Urban traffic management agencies are largely blind between fixed sensor loops and camera feeds. Those ground assets are expensive to install, easy to vandalise, and cover only the roads someone thought to instrument years ago. Satellite imagery — captured multiple times per day across an entire metropolitan area — fills that gap with a uniform, tamper-proof view of every arterial, interchange and car park simultaneously.
A constellation of sub-metre optical microsatellites, supplemented by X-band SAR for night and cloud-affected passes, delivers vehicle-level counts from which flow, density and average speed can be derived using computer-vision and kinematic models. At 30 cm resolution an individual vehicle is detectable; at 50 cm, vehicle class (truck, bus, private car) is recoverable. Pairs of passes separated by minutes give direct speed estimates without any roadside hardware.
The operational payoff is genuine: a city that owns this pipeline can feed congestion indices into adaptive signal control, reroute freight in near-real-time, validate the impact of new bus lanes within weeks of opening, and price road access dynamically — all without licensing data from a foreign commercial vendor who can change terms, drop coverage, or simply go offline. Sovereign control over the imagery archive also means longitudinal analysis — decade-scale traffic trend modelling — stays inside national jurisdiction.
Frequently asked
Can satellite imagery actually count moving vehicles, or only parked ones?
Modern high-resolution optical sensors (0.5–0.7 m GSD) can detect and classify vehicles in motion, though very high-speed vehicles may show motion blur in push-broom imagery. SAR-based techniques such as along-track interferometry are specifically designed to estimate vehicle velocity. In practice, most operational systems combine multi-pass optical imagery with AI to infer flow rates rather than count individual moving cars in a single frame.
How does this differ from using Google Maps or Waze traffic data?
Navigation-app traffic data is derived from mobile-phone GPS signals and is only available where users have opted into tracking — coverage is therefore proportional to smartphone penetration and skewed toward higher-income commuters. Satellite-based estimation covers all vehicle classes on all road types regardless of who is carrying a phone, and the raw data is held by the nation rather than a US technology company. For sovereign transport planning, that independence matters enormously.
What orbit and satellite type should a national programme consider?
LEO constellations at 450–550 km altitude using microsatellites (50–150 kg) with electro-optical payloads in the 0.5–1 m resolution class are the standard architecture. A minimum viable constellation for a mid-size nation's capital is typically 4–6 satellites to achieve 90-minute revisit. Adding one or two SAR microsatellites (cf. ICEYE or a domestically built equivalent) extends all-weather capability without requiring a full GEO programme.
Is the data good enough for traffic signal optimisation in real time?
Not on its own. Satellite revisit intervals are too long for closed-loop signal control, which requires sub-second sensor feedback. The appropriate use of satellite data in traffic management is strategic: identifying chronic congestion nodes, validating transport model assumptions, planning new road or transit infrastructure, and monitoring the effect of policy interventions over days and weeks. Couple it with ground sensors and 5G edge compute for real-time signal timing.
What does a sovereign programme cost compared to buying a commercial data subscription?
A four-satellite LEO microsatellite constellation with a ground segment and five-year operations budget typically runs $80–150 million for a mid-income nation — comparable to a five-year enterprise contract with a major commercial EO provider but with full data ownership, no export-control risk, and a domestic industrial base as a co-benefit. The World Bank and ESA both offer co-funding mechanisms that can reduce first-unit cost by 20–40%.
How does weather affect the programme, and how do you manage it?
Cloud cover is the primary operational risk for optical payloads. Cities in temperate or arid climates typically see 60–80% clear-sky availability; tropical cities may drop to 30–40%. The standard mitigation is a hybrid optical-plus-SAR constellation — SAR penetrates cloud and operates at night — and algorithmic fusion with complementary data sources such as Spire's GNSS-RO atmospheric profiles to forecast imaging windows.
Which international bodies set the rules for operating these satellites?
Spectrum coordination is governed by the ITU Radio Regulations and ITU-R Recommendations (particularly the S-series for FSS/NGSO). Orbital debris mitigation must comply with UN COPUOS guidelines adopted in 2018 and, for many markets, national licensing conditions aligned with those guidelines. Earth observation data formats and metadata are standardised by ISO/TC 211 and the Open Geospatial Consortium, ensuring that satellite-derived traffic datasets can be exchanged with municipal GIS platforms.
Can smaller or lower-income cities justify the investment?
A single national programme can serve dozens of cities simultaneously from the same constellation — the marginal cost of adding a city to the analysis pipeline is essentially software and analyst time, not additional hardware. Regional consortia (analogous to EUMETSAT's model for meteorology) can further pool launch and operations costs across neighbouring states, bringing per-city economics within reach of smaller transport budgets.