolers (120 K to 160 K)
Spectrometer Core	Precision mechanical Michelson interferometer with active laser metrology and dynamic alignment	Static spatial heterodyne circuits, photonic integrated circuits (PICs), or MMIC frontends
Orbital Revisit Frequency	12 to 24 hours per individual satellite platform	30 to 60 minutes via orbital planes with 6 to 18 distributed spacecraft
Development and Build Time	12 to 16 years from initial Phase A to launch	2 to 4 years per spacecraft generation
Redundancy Architecture	Internal cold and hot hardware redundancy (duplicate electronics, dual motors)	Constellation-level systemic redundancy (individual satellite failure tolerated by the network)

Future Outlook and Scientific Impact

The operational deployment of miniaturized sounding technologies will fundamentally reshape numerical weather prediction, disaster monitoring, and atmospheric climate science over the coming decades.

Disaggregated and Hybrid Observing Systems

The future of spaceborne atmospheric sounding will not involve a total abandonment of flagship satellites. Instead, operational meteorology is transitioning to a hybrid observing architecture. In this framework, large flagship satellites (such as MetOp-SG and the United States' JPSS series) serve as high-precision "anchor stations." These flagships carry complex, highly calibrated instruments that supply reference data with absolute radiometric accuracy.

Surrounding these anchor flagships will be constellations of dozens of small, miniaturized satellites (such as EPS-Sterna). The small sounders will cross-calibrate their sensors whenever their orbits intersect the anchor satellites' ground tracks. By using flagship data as an on-orbit calibration reference, the small satellites can drift slightly in calibration without degrading the wider forecasting model. In return, the constellation provides high-frequency observations every 30 minutes, capturing rapidly developing storms, boundary layer transformations, and localized wind and moisture fields.

Edge Computing and On-Board AI Processing

Miniaturized sounders deployed in dense constellations produce massive total volumes of raw data. A constellation of 20 hyperspectral infrared sounders, each generating hundreds of spectral bands every second, easily overwhelms the download capacity of polar ground station networks and small-satellite S-band or X-band downlinks.

To resolve this communication bottleneck, ESA is investing in radiation-tolerant, high-performance edge computing modules and artificial intelligence (AI) accelerators for small spacecraft. On-board edge computing units allow the spacecraft to perform autonomous data processing directly in orbit, including:

• Real-time mathematical Fourier transformation and spectral calibration on field-programmable gate arrays (FPGAs).
• Autonomous cloud clearing: detecting and isolating cloud-contaminated optical pixels and extracting clear-sky radiances before downlinking.
• Data compression: transforming thousands of correlated spectral radiance channels into compact principal component analysis (PCA) scores or empirical orthogonal functions (EOFs). This cuts downlink bandwidth demands by over 90 percent without removing vital meteorological signals.
• Autonomous event triggering: onboard machine-learning models can analyze vertical profiles for abrupt, anomalous changes in moisture or temperature that signal rapid cyclogenesis. The satellite can then autonomously task neighboring satellites in the constellation to focus on the emerging severe weather threat.

Broader Scientific Discoveries

Beyond daily weather forecasts, miniaturized atmospheric sounders will open new frontiers across atmospheric chemistry and planetary research. Affordable limb sounders deployed on CubeSat swarms will allow scientists to trace global transport cycles of greenhouse gases, monitor wildfire smoke plumes penetrating the stratosphere, and track the recovery of the Antarctic ozone hole with fine 3D spatial resolution.

Furthermore, these miniaturized optical, photonic, and microwave technologies are directly transferable to deep-space exploration. The tight mass, power, and volume limits of planetary probes—such as future landers, atmospheric entry probes, and orbiters sent to Mars, Venus, Titan, or the ice giants—prevent the use of multi-hundred-kilogram Earth-observation sounders. By qualifying robust, solid-state, flat-optic spectrometers and single-chip microwave sounders for Earth observation, ESA simultaneously equips future planetary exploration missions with the tools needed to profile alien atmospheres with unprecedented detail.

See also

• Arctic Weather Satellite
• Earth observation satellite
• Fourier Transform Spectrometer
• Infrared Atmospheric Sounding Interferometer (IASI)
• MetOp-SG
• Metasurface
• Monolithic Microwave Integrated Circuit (MMIC)
• Numerical weather prediction
• Photonic Integrated Circuit
• Radiometer
• Type-II Superlattice