Satellite Services for Weather Forecasting Market Analysis 2026

Key Takeaways

• The global satellite weather services market exceeded $2.5 billion in 2024 and is growing steadily
• Commercial operators like Spire Global and Tomorrow.io are disrupting traditional government-led forecasting
• Data fusion from LEO constellations is reshaping forecast accuracy at hyperlocal scales

The Business of Knowing What the Sky Will Do

Weather has always been worth money. Farmers, shippers, airlines, utilities, and insurers have paid for forecasts for centuries, and the appetite for that information has never shrunk. What’s changed is the infrastructure behind it. Satellites have become the dominant source of atmospheric data feeding weather models worldwide, and the commercial satellite services sector has grown into a multi-billion-dollar industry layered on top of government-operated space assets.

The relationship between satellites and weather prediction isn’t new. The United States launched TIROS-1 in April 1960, making it the first weather satellite to transmit useful imagery back to Earth. That single step cracked open a market that now includes dozens of government agencies, scores of commercial operators, and a growing ecosystem of data analytics companies translating raw satellite observations into actionable intelligence.

What’s happening in the 2020s is something qualitatively different from past eras. The cost of launching small satellites has dropped dramatically. SpaceX Falcon 9 rideshare missions have enabled constellations that would have been financially impossible a decade ago. Private companies are deploying sensors designed specifically for atmospheric measurement, rather than repurposing imagery satellites. And the buyers of weather data have multiplied well beyond traditional meteorology offices.

Market Size and Growth Trajectory

Measuring the satellite weather services market is genuinely tricky, partly because different analysts draw the boundaries differently. Some figures count only commercial data sales. Others fold in government procurement contracts. A few include downstream analytics products built on satellite data. Depending on who’s counting, estimates for 2024 cluster between $2.3 billion and $3.1 billion for commercially transacted satellite weather data and derived services.

What analysts do agree on is the growth rate. MarketsandMarkets put the compound annual growth rate for the broader weather forecasting services market at roughly 7.5% through 2028, while Grand View Research projected the satellite-specific segment growing faster than that, driven by constellation deployment and expanding enterprise demand. The World Meteorological Organization has separately documented a shift in member-state procurement patterns, with national meteorological services increasingly supplementing government satellite data with commercial purchases.

The split between government and commercial spending matters because it shapes who controls the market. Historically, national meteorological agencies were the primary customers, buying satellite data through intergovernmental arrangements or operating their own satellites entirely. That model still dominates total satellite weather spending by dollar value. But commercial procurement has been growing as a share, and the commercial segment is where most of the product innovation is happening.

Aviation alone represents a significant anchor for commercial satellite weather data. Airlines spend approximately $3 billion annually on weather-related services across all data types, and satellite-derived products account for a growing fraction of that. The Weather Company, now owned by Francisco Partners after IBM announced the sale in August 2023 and completed the transaction in February 2024, counts aviation as one of its largest verticals, with satellite data inputs central to turbulence forecasting, route optimization, and dispatch decision-making.

How Satellites Actually Feed Weather Models

Most public understanding of satellite weather is rooted in visible imagery, the kind shown on evening news broadcasts with swirling cloud masses and storm systems. That’s real and useful, but it represents a fraction of what satellites actually contribute to modern numerical weather prediction.

The data types that matter most to forecast models fall into several categories. Atmospheric soundings, which measure temperature and humidity at different altitudes throughout the atmospheric column, are arguably the most valuable. These come primarily from infrared and microwave instruments aboard geostationary satellites and low Earth orbit polar-orbiting platforms. Wind vectors derived from tracking cloud and water vapor movement contribute to understanding atmospheric dynamics. Surface observations including sea surface temperature, sea ice extent, and land surface conditions feed model boundary conditions.

Then there’s radio occultation, a technique that’s become increasingly important and commercially contested. When a GPS or GNSS signal passes through the atmosphere on its way from a navigation satellite to a receiver aboard a weather satellite, the signal bends. The amount of bending reveals temperature, pressure, and humidity profiles with remarkable precision across a broad vertical range. Spire Global has built a significant commercial business around this technique, operating a constellation of over 100 small satellites in LEO, each equipped with GNSS radio occultation receivers. The company sells this data to national weather services including those of the United States, Germany, and the United Kingdom.

EUMETSAT, the European Organisation for the Exploitation of Meteorological Satellites, manages the Meteosat series in geostationary orbit and coordinates access to data from the Metop polar-orbiting series. Both contribute soundings, imagery, and derived products that feed the European Centre for Medium-Range Weather Forecasts (ECMWF) model, widely regarded as the world’s most accurate global weather prediction system. The ECMWF’s track record on major weather events including Hurricane Dorian in 2019 and Winter Storm Elliott in December 2022 demonstrated the model’s value at lead times that were simply not achievable before modern satellite observing systems existed.

The Major Satellite Operators

Government agencies operate the backbone infrastructure, but the commercial layer is where growth is concentrated. The landscape of operators has changed substantially since 2018 and continues to evolve.

NOAA operates the United States’ primary weather satellite fleet, including the GOES-East and GOES-West geostationary satellites and the JPSS polar-orbiting series. GOES-18, launched in March 2022, carries the Advanced Baseline Imager providing 16-band multispectral imagery that has materially improved severe weather monitoring over the western hemisphere. The Joint Polar Satellite System, particularly NOAA-20 and the more recent Suomi NPP, provides twice-daily global coverage with the ATMS microwave sounder and CrIS infrared spectrometer.

NASA operates research satellites like the Global Precipitation Measurement mission, a joint effort with the Japan Aerospace Exploration Agency (JAXA), that measure global rainfall and snowfall with unprecedented precision. The GPM Core Observatory, launched in February 2014, generates data used both in real-time forecasting and in long-term climate model validation.

On the commercial side, three companies have emerged as the dominant providers of novel satellite-derived weather data: Spire Global, Tomorrow.io, and GeoOptics.

Spire Global went public through a SPAC merger in August 2021. Its business model is built around selling space-based data as a service, with weather being its largest revenue vertical. The company generated approximately $105.7 million in total revenue during fiscal year 2023, with weather data contracts representing a substantial portion. Its customers include the US Air Force, Environment and Climate Change Canada, and multiple European national meteorological services.

Tomorrow.io has taken a different approach, positioning itself not just as a data provider but as a full-stack weather intelligence company. It launched its own commercial weather satellites beginning with the Tomorrow-R1 radar satellite in 2023, the first commercial weather radar in space. The company counts Delta Air Lines, Uber, and the National Football League among its commercial customers. Its platform synthesizes satellite observations with numerical models and AI-based post-processing to generate hyperlocal forecasts at resolutions that traditional models can’t match.

GeoOptics operates a smaller constellation focused specifically on GNSS radio occultation, selling data through the NOAA Commercial Weather Data Pilot program, which began evaluating commercial radio occultation data in 2016. The program eventually validated that commercial data was operationally useful, paving the way for ongoing procurement contracts.

Planet Labs, primarily known for Earth observation imagery, has become relevant to weather-adjacent applications through its daily global imaging cadence. The company’s Dove constellation of over 200 small satellites produces daily revisit imagery used in agricultural monitoring, flood mapping, and wildfire assessment, all of which intersect with operational meteorology and emergency management applications that pull from satellite weather data services.

Maxar Technologies, acquired by Advent International in 2023, provides high-resolution satellite imagery used in post-event assessment and has entered weather-adjacent markets through its analytics products. Its WorldView satellites aren’t weather instruments per se, but the company’s geospatial analysis capabilities are increasingly integrated with meteorological data in enterprise applications.

The Radio Occultation Race

It’s worth examining radio occultation more closely because this relatively obscure measurement technique has become the most commercially contested piece of the satellite weather data market. Government agencies have depended on it for years, primarily through the COSMIC/FORMOSAT-3 constellation operated jointly by NOAA and Taiwan’s National Space Organization, and its successor COSMIC-2, which launched six spacecraft into equatorial orbit in 2019.

The commercial version of this technology arrived through companies willing to add GNSS receivers to small satellites as secondary payloads, dramatically reducing per-profile costs compared to dedicated government missions. Spire has been the most aggressive. Its constellation currently generates over 10,000 atmospheric profiles per day, compared to roughly 2,000-3,000 profiles per day from COSMIC-2. NOAA has independently assessed that assimilating commercial radio occultation data alongside its government sources has reduced 24-hour forecast errors measurably, though the agency has been careful not to overstate the magnitude.

GeoOptics published results in 2022 showing its data quality was statistically indistinguishable from COSMIC-2 profiles after quality control processing, validating the small-satellite approach for this application. The implication for the market is significant: radio occultation data, once the exclusive domain of expensive dedicated government missions, has become a commodity that commercial operators can produce at scale, creating price competition and expanding total data volume.

The European Space Agency’s Aeolus wind lidar satellite, launched in August 2018 and deorbited in July 2023, demonstrated that direct wind measurement from space was technically feasible. Its successor, Aeolus-2, is planned by ESA as part of a longer-term observing strategy. On the commercial side, Orbital Micro Systems and others have been developing passive microwave instruments for small satellites, competing in a different corner of the atmospheric observation market.

Enterprise Demand and the Expansion of Weather Intelligence

The customer base for satellite weather services has expanded well beyond meteorology agencies and airlines. Insurance, agriculture, energy, retail, and logistics have all developed sophisticated appetites for weather intelligence that incorporates satellite data.

The insurance industry’s relationship with satellite weather data runs through catastrophe modeling. RMS (now part of Moody’s ) and AIR Worldwide (now part of Verisk ) incorporate satellite observations into the physical models that estimate potential losses from hurricanes, hail, floods, and wildfires. When Hurricane Ian struck Southwest Florida in September 2022 as a Category 4 storm, satellite data from GOES-16 provided the 30-second scanning cadence needed to track rapid intensification, while post-event satellites enabled rapid damage assessment. The economic loss from Ian ultimately exceeded $110 billion, and the insurance industry’s ability to triage claims and estimate total exposure drew partly on satellite-derived assessments.

Agriculture represents the largest enterprise vertical by contract volume, if not always by individual deal size. Precision agriculture platforms from companies like The Climate Corporation (now Bayer’s Digital Farming division) and Granular integrate satellite soil moisture, evapotranspiration estimates, and vegetation index products alongside weather model outputs to generate field-level management recommendations. The satellite weather inputs in these platforms come from a mix of government sources, commercial data providers, and reanalysis datasets from ECMWF and NASA’s MERRA-2 reanalysis product.

The energy sector’s exposure to weather variability has intensified with the growth of wind and solar power generation. Grid operators managing renewable-heavy portfolios need accurate day-ahead and hour-ahead forecasts to balance supply and demand. Satellite-derived solar irradiance data, particularly from EUMETSAT’s Meteosat Second Generation series, has become a standard input for solar power forecasting across Europe. In the United States, Solargis and Clean Power Research incorporate GOES satellite irradiance products into commercial forecasting tools sold to utilities and independent power producers.

The Competitive Structure of the Market

The satellite weather data market doesn’t fit neatly into a single competitive model. It’s more like several overlapping markets, each with different cost structures, customer types, and competitive dynamics.

At the data layer, government agencies (NOAA, EUMETSAT, JAXA, IMD, CMA ) produce the bulk of global satellite weather observations and distribute most of them freely under open data policies. NOAA’s data-sharing policies mean that most of its satellite products are publicly available at no cost. EUMETSAT operates a tiered access model, with registered users in member states getting full access and others paying licensing fees. This means commercial providers aren’t really selling access to data that governments already provide for free. They’re selling data types governments don’t provide, at resolutions or frequencies governments don’t achieve, or in formats and integrations that require commercial engineering effort.

The analytics layer sits above raw data, and this is where the highest-margin commercial activity is concentrated. Companies like DTN, Accuweather, The Weather Company, and Tomorrow.io transform raw satellite observations into decision-specific intelligence. The margins on these products are better than on raw data sales, but the competitive dynamics are different. Customers at this layer are buying outcomes, not observations, and the barriers to entry include proprietary model architectures, customer relationships, and integration with enterprise software systems.

DTN, which has been owned by TBG AG since 2017, serves agriculture, energy, and transportation with weather-dependent decision tools. Its acquisition of several smaller data companies over the past several years has assembled a capability spanning satellite data access, numerical model post-processing, and enterprise software delivery. The company’s agricultural weather products serve grain traders, elevator operators, and commodity merchants for whom a single bad forecast can translate into millions of dollars of exposure.

A third competitive layer exists around ground-truth integration and AI-based post-processing. Satellite observations have gaps, particularly below cloud cover for visible and near-infrared instruments, and at low altitudes in the boundary layer where microwaves have limited sensitivity. Several companies have built businesses around fusing satellite data with surface observations, radar networks, and numerical model output to fill these gaps. StormGeo, MeteoGroup (now part of DTN), and Weathernews all operate in this space.

Government Procurement and Commercial Data Programs

The United States government’s commercial weather data procurement programs have been pivotal in establishing the market’s viability. The NOAA Commercial Weather Data Pilot, launched under the Commercial Space Launch Competitiveness Act of 2015, created a mechanism for NOAA to purchase and evaluate commercial satellite weather data. This wasn’t purely altruistic; NOAA recognized that commercial radio occultation data could supplement its government sources at lower cost than operating additional government satellites.

The program worked. By 2019, NOAA had awarded contracts to Spire Global and GeoOptics for radio occultation data, and subsequent evaluations confirmed their operational value. Congress subsequently appropriated funding for ongoing commercial data procurement, and NOAA has continued expanding its commercial data portfolio. The FY2024 appropriations for NOAA included approximately $10 million designated specifically for commercial data acquisition, a figure some atmospheric scientists have argued should be substantially higher.

The European approach has differed. EUMETSAT and ESA have relied more heavily on government-to-government agreements and on the Copernicus programme, the EU’s Earth observation initiative that operates the Sentinel satellite series. Copernicus Sentinel-6 Michael Freilich, launched in November 2020, measures sea surface height with applications in ocean state forecasting and storm surge modeling. The program’s open data policy has made Copernicus one of the world’s most-used sources of Earth observation data, but it has also shaped a European market where commercial providers face more competition from free government data than their American counterparts do.

Japan’s JMA operates the Himawari-8 and Himawari-9 geostationary satellites, covering the western Pacific with Advanced Himawari Imager instruments that are technically equivalent to GOES-R series capabilities. JMA’s data sharing policies have made Himawari data available to national meteorological services across Asia and the Pacific, and commercial providers selling into Asian markets work in an environment where much foundational satellite data is freely accessible.

China’s CMA operates the FY (Fengyun) satellite series, which has expanded significantly in capability since the 2010s. Fengyun-4A, launched in December 2016, and Fengyun-4B in 2021, provide geostationary coverage over the Asia-Pacific with capabilities roughly comparable to GOES-16. China has simultaneously been developing its own commercial space sector, with companies like ADA Space entering the small satellite market. Whether Chinese commercial weather satellite operators will penetrate international markets remains an open question with geopolitical dimensions that extend well beyond market economics.

Small Satellites and Constellation Economics

The economics of satellite constellation deployment have changed enough in the past decade that statements made in 2015 about per-satellite costs are largely obsolete. SpaceX’s Transporter rideshare program, which began regular flights in 2021, enables operators to place satellites in sun-synchronous orbit for roughly $6,000 per kilogram. A 10-kilogram small satellite carrying GNSS radio occultation hardware can reach orbit for under $100,000 in launch costs, compared to hundreds of millions for a dedicated government weather satellite.

This doesn’t mean small constellations are cheap to operate at scale. Software development, ground station networks, data processing pipelines, and customer integration all add cost that doesn’t scale down linearly with satellite size. But the entry barriers for new operators have dropped enough that the competitive set is meaningfully larger than it was five years ago.

Spire’s experience illustrates the unit economics. The company has stated that each of its LEMUR-2 satellites costs approximately $500,000 to manufacture and deploy, including launch. With over 100 satellites, that’s roughly $50 million in constellation capital investment, supplemented by ongoing replenishment launches. Against that, Spire generated $105.7 million in total revenue for 2023, suggesting the model can work financially, though the company has not yet achieved sustained profitability.

Tomorrow.io’s investment in actual radar hardware aboard its commercial satellites represents a more capital-intensive bet. Weather radar from space has been a longstanding gap in the observation system because traditional radar doesn’t work in geostationary orbit and requires substantial power in LEO. The company’s Tomorrow-R1 satellite uses a synthetic aperture radar approach to overcome some of these limitations, and if the approach proves operationally useful, it could represent a meaningful advance in precipitation monitoring over data-sparse oceanic and polar regions.

Artificial Intelligence and Satellite Data Integration

The past three years have seen a convergence between large-scale machine learning and satellite weather data that is reshaping what forecast products can look like. Traditional numerical weather prediction relies on physics-based differential equations describing atmospheric dynamics, running on massive supercomputers. These models are powerful but computationally expensive and have certain systematic biases that satellite observations partially correct through data assimilation.

Machine learning approaches trained directly on large satellite and reanalysis datasets have demonstrated surprising skill at forecast lead times where physics-based models have historically excelled. Google DeepMind’s GraphCast model, published in November 2023, produced 10-day global forecasts with competitive skill against the ECMWF operational model at a fraction of the computational cost. Huawei’s Pangu-Weather model, published in July 2023, showed similar results. Both models were trained on ERA5, a reanalysis dataset produced by ECMWF that synthesizes decades of historical satellite observations, radiosondes, and surface data.

The satellite weather data market implications of this AI shift are genuinely uncertain, and this is a case where confident predictions seem unwarranted. If AI models can produce competitive forecasts from compact, coarse-resolution satellite observations, the value of comprehensive high-resolution satellite data for real-time forecasting could change in ways that favor different data types or different commercial operators. Alternatively, if the best AI forecast systems turn out to require very high-quality observation inputs to remain accurate under unusual atmospheric conditions, the premium on comprehensive satellite coverage could increase. The answer probably depends on specifics that won’t become clear until these systems have been running operationally through multiple major weather events.

What’s already clear is that AI has changed the analytics layer of the market. Companies that process and sell satellite-derived weather products have integrated machine learning throughout their pipelines, from cloud screening and quality control to bias correction and spatial downscaling. The Weather Company deployed its Deep Thunder high-resolution modeling system, which uses machine learning post-processing extensively, across several markets before spinning out from IBM. The integration of satellite observation data with AI-driven post-processing is now standard practice in commercial weather analytics.

Regional Market Dynamics

The satellite weather services market isn’t globally uniform. Regional differences in government data policy, satellite coverage quality, and enterprise customer concentration create distinct competitive environments across major geographies.

North America represents the most commercially mature market. NOAA’s procurement programs have established a validated commercial data supply chain, enterprise customers in agriculture, energy, and aviation are sophisticated buyers, and the technology ecosystem around weather analytics is dense. The United States accounts for roughly 40% of global commercial satellite weather data spending by most estimates, though that concentration has been slowly declining as other markets develop.

Europe is the second-largest market, shaped by EUMETSAT’s extensive government data provision and Copernicus’s open data philosophy. Commercial differentiation in Europe tends to focus on specialized applications, sector-specific analytics, and integration with enterprise systems rather than on raw data sales. MeteoGroup, founded in 1997 and subsequently acquired by TBG AG in 2018 and integrated into DTN, built a successful European commercial weather business by focusing on road weather, renewable energy forecasting, and maritime applications.

Asia-Pacific is growing fastest, driven by Japan, Australia, South Korea, and India. The Indian Space Research Organisation (ISRO) has been expanding India’s INSAT geostationary weather satellite capabilities, and the IMD’s reliance on satellite data has grown substantially with successive monsoon forecast improvements. South Korea’s KMA operates the GK2A geostationary satellite launched in December 2018, which covers the Asia-Pacific with Advanced Meteorological Imager capabilities.

Australia’s Bureau of Meteorology depends heavily on satellite observations given the country’s sparse surface observing network relative to its land area. Commercial weather services for agriculture, mining, and offshore energy have developed around this dependence on satellite data, making Australia a disproportionately active market for commercial satellite weather applications relative to its population.

Market Segmentation by Application

Breaking the market down by end-use application reveals where the growth is concentrated and where pricing power sits.

Aviation has historically been the largest single commercial segment, partly because weather-related flight delays and cancellations cost the US airline industry alone approximately $8 billion annually according to FAA estimates. The value proposition for better weather intelligence is easy to quantify, and airlines have procurement processes calibrated to spend money on certified, quality-assured data products. This pushes commercial providers serving aviation toward investment in data validation and regulatory compliance that raises barriers to entry but also creates durable customer relationships.

The energy transition is reshaping the energy segment rapidly. As recently as 2015, weather services for the energy sector were dominated by demand forecasting for heating and cooling loads. Today, solar and wind power forecasting has become equally important and is growing faster. Vaisala, the Finnish meteorological instruments and data company, has built a significant renewables forecasting business using satellite irradiance and wind data, with operations across Europe, the Americas, and Asia-Pacific. The company’s solar resource assessment products use satellite-derived irradiance data from multiple sources to estimate long-term energy yield for project financing purposes.

Data Quality, Validation, and the Trust Problem

Weather forecasting has always been a high-trust activity. Decisions made on the basis of a forecast can be irreversible, whether that’s evacuating a coastal city ahead of a hurricane or dispatching crews to pre-position repair equipment before an ice storm. Errors in satellite-derived data that feed into forecasts can cascade into forecast errors that affect real decisions.

This creates a quality problem for commercial satellite data providers that doesn’t exist in the same way for companies selling, say, retail analytics or supply chain data. Errors in weather data can contribute to decisions that cost lives. National meteorological services are acutely aware of this, which is why NOAA’s commercial data programs have included rigorous quality assessment phases before incorporating commercial data into operational models.

The validation methodology for Spire’s radio occultation data, conducted by NOAA’s Environmental Modeling Center, compared commercial profiles against radiosonde balloon observations and found the data quality statistically comparable to COSMIC-2. This kind of independent validation has been essential for commercial providers to win contracts with operational agencies. Customers who are making life-safety decisions need external validation, not just the vendor’s own quality metrics.

The trust problem extends to forecast products themselves. When Tomorrow.io or AccuWeather sells a forecast to an airline or logistics company, the customer is relying on a chain of satellite observations, model runs, and post-processing steps whose internal details they can’t fully audit. Unlike pharmaceutical products, there’s no regulatory approval process for commercial weather forecasts in most countries. The liability structure in commercial weather service contracts typically limits provider exposure to service fees paid, not consequential damages, which means the financial incentive to invest in quality assurance beyond what customers can observe may be suboptimal.

Investment Flows and M&A Activity

The period from 2019 through 2023 saw substantial venture capital and growth equity flow into satellite weather services startups, partly driven by enthusiasm about commercial space broadly and partly by genuine business development. Spire Global raised over $200 million in total funding before its 2021 SPAC listing. Tomorrow.io raised $77 million in a Series D round in 2021, led by Stonecourt Capital.

The SPAC wave that carried several space companies to public markets in 2020 and 2021 produced mixed results. Spire’s stock declined significantly from its SPAC listing price as the company worked to demonstrate a path to profitability that the market initially found unconvincing. The experience was not unique to Spire; several commercial space companies that listed via SPACs in 2020-2021 subsequently traded well below their listing prices.

Post-SPAC, the investment environment has been more selective. Strategic acquisitions have become more common than new VC rounds. TBG AG’s acquisition of MeteoGroup in 2018, which was subsequently integrated into DTN, represents one model of market consolidation through corporate combination. Accuweather’s partnership expansions and licensing agreements with satellite data providers represent a different model through contracts rather than corporate combinations.

One clear trend in M&A is large data and analytics conglomerates absorbing weather capabilities. Verisk’s ownership of AIR Worldwide positions it to integrate satellite weather data with property exposure databases and claims data for insurance analytics. Moody’s acquisition of RMS in 2021 for approximately $2 billion reflected the same logic: catastrophe risk assessment requires weather intelligence, and the combination of financial analytics infrastructure with meteorological modeling capability has strategic value.

Gaps and Limitations in Current Satellite Weather Systems

Despite the progress, real gaps remain in the global satellite observing system, and those gaps have commercial implications.

The Southern Hemisphere is systematically under-observed relative to the Northern Hemisphere. Fewer surface stations, fewer radiosondes, and relatively sparse commercial satellite data procurement in South America, sub-Saharan Africa, and the Southern Ocean mean forecast models initialize less accurately over these regions. Improvements in Southern Ocean coverage have been identified by ECMWF and NOAA as having disproportionate value for improving global model accuracy, because errors in the Southern Ocean propagate eastward and can degrade Northern Hemisphere forecasts at lead times beyond 7-10 days.

Boundary layer observations remain problematic from space. The lowest 2 kilometers of the atmosphere, where most human activity takes place and where surface weather phenomena originate, are poorly sampled by existing satellite instruments. Microwave sounders have limited vertical resolution at low altitudes. GNSS radio occultation signals below about 1-2 km altitude are affected by superrefraction and ducting that complicate interpretation. This gap matters for local severe weather forecasting, where convective initiation, fog formation, and low-level wind shear have enormous practical importance.

The tropics are another challenging region. Global weather models initialize less accurately in the tropics than in the extratropics, partly because the dynamical relationships that make temperature and pressure observations so useful for constraining the flow field are weaker in the tropics. Tropical weather systems, including the mesoscale convective systems that drive much of the world’s rainfall, are not yet well-predicted at extended range despite satellites providing dense tropical coverage.

Fast-moving data gaps are a third limitation. Satellites in low Earth orbit pass over any given location only twice per day in typical polar-orbiting configurations, or at 10-minute intervals for full-disk imagery from geostationary. For some applications, this is adequate. For nowcasting severe weather at the scale of individual thunderstorm cells, it’s not always sufficient. Ground-based radar fills this gap over land in developed countries, but oceanic coverage and coverage in countries without dense radar networks remains inadequate.

Policy, Regulation, and Spectrum

The satellite weather services market operates within a framework of international coordination that most commercial participants take for granted but that shapes competitive dynamics in important ways.

The World Meteorological Organization’s Space Programme coordinates the global satellite observing system through the Coordination Group for Meteorological Satellites (CGMS), which includes both government agency operators and, increasingly, commercial constellation operators. Commercial companies that want their data recognized as meeting operational quality standards and potentially used by national weather services need to engage with this technical framework, which was designed around government-to-government cooperation but has been gradually adapting to commercial participation.

Radio frequency spectrum is a physical constraint on satellite weather observation that has no commercial solution. Many weather satellite instruments depend on passive microwave frequency bands that are protected from interference under ITU Radio Regulations. The growth of commercial satellite communications, including the proliferation of broadband LEO constellations from SpaceX (Starlink ), Amazon’s Project Kuiper, and others, has intensified spectrum management challenges. SpaceX’s Starlink satellites numbered over 6,000 in orbit by early 2025, creating a changed electromagnetic environment that requires ongoing management to protect weather sensor performance.

The NOAA Commercial Data Program operates under a legal framework established by the Commercial Space Launch Competitiveness Act and subsequent agency policy, but that framework doesn’t fully resolve questions about data exclusivity, government use rights, and pricing that continue to complicate contract negotiations. When a commercial company sells weather data to NOAA, the government typically receives unlimited rights to use the data internally, which means the commercial provider cannot also sell the same data exclusively to other government customers without restriction. Navigating these arrangements requires commercial legal expertise that adds cost and complexity to commercial data businesses.

The 2025-2030 Market Outlook

Several technology and policy developments in progress will shape the market over the next five years in ways that seem reasonably predictable, alongside others that don’t.

NOAA’s next-generation geostationary program, GeoXO, is in early development phases with initial operations targeted for the mid-2030s. GeoXO satellites will carry a much larger complement of instruments than current GOES-R series, including hyperspectral sounders, ozone and trace gas sensors, and lightning mappers with expanded coverage. This doesn’t directly affect the commercial market in the near term, but it does signal NOAA’s long-term investment in government satellite infrastructure.

On the commercial side, the most significant near-term developments involve the expansion of existing constellations rather than dramatic new entries. Spire has consistently stated intentions to grow its constellation beyond 150 satellites, which would further increase radio occultation data volume. If the ECMWF’s own assessments that radio occultation is among the highest-value data sources for global NWP are correct, increased volume should translate into continued demand growth.

Machine learning forecast systems are moving from research demonstrations to operational deployment, and this is where the market trajectory becomes genuinely uncertain. If models like GraphCast or Pangu-Weather prove durable and accurate across diverse weather regimes, the marginal value of incremental satellite observations could change in ways that favor different data types or acquisition frequencies. The commercial satellite weather data market has been built partly on the premise that more data is better; if AI models reach diminishing returns on data volume faster than expected, the economics of constellation expansion change.

The insurance and reinsurance segment looks like the most reliable growth driver through 2030. Climate change is increasing the frequency and intensity of weather extremes that drive insured losses, and the insurance industry’s need for better spatial and temporal resolution in weather event assessment is growing. Satellite hail detection products, flood inundation mapping from SAR satellites, and post-wildfire burn scar mapping from optical satellites are all commercially active niches with expanding demand. CoreLogic, Verisk, and Moody’s RMS all have active satellite data integration programs aimed at improving loss estimation in these categories.

Private Equity and the Maturation of the Sector

Commercial satellite weather is transitioning from a startup-dominated sector to one where private equity and large corporate strategics are the primary capital allocators. This is a sign of market maturation, but it also changes incentive structures in ways that may not uniformly benefit data quality or innovation.

When PE firms like TBG AG own DTN or Advent International holds Maxar, the optimization logic shifts toward cash flow generation and portfolio positioning rather than technology development. This isn’t inherently negative; operational discipline and customer focus are valuable. But it can reduce the appetite for long-cycle technology investments like developing novel satellite sensor types or funding the ground infrastructure needed to support next-generation data products.

The counter-argument is that government R&D programs and university research will continue funding speculative sensor development, with commercial operators then deploying proven technologies. This has been the pattern for GNSS radio occultation, where government research via COSMIC established the technique’s value before commercial operators built operational constellations. It may be the pattern for other techniques currently in development, including infrared laser occultation and space-based Doppler radar.

Summary

The satellite services market for weather forecasting has matured from a government-dominated utility into a commercially layered ecosystem with meaningful private sector participation at the data, analytics, and decision-intelligence levels. The commercial layer is being shaped by falling launch costs, expanding enterprise demand, and a technology inflection around machine learning that introduces real uncertainty about which data types will carry the most value in the next decade.

The clearest opportunity in the near term sits where it has sat for the past several years: in the intersection of high-value decision contexts, such as insurance catastrophe assessment, renewable energy dispatch, and precision agriculture, with data products that government satellite systems don’t provide at adequate resolution or frequency. Commercial providers who can demonstrate measurable forecast skill improvements or cost savings in these contexts will continue to win contracts regardless of what happens with AI-based global forecast models.

What’s underappreciated in most market analyses is the degree to which the global satellite weather observation system’s continued improvement depends on the policy choices of a handful of governments. NOAA’s commercial data procurement program is good policy, but its funding has been modest relative to its potential value. EUMETSAT’s open data policies have made Copernicus one of the most impactful public investments in Earth observation anywhere, but the program’s budget is contested in EU negotiating processes that have little to do with meteorological science. The commercial market exists within and alongside this public infrastructure, and its long-term trajectory is shaped as much by government budget decisions as by SpaceX launch prices or AI model accuracy benchmarks.

Appendix: Top 10 Questions Answered in This Article

What is the current size of the satellite weather services market?

The global satellite weather services market exceeded $2.5 billion in estimated commercial activity in 2024, with analyst projections clustering between $2.3 billion and $3.1 billion depending on how commercial transactions are defined. Growth rates for the segment are estimated at roughly 7.5% compound annually through 2028, driven by expanded enterprise demand and commercial constellation deployment.

Which companies are the leading commercial providers of satellite weather data?

Spire Global, Tomorrow.io, and GeoOptics are the most prominent commercial satellite weather data providers as of 2025. Spire operates a constellation of over 100 small satellites generating GNSS radio occultation profiles, while Tomorrow.io has launched commercial weather radar satellites and sells full-stack weather intelligence products. Established analytics players including DTN, The Weather Company, and AccuWeather distribute satellite-derived products to enterprise customers across multiple sectors.

What is radio occultation and why is it commercially important?

Radio occultation is a measurement technique in which signals from GPS and GNSS navigation satellites are detected as they pass through the atmosphere en route to a receiver aboard an orbiting satellite, with the signal bending revealing temperature, pressure, and humidity profiles. The technique is commercially important because it delivers atmospheric soundings comparable in quality to expensive dedicated government missions, but at a fraction of the cost when deployed on small satellites. NOAA’s independent validation confirmed commercial radio occultation data is operationally useful for improving numerical weather prediction.

How do national meteorological agencies buy commercial satellite weather data?

In the United States, NOAA’s Commercial Weather Data Pilot program, established under the Commercial Space Launch Competitiveness Act of 2015, created a procurement mechanism for evaluating and buying commercial satellite data. The program awarded contracts to Spire Global and GeoOptics after validation studies confirmed data quality. European agencies primarily access satellite data through EUMETSAT’s coordinated programs, while also purchasing some commercial products for specialized applications not covered by government satellites.

What role does machine learning play in satellite weather forecasting?

Machine learning has been integrated throughout satellite weather data processing pipelines, from quality control and cloud screening to bias correction and spatial downscaling of forecast products. Google DeepMind’s GraphCast model and Huawei’s Pangu-Weather system, both published in 2023, demonstrated that AI systems trained on historical satellite and reanalysis data can produce global weather forecasts competitive with physics-based numerical models at significantly lower computational cost. The long-term impact on commercial satellite data demand remains genuinely uncertain.

Which industries are the largest buyers of satellite weather services?

Aviation accounts for approximately 22% of commercial satellite weather service spending, making it the largest enterprise segment. Agriculture represents roughly 19%, driven by precision farming platforms that use satellite soil moisture, precipitation, and evapotranspiration products. The energy sector, particularly renewable energy forecasting for solar and wind power, accounts for around 18%. Insurance and reinsurance represents 14%, a share growing as weather-related insured losses increase with climate change.

How have falling launch costs changed the satellite weather data market?

SpaceX’s Transporter rideshare program reduced LEO launch costs to approximately $6,000 per kilogram, enabling small satellite operators to deploy weather observation instruments at costs that were unviable a decade ago. A 10-kilogram satellite with GNSS radio occultation hardware can reach orbit for under $100,000 in launch costs, compared to hundreds of millions for a traditional dedicated government weather satellite. This reduction enabled companies like Spire Global and GeoOptics to build commercial weather constellations on venture capital budgets.

What are the major gaps in current satellite weather observing capabilities?

The Southern Hemisphere is systematically under-observed, with fewer surface stations, radiosondes, and commercial satellite data inputs feeding global models over the Southern Ocean. The atmospheric boundary layer below roughly 2 kilometers remains poorly sampled by satellite instruments, creating gaps in local severe weather forecasting. Tropical weather systems including mesoscale convective complexes are still inadequately predicted at extended range despite good satellite coverage, partly because mid-latitude dynamical constraints don’t apply.

What is the GeoXO program and how will it affect the market?

GeoXO is NOAA’s planned next-generation geostationary satellite program, intended to succeed the current GOES-R series in the mid-2030s. The program includes hyperspectral sounders, trace gas sensors, and expanded lightning mapping instruments that will substantially increase atmospheric profiling capability from geostationary orbit. GeoXO is not expected to directly affect commercial market dynamics in the near term, but its development signals NOAA’s continued investment in government satellite infrastructure alongside commercial data procurement.

How does spectrum protection affect the satellite weather services market?

Passive microwave frequency bands used by weather satellite instruments are protected from radio interference under ITU Radio Regulations, but the rapid growth of commercial broadband LEO constellations including SpaceX Starlink, which exceeded 6,000 satellites by early 2025, has intensified spectrum management demands. Interference with passive weather microwave bands could degrade the quality of satellite soundings used in numerical weather prediction. This creates a structural tension between the commercial satellite communications sector and weather satellite operators that requires ongoing technical and regulatory management.