What Is a Weather Satellite and How Does It Work?

A weather satellite is a spacecraft equipped with instruments that measure energy radiating from the Earth, its atmosphere, and the Sun, then relay those measurements to ground stations where they feed into weather forecasts, storm warnings, and climate records. These satellites work by detecting electromagnetic radiation across many wavelengths, from visible light you can see with your eyes to infrared and microwave energy you cannot, and converting those signals into images and data profiles that reveal cloud patterns, temperatures, moisture, and wind. The concept sounds simple, but the engineering behind it involves two fundamentally different orbital strategies, dozens of instrument types, and a data pipeline that stretches from orbit to your phone’s weather app.

Two Orbital Strategies for Two Different Jobs

Weather satellites fall into two broad categories based on where they orbit, and each approach has distinct strengths. Geostationary satellites sit roughly 35,786 kilometers above the equator, matching the Earth’s rotation so they hover over the same spot continuously. This fixed vantage point lets them scan their entire field of view repeatedly, producing a new image every few minutes. That rapid refresh rate is critical for watching fast-developing storms, since a thunderstorm can spin up or a hurricane can change structure in the time between scans.

Polar-orbiting satellites, by contrast, circle the Earth at much lower altitudes, typically between 700 and 900 kilometers. They pass roughly from pole to pole, and as the planet rotates beneath them, successive orbits trace out strips that eventually cover the entire globe. Because they fly so much closer to the surface, their instruments achieve finer spatial detail. The trade-off is time: any given location might only be observed a couple of times per day rather than every few minutes. Research comparing the two approaches for wildfire monitoring has confirmed this fundamental tension. Geostationary sensors offer higher observation frequency that gives them significant advantages for tracking rapidly changing events, even though their coarser spatial resolution underestimates some measurements that polar-orbiting counterparts capture more precisely.¹

In practice, the global weather observation network depends on both types working together. The geostationary ring, with satellites spaced around the equator operated by the United States, Europe, Japan, China, India, South Korea, and Russia, provides the continuous watch. Polar orbiters fill in the polar regions that geostationary satellites cannot see well from their equatorial perch, and they deliver the high-resolution vertical profiles of temperature and humidity that numerical weather models rely on most heavily.

How Satellites Actually See the Atmosphere

Your eyes detect a narrow band of electromagnetic radiation called visible light. Weather satellite instruments work on the same principle but extend across a much wider range of wavelengths. The simplest channels image reflected sunlight, producing pictures that look like the cloud-swirled globes you see on the evening news. These visible-light images are sharp and intuitive, but they only work during daytime.

Infrared channels detect thermal energy emitted by the Earth’s surface, clouds, and atmospheric gases. Everything warmer than absolute zero radiates infrared energy, and the amount and wavelength depend on temperature. This is why infrared imagery works around the clock: the satellite does not need sunlight, just the heat signature of whatever it is looking at. By measuring infrared radiation in carefully chosen narrow wavelength bands, instruments can estimate cloud-top temperatures, which in turn reveal how tall a storm is. Colder cloud tops mean taller, more vigorous storms. Researchers have developed methods that use infrared measurements in specific bands near 8.55 and 12.0 micrometers to retrieve the temperatures at the tops of thin, icy cirrus clouds by working through the physics of how radiation passes through those clouds.²

Microwave channels round out the toolkit. Microwaves pass through clouds that block visible and infrared light, which makes them useful for measuring rainfall intensity, sea-surface temperatures beneath overcast skies, and the temperature and moisture structure of the atmosphere at different altitudes. Together, these three wavelength families give forecasters a layered view: visible light shows what the atmosphere looks like from above, infrared reveals thermal structure day and night, and microwaves peer through the clouds themselves.

Specialized Instruments Beyond Standard Imaging

Modern weather satellites carry far more than simple cameras. Some of the most impactful instruments are designed for very specific jobs that a general-purpose imager cannot do well.

Lightning mappers are a good example. The Geostationary Lightning Mapper on the GOES-R series of American weather satellites detects lightning flashes across the entire Western Hemisphere continuously, day and night, with a spatial resolution of about 8 kilometers and a data refresh rate of less than 20 seconds.³ Lightning activity is one of the earliest indicators that a thunderstorm is intensifying, and sudden jumps in lightning rates have been linked to severe weather like tornadoes and large hail. Having that information in near real time gives forecasters a head start on warnings.

Another class of specialized sensor uses radio signals from GPS and other navigation satellite constellations. As a navigation signal passes through the atmosphere on its way to a receiver on a low-orbiting satellite, the atmosphere bends and slows the signal. By measuring that bending precisely, scientists can extract vertical profiles of temperature, pressure, and humidity with high accuracy. This technique, called radio occultation, has become a significant data source for weather models. Companies like Spire have launched large constellations of small CubeSats specifically to collect radio occultation measurements of the atmosphere.⁴ The same signals can also probe the ionosphere, the electrically charged upper atmosphere, where perturbations can disrupt communications and navigation.⁵

Some meteorological satellite platforms also carry radiation monitors that track energetic particles from the Sun and cosmic rays. Europe’s next-generation Meteosat Third Generation and MetOp Second Generation satellites, for instance, include hosted payloads that contribute to operational space weather monitoring.⁶ Solar storms can damage satellites, overload power grids, and endanger astronauts, so piggybacking space weather sensors onto meteorological satellites is a practical way to expand monitoring coverage without launching dedicated missions.

From Raw Signal to Weather Forecast

The measurements a satellite makes in orbit are useless until they reach the ground and get translated into something a forecaster or a computer model can use. That pipeline is more involved than most people realize.

Raw instrument data are transmitted from the satellite to ground receiving stations using high-frequency radio links, typically in the X-band. The ESA’s EarthCARE mission, which carries a cloud radar, a lidar, a multispectral imager, and a broadband radiometer, offers a detailed look at this process. Instrument data arrive as source packets that get sorted into time order, annotated with location and timing information, and assembled into what is called a level 0 product: essentially, the raw numbers. From there, instrument-specific processors calibrate the data, correct for known sensor quirks, and attach geographic coordinates to every measurement, producing level 1 products that represent fully calibrated science measurements on the instrument’s native grid.⁷

Those calibrated measurements then enter the world of numerical weather prediction. Weather models divide the atmosphere into a three-dimensional grid and solve equations that describe how air moves, heats, cools, and exchanges moisture. Satellite data get blended into these models through a process called data assimilation, where the model’s current best guess of atmospheric conditions is adjusted to better match what the satellites actually observed. Getting this right requires fast, accurate models of how radiation travels through the atmosphere, along with sophisticated techniques to correct for any systematic biases in the satellite measurements.⁸ Over the oceans, where ground-based weather stations are sparse or nonexistent, satellite data are often the only observations feeding the model. Without them, forecast accuracy over the open ocean would collapse.

Tracking Tropical Cyclones and Severe Storms

One of the most consequential applications of weather satellite imagery is estimating the intensity of tropical cyclones. Over the open ocean, there are no weather stations to measure a hurricane’s wind speed directly. Aircraft reconnaissance covers some storms, primarily in the Atlantic basin, but vast stretches of the Pacific, Indian Ocean, and Southern Hemisphere rely almost entirely on satellites. The Dvorak technique, developed in the 1970s and used operationally by tropical cyclone warning centers worldwide since 1984, estimates storm intensity by analyzing patterns in enhanced infrared satellite imagery: the shape of the eye, the symmetry of the cloud shield, and how cold the cloud tops are.⁹

As satellite resolution has improved, the relationship between resolution and intensity estimates has gotten more nuanced. Recent work comparing the Dvorak technique applied to older 8-kilometer resolution imagery versus current 2-kilometer imagery found that the higher resolution yields higher intensity estimates for some hurricanes with well-developed eyes, and that difference is statistically significant. Interestingly, for hurricanes below a certain intensity threshold, the finer resolution actually introduced slightly higher error when checked against aircraft measurements.⁹ This is a useful reminder that more pixels do not automatically mean better answers. The techniques built around older instruments sometimes need recalibration when the hardware improves.

Detecting Volcanic Ash and Other Hazards

Weather satellites were designed for meteorology, but their infrared channels turn out to be remarkably useful for spotting hazards that have nothing to do with rain or wind. Volcanic ash is one of the most important. Ash clouds are invisible on standard visible imagery once they drift away from the eruption plume, and they are deadly to jet engines. Meteorological satellites can detect airborne ash using a technique based on the way volcanic particles interact with infrared radiation differently than water or ice clouds do.

An improved method uses three infrared bands centered near 3.9, 10.7, and 12.0 micrometers and combines the brightness temperature differences between them, scaled to maximize the contrast between volcanic ash and ordinary weather clouds. Testing with GOES satellite data showed this three-band approach provides better discrimination of ash from meteorological clouds than older two-band methods, and it has been adopted operationally at the Washington Volcanic Ash Advisory Center.¹⁰ Aviation authorities around the world depend on these satellite-based ash advisories to reroute flights and avoid catastrophic engine failures.

The same infrared instruments are used to detect and monitor wildfires, track dust storms crossing oceans, identify fog banks that pose risks to shipping and coastal airports, and measure sea-surface temperatures that influence hurricane development and fisheries. These are all extensions of the same basic capability: sensing radiation in carefully chosen wavelength bands and interpreting the differences.

Keeping Instruments Honest Through Calibration

A satellite instrument drifting even slightly out of calibration can corrupt years of data and throw off forecasts. Sensors degrade in the harsh environment of space, exposed to radiation, extreme temperature swings, and micrometeorite impacts. Maintaining measurement quality requires constant calibration, both against onboard references and against other satellites.

One particularly effective strategy involves inter-calibration between twin satellites. South Korea’s GK2A geostationary satellite carries an imager that can be cross-checked against a spectrometer on its companion satellite in the same orbital slot. Because the twin satellites observe the same locations at the same times under the same atmospheric conditions, researchers can apply strict collocation criteria that dramatically reduce the uncertainty introduced by mismatches in viewing geometry or scene content. Results from this approach show correlation coefficients above 0.99 between the two instruments, enabling detailed analysis of each sensor’s calibration characteristics and how they vary across space and time.¹¹ This kind of rigor matters not just for day-to-day forecasting but for building long-term climate records, where even tiny systematic errors can masquerade as climate trends.

CubeSats and the Changing Economics of Weather Observation

Traditional weather satellites are expensive, heavy, and take years to build. A flagship geostationary satellite can cost billions of dollars and take a decade from concept to launch. That model works, but it concentrates risk: if a launch fails or an instrument breaks, the gap in coverage can last years.

CubeSats are changing the picture. These standardized small satellites, built in units roughly the size of a loaf of bread, can be manufactured and launched quickly and cheaply. Spire’s LEMUR2 satellites, for example, use a 3U CubeSat form factor and carry radio occultation payloads that can operate at duty cycles as high as 98 percent.¹² By launching many of these small satellites into different orbital planes, a constellation can achieve global coverage with a temporal density that a handful of large satellites cannot match. If one CubeSat fails, the constellation barely notices.

CubeSat data are not a replacement for the big flagships. They lack the large imagers and heavy instruments that produce the detailed cloud imagery and atmospheric profiles that operational forecasting depends on. But they fill gaps, particularly in radio occultation measurements, and they provide a testing ground for new sensor technologies. The data from these commercial constellations are already being assimilated into operational weather models alongside observations from traditional government-operated satellites.

The Economic and Human Stakes

Weather satellites are not just scientific instruments. They are economic infrastructure. Satellite data feed into forecasts that influence decisions worth enormous sums every day: when to plant and harvest crops, how much electricity a utility needs to generate, whether to evacuate a coastal city, how to route container ships across the Pacific. Research examining the electric power industry found that satellite-derived forecasts have saved lives, reduced weather-related economic losses, and improved quality of life broadly, with the data becoming integral to the forecast process in the United States and internationally.¹³

The value is especially stark in severe weather warnings. Before the satellite era, a hurricane could approach a coastline with little advance warning beyond what coastal observers and ships could report. Today, tropical cyclones are tracked from formation to landfall, often over a week in advance, giving communities time to prepare and evacuate. Tornado warning lead times have improved steadily as satellite imagery, lightning data, and satellite-fed model forecasts give meteorologists a more complete picture of storm environments. The economic return on investment in weather satellites is estimated to be many times their cost, though pinning down an exact number is difficult because the benefits are diffuse and spread across nearly every sector of the economy.

Climate monitoring adds another dimension. Satellites have provided continuous measurements of sea-surface temperature, ice extent, atmospheric composition, cloud properties, and radiation balance for decades. Building credible climate data records from these observations requires careful attention to uncertainty, since the instruments, orbits, and processing methods change over time. The European Space Agency’s Climate Change Initiative has worked to generate climate data records that include rigorous uncertainty information to support applications in policy, climate modeling, and weather prediction reanalysis.¹⁴ Without satellite-based climate records, our ability to detect and attribute long-term changes in the Earth system would be far weaker than it is today.