Single ground station vs. satellite read: confirming a parametric trigger
A parametric contract is only as good as the thing it reads against. Write the trigger to a single anemometer, a single river gauge, a single rain cell, and you've made that one instrument the whole settlement mechanism. Most of the time it works fine. Then a storm takes the power line feeding the station, or the gauge sits twelve miles outside the actual damage swath, and now two parties are arguing about a number from a sensor that either stopped reporting or never saw the event the way the field did.
This is the question that comes up every time a cedant, a broker, or a counterparty pushes back on a station-based trigger: what's the alternative, and does it actually hold up better.
Where a single ground station breaks down
A point sensor measures one point. That's the whole design, and it's fine until the peril doesn't cooperate with the geometry. A few ways this goes wrong in practice:
Power and comms fail during the exact event the contract is meant to cover. Wind, flood, and wildfire all have a habit of taking out the infrastructure a ground station depends on right when you need the reading most.
The footprint doesn't line up with the station. Hail swaths, flood extents, and burn scars are irregular. A station thirty kilometers from the reference point can report calm conditions while the insured location is underwater or under hail.
There's no second source. When a counterparty disputes a station reading, and there's only one station, there's no independent check. You're arguing about whether the instrument was calibrated, whether it was maintained, whether the data feed dropped a few readings during the storm. None of that is resolvable after the fact.
Official bulletins, when they exist, often lag the settlement clock. A national met service confirming "yes, that was a Category 3 equivalent event in that district" three weeks later doesn't help a payout that's supposed to clear fast.
What a wide-swath satellite read actually adds
A satellite pass doesn't replace physics. Wind speed from a scatterometer and wind speed from an anemometer are measuring different things in different ways, and nobody should pretend otherwise. What a satellite read does add is coverage and a second, independent vantage point on the footprint itself.
Wide-swath multispectral imagery, at 10 to 30 meter ground resolution, captures a before/after comparison across the whole affected area, not a single coordinate. Triggered event-driven rather than on a fixed revisit schedule, the read happens on the event's timeline, not the satellite's. That means the imagery used to confirm the trigger is captured around the same window the loss event occurred, not whatever the archive happened to have on file.
For a product lead writing the next contract, the practical question isn't "satellite or station," it's whether the trigger definition has a second source built in at all. A station plus an independent before/after read of the same footprint gives both counterparties something to check the primary trigger against, without waiting on a government agency's monthly report or trusting one party's own damage assessment.
Where each one actually fits
Ground stations are still the right primary instrument for a lot of triggers, particularly wind speed, rainfall intensity, and anything with a long calibrated time series behind it that both sides already trust. The gap shows up when that one station goes dark, gets disputed, or sits outside the actual damage zone, and there's nothing else on record to check it against.
That's the specific gap a third-party trigger observation is built to close: an independent, before/after read of the peril footprint from wide-swath imagery, timed to the event, that either party can point to when the single-station number is the only thing on the table.
If the next contract you're structuring leans on one station for the whole payout decision, it's worth working out now what the second read looks like before the first claim comes in.