Wide-area surveillance finds the needle; persistent target watch keeps eyes on it. Once a site, vessel, convoy or installation is flagged as a priority, the intelligence problem shifts from discovery to continuity — any gap in coverage is an opportunity for the adversary to move, disperse, or reconstitute. A sovereign constellation dedicated to this mission ensures that the decision cycle never pauses because a commercial vendor has rescheduled a tasking or invoked a denial-of-service clause under third-party government pressure.
The satellite stack for persistent watch is a layered mix: SAR provides the all-weather, day-night backbone with sub-metre resolution, while optical imagers deliver the colour context needed for human analysts and automated change-detection models. RF sensors close the loop by detecting emissions from the target — radar lock-ons, communication bursts, engine harmonics — that confirm activity status even when no physical change is visible. A twelve-to-twenty satellite constellation at 500–550 km, tuned to a walker geometry over the target's latitude band, can achieve sub-ninety-minute revisit and, with on-board tasking updates, can tighten that to sub-thirty minutes on priority orbits.
The operational payoff is decision dominance. A commander who knows that a ballistic missile transporter-erector-launcher has not moved in six hours, or that a specific port gate opened at 0340 local time, can act or withhold action with confidence. That confidence evaporates the moment the watch depends on a foreign commercial operator who may throttle access during a crisis, lag delivery by hours, or simply lack coverage over the precise geographic cell that matters.
Frequently asked
How many satellites does a nation actually need to achieve persistent watch over a single high-priority site?
At LEO (~500 km), achieving continuous or near-continuous optical coverage of a single point typically requires 40–60 satellites in a purpose-designed Walker constellation, depending on minimum elevation angle and sensor field of regard. SAR or RF sensors with wider swaths can reduce that number but not eliminate it. Most mid-tier nations begin with a 6–12 satellite constellation that provides 3–5 revisits per day — operationally useful but not truly persistent.
Can't we just buy imagery from Planet, ICEYE or BlackSky instead of building our own constellation?
Commercial services offer fast time-to-capability but carry three structural risks: data can be withheld or prioritised away from your tasking during allied or commercial conflicts of interest; ground resolution and collection-scheduling are optimised for the vendor's global customer base, not your specific order of battle; and the intelligence derived never truly belongs to you — metadata, collection geometry and baseline imagery libraries remain on vendor infrastructure. For persistent watch of nationally sensitive sites, sovereign ownership eliminates all three risks.
What orbit should a persistent-watch constellation use?
LEO (400–600 km) is the default for sub-metre optical and SAR payloads because it maximises ground resolution and minimises atmospheric path length. Sun-synchronous orbits (~97.6° inclination) standardise illumination geometry, which is critical for automated change detection. Very Low Earth Orbit (VLEO, below 450 km) is an emerging option that improves resolution and reduces signal-to-noise for SAR but sharply increases atmospheric drag and thus propulsion and replacement costs.
How is 'persistent watch' different from 'wide-area surveillance'?
Wide-area surveillance sweeps broad geographies at moderate resolution to detect activity of interest — effectively a search function. Persistent watch is a track-and-stare function: once a site or object of interest is identified, it is cuued for repeated high-resolution revisits at the shortest possible interval. The two missions are complementary; wide-area surveillance typically generates the target list that feeds persistent watch tasking.
How do sovereign systems handle the legal constraints on space-based imaging — especially the Outer Space Treaty?
The 1967 Outer Space Treaty (Article IV) does not prohibit reconnaissance satellites — a position established by state practice since the early 1960s and confirmed by UN Resolution 41/65 (1986) on remote sensing principles. National legislation varies: sovereign operators must comply with their own domestic licensing regimes and, where relevant, ITAR/EAR for components. The key legal risk is not collection but dissemination — sharing imagery with non-allied parties may trigger bilateral treaty obligations.
What ground-processing infrastructure is needed to exploit persistent-watch data in near real time?
A credible sovereign system requires: (1) a network of at least 3–5 geographically distributed ground stations to minimise downlink latency; (2) automated ingest pipelines capable of orthorectification, atmospheric correction and change-detection within minutes of downlink; (3) a classified data lake with versioned baseline imagery for every watched site; and (4) analyst workstations connected to national intelligence networks. Cloud-native, GPU-accelerated processing is now standard — open-source frameworks such as GDAL/OGR and ORFEO Toolbox are widely used in national programmes.
How should a nation protect its persistent-watch satellites from jamming or physical interference?
Resilience requires a layered approach: frequency-hopping and spread-spectrum downlinks to defeat jamming; low-observable orbital profiles where operationally permissible; on-board encryption of all stored imagery to FIPS 140-3 or national-equivalent standards; and constellation disaggregation so that no single launch or anti-satellite intercept can destroy the capability. Nations should also maintain cold-spare satellites and rapid-launch agreements with domestic or trusted-partner launch providers.
How long does it take a mid-sized nation to field a basic persistent-watch capability from scratch?
Based on programmes including South Korea's 425 Project (5 satellites, 2023–2025 delivery), Israel's OPTSAT-3000 and UAE's Falcon Eye series, a realistic timeline from programme launch to first on-orbit satellite is 4–7 years for a custom design. COTS-based microsatellite approaches — using proven buses from suppliers such as SSTL, GomSpace or Satellogic — can compress this to 2–3 years for the first pair, with constellation build-out over 5 years. Budget typically ranges from $200M–$800M for a 6–12 satellite optical system.