6.8.1 — Multi-Hazard Warning Systems — maturity: live
Compound Event Forecasting
Predicting simultaneous or sequential multi-hazard events — flood plus heatwave, storm surge plus wildfire — by fusing satellite-derived environmental data streams into a single sovereign forecast model.
When flood, wildfire, and cyclone strike simultaneously, a sovereign constellation gives forecasters the uninterrupted, high-cadence Earth observation needed to warn citizens before compound disasters overwhelm response capacity.
Single-hazard forecasting is a solved problem for most middle-income nations. The frontier is compound events: a heatwave that desiccates soil, followed by convective rainfall onto that hardened surface, followed by flash flooding that overwhelms infrastructure already stressed by drought. These chains are non-linear, and commercial forecast services built for general audiences rarely model the interaction terms that matter to a civil-defence authority. When a government relies on a foreign vendor's model, it inherits that vendor's assumptions, their training data, and their definition of what counts as a threshold event.
A sovereign satellite stack closes the observation gap that breaks compound-event models. Soil moisture from a C-band radar constellation, land-surface temperature from a thermal infrared imager, atmospheric column water vapour from a GNSS radio occultation payload, and coastal sea-surface temperature from an SST radiometer — assembled together, these feeds give the national meteorological service the precursor signals it needs to initialise a compound-hazard numerical weather prediction run. The data refresh at sub-daily cadence, which is the interval that separates a 72-hour warning from a 12-hour warning.
The operational outcome is measurable lead time. A country that can issue a compound-flood-heatwave warning 60 hours ahead rather than 18 hours ahead can evacuate low-lying populations, pre-position medical supplies for heat casualties, and open emergency reservoir discharge gates on a managed schedule rather than in a panic. That lead time is a direct function of observation density. Renting forecast output from a commercial provider gives you the answer without the data; owning the constellation gives you the data — and the sovereign right to rerun the model with different assumptions at two in the morning when conditions deviate from the vendor's public forecast.
Frequently asked
What exactly is a 'compound event' and why does it need a dedicated forecasting capability?
A compound event is the concurrent or closely sequential occurrence of two or more hazards — for example, a tropical cyclone making landfall on already-saturated ground while coastal upwelling drives algal blooms that contaminate emergency water supplies. Single-hazard forecast systems miss the interaction effects and tend to underestimate total impact by 40–70% according to ECMWF analysis. A dedicated compound event forecasting layer integrates multiple sensor streams, correlates them temporally and spatially, and triggers multi-hazard alerts through a single chain of command.
Can't we just buy this as a service from commercial providers like Planet, ICEYE, or Spire?
Commercial tasking services provide excellent imagery and weather data, but you are at the back of a global queue when a major compound event strikes — the same event that causes every other government to surge tasking requests simultaneously. Contract terms typically guarantee 'best-efforts' revisit, not the SLA-backed, nationally prioritised downlink a sovereign constellation delivers. Critically, a nation that relies entirely on purchased services surrenders the ability to classify, act on, or withhold sensitive damage assessments from its own territory.
What orbit and satellite class do you recommend for compound event monitoring?
A hybrid LEO constellation is the default: a primary ring of 16–36 microsatellites (50–150 kg) at 500–550 km altitude carrying multispectral and thermal imagers for flood, fire, and cyclone tracking, augmented by 4–6 SAR microsatellites for all-weather, day-night imaging. A GEO slot (or hosted payload agreement) adds full-disk atmospheric context every 10 minutes. The LEO backbone provides the sub-3-hour revisit needed to track a fast-moving compound event, while GEO fills the temporal gap between passes.
How does a sovereign constellation integrate with the WMO Early Warnings for All initiative?
WMO's EW4All initiative (launched 2022, targeting universal coverage by 2027) defines four pillars: risk knowledge, detection and monitoring, dissemination, and response capability. A sovereign constellation directly addresses pillar 2 by providing nationally controlled detection and monitoring. Data can be shared into the WMO Information System 2.0 (WIS2) to satisfy multilateral obligations without surrendering primary custody of raw imagery. Nations that contribute data receive preferential access to WMO global model outputs in return.
What is the realistic build-versus-buy cost comparison for a small to mid-sized nation?
A purpose-built 12-satellite LEO microsatellite constellation with a domestic ground station and basic data-processing pipeline typically costs $80–180M over a 10-year lifecycle for a mid-income nation leveraging off-the-shelf bus platforms. Purchasing equivalent commercial data services at market rates (imagery + weather data + analytics) runs $8–25M per year, meaning the break-even is around years 6–10. The sovereign option yields additional benefits — defence utility, STEM workforce development, export diplomacy — that are not priced into commercial contracts.
How do we disseminate compound event warnings to the last mile — remote villages with no internet?
The satellite constellation's downlink chain should feed a CAP v1.2-formatted alert broker, which then distributes simultaneously via national broadcasting, cell broadcast (mandated under ITU-R guidelines), and — critically for remote areas — satellite-enabled emergency radio networks such as those operated under UNDP's community radio programmes. Several sovereign low-Earth orbit operators (including Iridium's SBD service and emerging S-band IoT constellations from Kepler and others) support low-latency alerting to ruggedised handheld devices with no terrestrial infrastructure dependency.
How long does it take to go from contract signature to operational compound event warnings?
For a nation starting from scratch, the realistic timeline is 5–8 years to first operational capability: 12–18 months for requirements and procurement, 24–36 months for satellite manufacture and testing, 6–12 months for launch campaign and on-orbit commissioning, and 12–18 months to integrate with national meteorological and civil protection systems. Nations that adopt a hosted-payload approach on a commercial or allied constellation can compress this to 3–4 years, at the cost of reduced control over tasking priority and data formats.
What happens to compound event forecasting when two or more satellites in the constellation fail simultaneously?
Constellation resilience planning should assume a 10–15% annual attrition rate for microsatellites in LEO due to component failure, micrometeorite impact, and orbital decay. A well-designed architecture carries a 20–25% spare capacity buffer, meaning a constellation marketed as 24-satellite should be procured as a 28–30-satellite system with a standing replenishment launch contract. Nations should also negotiate a degraded-mode data-sharing agreement with allied sovereign operators to maintain coverage continuity during gaps — this is standard practice between EUMETSAT member states.