Sovereign weather capability

Most national meteorological agencies ingest satellite data they cannot verify, reprioritise or retain when a foreign operator changes policy. This page sets out the realistic ladder from receiving international data locally to contributing original observations to global models to owning the sensors that produce them.

The dependence this ends: Forecast inputs that arrive through someone else's agency

The dependency most agencies do not notice until it matters

National meteorological agencies have quietly become dependent on a handful of foreign satellite operators for the vertical atmospheric profiles that make modern numerical weather prediction work. EUMETSAT, NOAA and the Chinese FY series provide the bulk of global sounding and radio-occultation data. That data is shared under WMO frameworks, and the sharing has been reliable. It is also discretionary. An operator facing domestic political pressure, export-control review or simple bandwidth constraints can throttle, delay or suspend access. Most agencies have no fallback and no visibility into when the next delivery will arrive.

The post-2022 environment has sharpened this. Governments that once treated foreign satellite services as neutral infrastructure now treat them as strategic dependencies. The lesson from undersea cable sabotage, GNSS jamming over conflict zones and commercial imagery shutter-control decisions is the same in each case: the service works until the day it does not, and the failure is never announced in advance. Weather forecasting is not immune to that dynamic. A 48-hour forecast gap during a tropical cyclone, a flood event or a wildfire season is not an inconvenience. It is a public-safety failure with a political cost attached.

What sovereign weather capability actually means

Sovereignty in this context does not require launching a geostationary meteorological satellite on day one. It means building a progressively independent position across three distinct functions: receiving international data on your own terms, contributing original observations that improve global and regional models, and eventually owning sensors whose tasking you control entirely.

GNSS radio-occultation is the most tractable entry point. A small satellite carrying an RO receiver tracks the bending of GPS, GLONASS or Galileo signals as they pass through the atmosphere at the limb of the Earth. Each occultation event produces a vertical profile of refractivity, from which temperature and humidity can be retrieved to roughly 100-metre vertical resolution through the troposphere. The physics is well understood; the COSMIC-2 constellation operated by NOAA and Taiwan's NSPO has demonstrated that small platforms in low-Earth orbit can produce operationally useful soundings at scale. A national 3-to-6 satellite RO constellation contributes real data to global assimilation systems, which means the country gains standing in WMO data-exchange negotiations and reduces its net dependency on others.

Atmospheric sounders on slightly larger platforms add passive microwave and infrared channels that provide cloud-penetrating temperature and humidity retrievals. These are more expensive and more capable. A direct-receive station, meanwhile, captures full-resolution data from international systems such as Metop, NOAA-20 and Suomi-NPP as they pass overhead, without waiting for the data to be processed and redistributed through foreign ground networks. That alone can reduce latency from hours to minutes.

The ambition ladder: three honest levels

The pathfinder level is a direct-receive station combined with in-country processing. No satellite is required. The agency receives and processes international data domestically, builds the engineering and data-science capacity to ingest it, and establishes the ground infrastructure that a future national satellite will use. This is achievable in 18 to 30 months and represents the lowest-cost entry. It ends the dependency on foreign data-distribution portals and gives the agency control over archiving and reprocessing.

The operational constellation level adds one to six small satellites carrying GNSS-RO payloads, optionally with atmospheric sounder instruments on the larger platforms. Missions of this class, based on publicly reported programmes including COSMIC-2 and the EUMETSAT EPS-SG preparatory work, involve per-satellite costs that vary widely with instrument complexity, but small-satellite RO missions have been developed for budgets in the low tens of millions of dollars per satellite at the lower end of capability. A three-satellite constellation in complementary orbital planes provides meaningful global coverage contribution and regional revisit improvements. Timeline from contract to first operational data: 36 to 54 months, depending on payload maturity.

The full sovereign programme level integrates the constellation with a national forecast model that assimilates domestic observations, a mission-control centre staffed by national operators, and a data policy that allows the country to decide what it shares, with whom and on what terms. This is a multi-year institutional project as much as a technical one. It requires sustained investment in meteorological science alongside the engineering programme. Countries that have reached this level, including India with INSAT and the Kalpana series operated by ISRO, did so over decades. The realistic horizon for a country starting today is ten to fifteen years to genuine operational independence, shorter if the pathfinder and constellation phases are executed without pause.

What you end up owning, and where the limits are

At handover, the customer holds physical title to the ground station hardware, the satellite bus and payload documentation, source code for data-processing pipelines, spectrum filings registered in the national administration's name, and trained national operators who have run the system under supervision for at least one full operational season. Satellize structures contracts so that source-access terms and hardware audit rights are agreed before signature, not negotiated after delivery.

The limits are real and worth stating plainly. A three-to-six satellite RO constellation does not replace a geostationary imager. It contributes soundings; it does not provide visible or infrared imagery of developing convective systems. Cloud cover does not block RO retrievals the way it blocks optical sensors, but the horizontal resolution of an occultation profile is coarse, roughly 200 to 300 kilometres along the ray path, which means mesoscale features can be missed. A direct-receive station captures data only when an international satellite is overhead and transmitting; it does not give the agency any control over what that satellite observes or how it is tasked. And a national forecast model is only as good as the data assimilation expertise behind it. The hardware is the easier part.

How the programme is structured

Satellize delivers the programme under a single contract with a single accountable engineer. Launch is arranged and integrated with launch-vehicle partners; the customer does not manage multiple prime contractors. The ground station, processing infrastructure and satellite operations are designed from the start to be handed over to a national team, with staged transfer of operational authority beginning at first light and completing after a defined period of joint operations.

The Tonga sovereign-comms restoration after the 2022 Hunga Tonga cable break, and the subsequent Tonga crop-estimation analytics programme, illustrate the same principle applied to a smaller island-state context: the capability was built to be operated by Tongans, not managed remotely by Satellize indefinitely. The same logic applies here. A meteorological capability that requires a foreign contractor to keep the lights on is not sovereign. It is outsourced under a different name.

What this mission is built from

GNSS radio-occultation payloads: Primary science instrument producing vertical atmospheric profiles for assimilation into national and global forecast models.
Atmospheric sounders and gas spectrometers: Secondary payload on larger platforms providing passive microwave and infrared temperature and humidity retrievals, including through cloud.
6U CubeSat platforms: Low-cost bus platform for pathfinder RO satellites, enabling early in-orbit demonstration of national observation capability before full constellation commitment.
Direct-receive stations: Ground infrastructure capturing full-resolution data from international meteorological satellites on overhead pass, reducing data latency from hours to minutes.
In-country data processing: Data ingest, quality control and model-assimilation pipelines operated on national infrastructure, keeping raw observations and derived products under domestic control.

What you end up owning

Satellite bus and payload hardware, with full documentation and source-access terms agreed at contract signature
Direct-receive ground station, physically located on national territory and registered to the national administration
Spectrum filings and orbital slots registered in the country's name with the ITU
Source code and operational documentation for data-processing and model-assimilation pipelines
Trained national satellite operators and meteorological data engineers who have run the system through at least one operational season
Archive of all observations collected by national sensors, held on domestic infrastructure under national data policy

Operational authority transfers in stages: the national team shadows Satellize engineers from integration through launch and early operations, takes primary responsibility for routine operations after a defined commissioning period, and assumes full independent control before the contract closes. Satellize retains no ongoing access to mission data or systems after handover unless the customer requests a separate support arrangement. Launch-vehicle and some bus-platform partners have their own post-delivery relationships with the customer, which are documented and disclosed before contract signature.

Programme parameters

Pathfinder phase (direct-receive + processing)	1 ground station, no satellite required, 18 to 30 months to operational data
Constellation phase (RO satellites)	3 to 6 satellites in complementary low-Earth orbit planes, 36 to 54 months from contract to first operational soundings
Orbital altitude (RO constellation)	Typically 500 to 550 km, consistent with COSMIC-2 and similar operational programmes
Vertical resolution (RO retrieval)	Approximately 100 m through the lower troposphere, degrading above the tropopause
Horizontal resolution (RO profile)	200 to 300 km along ray path; not a substitute for high-resolution imagers
Direct-receive latency improvement	International data available within minutes of overhead pass versus hours via redistribution portals
Minimum national operations team	4 to 8 trained satellite operators and data engineers for a small constellation; more for a full programme
Full sovereign programme horizon	10 to 15 years from pathfinder start to genuine operational independence, based on precedents including ISRO INSAT series

Sources

One contract, one accountable engineer

Commissioned as one programme, not a stack of contracts: spacecraft, launch, ground segment, mission control, training and handover are priced together. Source-access terms and audit rights are agreed in writing before signature. Request a programme scoping session.