Blog

Optical vs. SAR for flood mapping: picking the right imagery when skies won't clear

You've got a flood event unfolding and a deck of cloud sitting over the whole basin. The question that decides what you can hand to response coordination in the next few hours isn't resolution or price. It's which sensor can see through what's overhead right now.

Why SAR gets the call when the sky won't cooperate

Synthetic aperture radar sends its own signal down and reads what bounces back, so cloud cover and darkness don't matter to it the way they do to a camera. That's the whole appeal during an active storm system: a SAR pass can get you a water mask while the rain is still falling, which an optical sensor simply cannot do.

The catch is what you get back. Radar returns are a function of surface roughness, not color or spectral reflectance, so standing water shows up as a smooth, dark signature that has to be told apart from other smooth, dark things: shadowed terrain, parking lots, flat bare soil after rain. That separation takes trained interpretation or a calibrated processing chain, not a quick visual read. And the SAR constellations with usable revisit for a given basin are a smaller list than the optical fleet, so "radar is available" and "radar is tasked over your flood right now" are two different sentences.

Where optical still does the job better

Once there's a genuine cloud-free look, multispectral optical imagery is easier to turn into something a coordination cell can act on fast. Water absorbs strongly in the near-infrared band, which gives a clean, high-confidence boundary between wet and dry ground without the roughness ambiguity SAR carries. It also reads like a photograph, so anyone on the response side can eyeball the extent map and trust what they're looking at without a radar specialist walking them through it.

The tradeoff is obvious and it's the one every duty officer already knows: optical needs a break in the weather. For a frontal system parked over a river valley for three days, that break might not come until the water's already receding, which is too late for the decisions that matter most, like which roads to close and which neighborhoods to evacuate first.

Which sensor to task

In practice this comes down to your event, not a general rule. If the forecast shows the cloud deck clearing within the operational window you care about, usually the first 24 to 48 hours, tasking optical and waiting for that pass gets you a map that's faster to interpret and easier to brief once it lands. If the system is slow-moving and skies are staying socked in past that window, SAR is the only sensor that's going to produce anything, and it's worth the extra interpretation step to have a water mask instead of nothing.

It's also worth saying plainly: these aren't interchangeable products with the same turnaround. A SAR-based delineation pipeline is a different build than an optical one, with its own processing and its own skill requirement on the output side. A site that promises either sensor on demand, regardless of conditions, is promising more than most operations can stand up.

Rapid Flood Mapping works the optical side of that choice. The moment a cloud-free pass comes in over your affected area, it turns medium-resolution multispectral imagery into a standing-water extent map, delivered as a numbered situation brief you can hand straight to response coordination, instead of someone on your team spending hours delineating the boundary by hand from whatever image happened to come in.

If your event is heading toward a clear sky window, that's the moment to have this queued up and ready to go.

Back to the blog